Amino Acid Chelation in Human and Animal Nutrition (PDFDrive)

AMINO ACID
CHEL ATION
IN
HUMAN AND
ANIMAL
NUTRITION
AMINO ACID
CHEL ATION
IN
HUMAN AND
ANIMAL
NUTRITION
H. DeWAYNE ASHMEAD
Boca Raton London New York
CRC Press is an imprint of the

Taylor & Francis Group, an informa business
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2012 by H. DeWayne Ashmead

CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works

Version Date: 20111129
International Standard Book Number-13: 978-1-4398-9768-3 (eBook - PDF)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any copyright material has
not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, microfilming, and recording, or in any information storage or retrieval system,
without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the CCC,
a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used
only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at

http://www.crcpress.com
Contents
Foreword...................................................................................................................vii
Introduction................................................................................................................ix
About the Author.......................................................................................................xi
Chapter 1 The Fundamentals of Mineral Nutrition............................................... 1
Chapter 2 The Chemistry of Chelation................................................................ 19
Chapter 3 The History of Nutritional Chelates.................................................... 35
Chapter 4 The Requirements for a Nutritionally Functional Chelate.................. 49
Chapter 5 The Development of Analytical Methods to Prove Amino Acid

Chelation............................................................................................. 61
Chapter 6 Absorption of Amino Acid Chelates from the Alimentary Canal...... 81
Chapter 7 The Pathways for Absorption of an Amino Acid Chelate...................97
Chapter 8 The Absorption of Amino Acid Chelates by Active Transport........ 117
Chapter 9 The Absorption of Amino Acid Chelates by Facilitated Diffusion.... 135
Chapter 10 The Fate of Amino Acid Chelates in the Mucosal Cell.................... 153
Chapter 11 The Uptake of Amino Acid Chelates into and out of the Plasma..... 171
Chapter 12 Tissue Metabolism of Amino Acid Chelates.................................... 185
Chapter 13 Some Metabolic Responses of the Body to Amino Acid Chelates..... 201
v
vi Contents
Chapter 14 Toxicity of Amino Acid Chelates...................................................... 223
Chapter 15 The Absorption and Metabolism of Amino Acid Chelates............... 233

Foreword
Mineral bioavailability has historically been “the black box” of micronutrient metab-
olism. Dietary intake of a mineral micronutrient in sufficient quantities to meet
dietary reference intakes does not always ensure adequate metabolizable mineral
at the tissue level. Minerals are by nature ionic and form complexes and chemical
compounds quite readily. The pathway from the food or supplement in which they
are contained to their target cells in the body provides multitudinous opportunities
to interact with their immediate chemical environments. The foodstuffs with which
they are ingested, the acidic and chemical milieu of the digestive tract, the absorptive
surface and interface of the gastrointestinal tract, the ions in the plasma, and ulti-
mately the cellular matrix to which they are delivered can interact to influence the
ultimate efficacy of the structural, metabolic, or catalytic roles of the dietary min-
eral. The seemingly large doses of mineral supplements needed to correct a dietary
mineral deficiency can be explained in terms of the “inefficiency of absorption”
or, in broader terms, the lack of “bioavailability” of the particular mineral supple-
ment. Mineral nutritionists have long sought chemical forms of minerals that evoke
a greater or more positive response at the target tissue. Two important historical
examples of mineral nutrition research that continue to be pursued today are calcium
supplementation to influence bone mineralization and iron supplementation to influ-
ence blood hemoglobin levels. Not all covalently bound minerals ionize sufficiently
to release their mineral counterpart optimally at the sites of absorption in the gut.
Mineral absorption from the gut is a complex topic, considering the various routes
that are available (e.g., passive absorption, facilitated absorption, active transport)
to account for the disappearance of the mineral from the gut and its appearance in
the plasma.
Enter the concept of supplying the mineral in an ionic or covalently bound pro-
tective amino acid matrix (chelate) with a stability factor that helps to circumvent
ionization issues and delivers the mineral to sites of absorption in the intestinal brush
border. Certain amino acids form soluble complex molecules with metal ions, thus
“protecting” the ions so that they cannot react with other elements or ions prior to
arriving at the absorptive site in the gut. The chelated mineral ligand can then be
either passively absorbed, subsequently released to its transporter, or in some man-
ner “escorted” through the absorptive surface of the gut to permit a more rapid and
quantitative transfer of the mineral from the intestinal contents, across the intesti-
nal villi and into the blood. The principle of chelation extends well beyond amino
acid chelates and is well documented in organic and inorganic chemistry. This book
explores the chelation principles as applied to the biochemistry of mineral absorp-
tion and metabolism, specifically focusing on the formation and absorption of amino
acid metal chelates.
The progress and development of amino acid mineral chelates has not been with-
out controversy. Although the improved bioavailability of some amino acid mineral
chelates is generally accepted, it has not been clearly understood exactly why these
vii
viii Foreword
chelates provide improved absorption. Early studies of the nutritional aspects of the
bioavailability of mineral chelates occurred during the 1960s and 1970s when ana-
lytical techniques suggested, but did not permit, direct implication of chelates in
improved absorption and transfer of mineral across the gut. Over the intervening
years, considerable indirect evidence and some direct evidence of enhanced bioavail-
ability was gained through numerous animal and a few human feeding trials. Much
of this early experimental information was initially studied with an agricultural
emphasis and published in related animal nutrition venues and proprietary in-house
publications sponsored by early innovators of chelated mineral products such as
Albion Laboratories. Some of these publications were not widely read by or acces-
sible to mineral researchers due to the early emphasis in livestock applications and
publication venues that were not readily available or read by those in the human min-
eral nutrition field. By publishing this book, Ashmead makes this information more
readily available to a wide audience.
In this book, DeWayne Ashmead provides a historical account of the theory and
application of chelates to mineral nutrition. Much of the pioneering early work was
accomplished by DeWayne’s father, the late Harvey Ashmead. Albion Laboratories
is a family-owned and operated business, and at first glance, one might imagine that
the content of this book would be a treatise on the nutritional superiorities of min-
eral amino acid chelates. That preconceived notion would be a mistake. This book
is a scholarly compendium that not only provides the historical context of chelates
but also explains the chemistry of chelation and the formation of amino acid min-
eral chelates in considerable detail. The book contains a well-developed introduc-
tion and discussion to the complexities of mineral bioavailability. Ashmead then
progresses to review the analytical methodology necessary to establish that one is
indeed working with a true chelate prior to engaging in direct feeding comparisons
of amino acid mineral chelates versus inorganic forms of the mineral in question.
Tabular and graphical data from feeding trials previously published in the literature
as well as some extracted from some difficult-to-access publications and previously
unpublished work are presented in the chapters on amino acid mineral chelates. The
concept and criteria for the development of a “nutritionally functional” metal chelate
are presented and discussed.
Although the main focus of this book is on the ingestion of amino acid metal
chelates as a way to optimize mineral absorption, the book also provides a good
fundamental discussion of chelation chemistry. Ashmead provides not only his inter-
pretation of the results of numerous studies of animal and human amino acid mineral
chelate digestion and absorption but also alternative interpretations. One cannot help
but admire the clarity of writing and the logical and stepwise development of the
material in this book. This reference should be invaluable to bioinorganic mineral
researchers and others seeking to enhance mineral bioavailability to support opti-
mal health and productivity.
Wayne Askew, PhD

Professor, Division of Nutrition
University of Utah
Introduction
In the early 1960s, a study was conducted in which gestating rats were given diets
containing the same mineral content of mineral salts or amino acid chelates. The
young from the group that was given amino acid chelates had a much higher survival
rate and grew faster. This type of study was then extended to dairy cows. Here, it was
found that both milk and butterfat productions were higher in the group receiving
amino acid chelates. This type of study was then extended to laying hens; greater
production and fewer broken eggs were observed from the group receiving amino
acid chelated minerals. Other researchers conducted a study with gestating sows.
This study showed that the group receiving amino acid chelated iron had higher
birth weights, lower mortality, and greater weight gains than those given the normal
iron dextran treatment. These studies initiated many others on the absorption of
amino acid chelated metals. The studies consistently demonstrated that amino acid
chelates were absorbed better and improved some aspect of health in humans and
other treated animals.
Although chelation was first observed over 100 years ago, it has only been in the
last 50 years that scientists discovered the nutritional benefits of amino acid chelates.
This book examines the reasons for those benefits, the chemistry of chelation, the
analytical methods that have been used to prove or verify chelation, and a detailed
discussion of the absorption and metabolism of various metal amino acid chelates
compared to mineral salts. The requirements for nutritionally functional chelates and
their absorption are discussed in this text. For a chelate to be formed, a metal must
be a member of a heterocyclic ring. When an amino acid forms a chelate, the car-
boxylate anion forms a bond with a positively charged metal. This places the amine
group in perfect position to share its pair of electrons with the metal to form a bond
to the metal and create a heterocyclic ring or chelate. Depending on the charge on the
metal, this process can be repeated one or more times. The structure of this chelate
can be proven by x-ray crystallography and strongly indicated by Fourier transform
infrared (FT-IR) spectroscopy.
It is logical to conclude that the amino acids, which surround the metal, protect
the metal from reactions that can greatly inhibit its absorption. Some of the reac-
tions that produce precipitation of the metals are reactions with phosphates, phytic
acid, and other substances commonly found in the gut. This protection of the met-
als is related to the stability of different amino acid chelates. More stable amino
acid chelates provide better protection against precipitation. It is also logical that in
lower pH environments the amine portion of the amino acid could accept a proton.
The pair of electrons that provided the bond to the metal is now used to bond to the
proton. When this happens, the protonated amine carries a positive charge and the
chelate ring is broken. This produces a chelate/complex rather than a chelate, but
Dr. Ashmead explains how this allows the metal amino acid chelate/complex to be
attracted to negatively charged transport molecules and thus be absorbed through
ix
x Introduction
active transport. The relationship between absorption through passive diffusion,

facilitated diffusion, as well as active transport is explained.
A study to determine the fate of amino acid chelates used a radioactive isotope
of the metal and another radioactive isotope in the amino acids. There appeared
to be some division of the metal and the amino acids in the mucosal tissue due to
hydrolysis. Differences in the amount of hydrolysis of the amino acid chelates in the
mucosal tissue are explained on the basis of the stability of the amino acid chelates.
Regardless of how much hydrolysis occurs in the mucosal tissue, some of the amino
acid chelate or chelate/complex appeared to be transferred to the plasma intact. The
metabolism of these amino acid chelates has been shown to produce responses in
performance or production of the animals being tested, and because of greater tissue
retention, these amino acid chelates can provide long-term positive responses.
Increased absorption of amino acid chelates has been observed many times in
tests where a radioactive isotope of the metal is given to the animal as an amino acid
chelate or as a mineral salt. After dosing, the amount of mineral that is absorbed by
various tissues and organs can be accurately determined. These tests demonstrate
that amino acid chelates provide better mineral absorption than when these minerals
are given as salts. Even though amino acid chelated minerals have greater absorption
than mineral salts, to be effective these amino acid chelates must be bioavailable. A
detailed explanation of why this occurs is found in this book.
Bioavailability of minerals is sometimes more difficult to determine, but this is
usually done by comparing some aspect of health or production when different types
of minerals are given. Many studies are reviewed that range from improving iron
deficiency anemia in human infants, to milk production in cows, to improved sur-
vival of baby pigs. These studies all showed that when amino acid chelated minerals
are in the diet, the response is improved health or production.
Although introduction of amino acid chelates in mineral nutrition initially met
with considerable skepticism and controversy, greater absorption and bioavailabil-
ity of amino acid chelated minerals compared to nonchelated minerals has been
well documented. This book reviews many of the studies that provided information
on the comparison of amino acid chelates and nonchelated minerals. These studies
were conducted using many different animals, including humans, under a variety of
conditions, and amino acid chelates consistently provided improved responses that
resulted from better absorption and bioavailability of the minerals being tested.
Boyd R. Beck, PhD

Retired Professor of Chemistry
Snow College, Ephraim, Utah
About the Author
Dr. H. DeWayne Ashmead, president of
Albion Laboratories Incorporated, has been
involved in research related to amino acid
chelates since the 1960s. The results of his
research and the research that he and his
father, the late Dr. Harvey Ashmead, directed
have been published in seven books authored
by Dr. Ashmead. He has also published over
25 peer-reviewed journal articles and over
60 magazine articles on the same subject. In
addition, he has authored chapters on chela-
tion in several books. His research has also
led to 18 patents.
Dr. Ashmead received his BS degree in
business in 1969 and his PhD degree in clini-
cal nutrition in 1981. He sits on the board
of directors of his own company, Albion
Laboratories, as well as the boards of a bank, a hospital, and two universities. He has
been recognized with an honorary doctorate of humanities by Weber State University. In
2006, he was honored by Ernst & Young as the regional Entrepreneur of the Year in
the area of health sciences. In 2008, he received the State of Utah Governor’s Medal
for Science and Technology. He is a member of several professional organizations.
xi
1 The Fundamentals
of Mineral Nutrition
During the Italian Renaissance, Leonardo da Vinci (1452–1519) wrote, “If you do
not supply nourishment equal to the nourishment departed, life will fail in vigor; and
if you take away this nourishment, life is utterly destroyed.”1 The science of nutrition
is thus the science of nourishing the body.
The body is, to a degree, the product of its nutrition. Nutrition begins with the
intake of foodstuffs. They undergo digestion, which transforms those foodstuffs
into basic nutrients. The nutrients are then passed through the gastrointestinal tract
wall into the blood and ultimately the cells that compose the body, where these
nutrients carry out their life- and health-sustaining functions. If the foodstuffs
contain inadequate or unbalanced nutrients, the body responds by not performing
at peak efficiency, which is another way of saying that the metabolic processes
within the body cells are compromised. This interruption of function is manifest
as insufficient energy, poor growth, morbidity, and if too severe, mortality of the
whole body.
When considered in its most basic terms, nutrition is the optimal intake of pro-
teins, carbohydrates, lipids, vitamins, minerals, and water. Depending on the author-
ity consulted, these six nutrient groups carry out three or four basic functions: (1)
They serve as a source of energy for the body; (2) they are essential for the growth
and maintenance of body tissue; (3) they regulate body processes; and (4) they are
required for sexual reproduction of the body.
A closer examination of these functions reveals that energy comes from the catab-
olism of carbohydrates, lipids, and protein. The metabolic processes required to
extract the energy requires the presence of certain vitamins and minerals in specific
enzymes along with sufficient water to facilitate the resultant enzymatic reactions
required to convert the carbohydrates, lipids, and protein into energy. Figure 1.1 pro-
vides a simplified illustration of those relationships.2
Enzymes are proteinaceous molecules that catalyze biochemical reactions. The
presence of specific amino acids and their exact order in the enzyme molecule will
govern the reaction that the enzyme molecule catalyzes. Each amino acid contains
a carboxyl group, an amine group, and its radical which is attached to the α-carbon.
The radical, or R group, is the unique portion of the molecule that separates each kind
of amino acid from every other kind. The active site in the enzyme is so arranged
that it can bind to a specific substrate (the reactants, i.e., the energy nutrients) through
the amino acid R groups. In some enzymes, the active site will promote the bending
of the substrate in such a way that it accelerates a certain reaction. In other enzymes,
the R groups attach to, or chemically react with, the substrate, which enhances the
rate of the enzymatic reaction.3
1
2 Amino Acid Chelation in Human and Animal Nutrition
Vitamins Minerals
BIOTIN CALCIUM
Lipid Metabolism Pancreatic Lipase
NIACIN PHOSPHORUS
Lipid Metabolism ATP
RIBOFLAVIN
Energy MAGNESIUM
Glycogenesis Energy Expenditure
PANTOTHENIC ACID SULFUR

Activates Coenzyme A Fatty Acids Catabolism
THIAMIN Carbohydrate IODINE

Glucose Metabolism Thyroxin
Protein Fat
FOLACIN POTASSIUM
Amino Acid Metabolism Glucogenesis
SODIUM
VITAMIN A, D & E
Glucose Absorption
Oxidative Stress
VITAMIN B6 Enzymes MANGANESE
Transamination Fatty Acid Synthesis
VITAMIN B12 COPPER

Conversion of Water Cytochrome Oxidase
Monosaccharides to Energy IRON
VITAMIN C Oxidation
Carnitine Synthesis ZINC
VITAMIN E Protein Synthesis
Transamination CHROMIUM
Glucose Tolerance
VANADIUM
Glucose & Lipid Metabolism
FIGURE 1.1 The interrelationships of vitamins, minerals, and water on the enzymes
required to extract energy from carbohydrates, lipids and protein. (Redrawn from Ashmead,
HD, Conversations on Chelation and Mineral Nutrition (New Canaan: Keats) 26, 1989.)
A small number of enzymes, such as pepsin or trypepsin, are composed exclu-

sively of protein and nothing else. Most enzymes, however, are composed of complex
proteins (the apoenzyme) linked to a nonprotein group (prosthetic groups). When the
prosthetic group can be readily removed from apoenzyme, that prosthetic group is
called a coenzyme. The enzyme functions only when the apoenzyme and prosthetic
groups are joined together.
In other enzymes, the protein portion of the molecule may have a simple metal
ion attached to it. When the metal is removed or substituted, the enzyme loses or
decreases its activity. If it is replaced, the catalytic properties of the enzyme return.
Not all trace minerals function as activators in an enzyme. Some are incorporated
into the apoenzyme, while others are parts of the prosthetic groups. The roles of
specific metals in either accelerating or inactivating enzymatic activity cannot be
overemphasized. Excesses or deficiencies of these essential elements can affect the
rate of catalytic action of the affected enzymes.4
Like the trace elements, specific vitamins also function primarily as coenzymes.
Structurally, most vitamins are part of the apoenzyme and are usually responsible
for the attachment of the enzyme to the substrate.5
Enzymatic reactions generally require the presence of water. Most minerals must
be ionized to function within the apoenzymes. Many of the vitamins are water solu-
ble and require the presence of water to enter into the enzyme system.
The Fundamentals of Mineral Nutrition 3
As can be seen from this synopsis, while minerals are not a direct source of
energy, their involvement in extracting energy from specific nutrients is critical. One
must think of minerals not only in the nutrient sense but also in the biochemi-
cal sense. Most of the roles played by minerals, particularly the trace elements,
are biochemical.
An exception to the statement is the second function of nutrients, the growth and
maintenance of body tissues, which is a structural role. As the infant grows to adult-
hood, protein, minerals, and other nutrients play direct roles. Soft tissue is composed
mostly of protein and to a lesser degree lipids. Hard tissue is primarily created from
minerals. Minerals also play indirect roles in creating body tissue. The consump-
tion, digestion, and reconstruction of protein for body tissues require enzymatic pro-
cesses. As previously described, certain enzymes have specific vitamins or minerals
that are integral parts of the enzymes or serve as cofactors. Water also plays a role
in creating body tissues.
Once maximum growth is achieved, the body must maintain itself. Tissues wear
out and are replaced. Bones dissolve and remineralize. Soft tissues, organs, and the
like are continually rebuilt as nutrients are ingested and absorbed. Maintenance of
body tissue is a 24-hour-a-day process.
Regulation of body processes will generally involve biochemical processes
requiring protein, minerals, vitamins, and water. All of these nutrients have one or
more direct roles in establishing acid/base balance, creating hormones, controlling
osmotic pressure, moving nutrients into body cells, and so on. Much of the regula-
tion of body processes is accomplished by enzymatic activity. There are, however,
requirements for some nutrients in their ionic, uncomplexed forms (e.g., sodium
and potassium) in body fluids. Regulatory processes can become very complicated
depending on the requirements for functionality. Many of these processes have need
of more than one sequential biochemical or enzymatic chain reaction to achieve the
overall desired control.
The final role of nutrients is for sexual reproduction. In a sense, reproduction can
be included in one or more of the other three roles since all are involved in reproduc-
tion. Energy is required; creation of new tissue is required; and hormonal changes
must take place for reproduction to occur. Thus, protein, carbohydrates, lipids, vita-
mins, minerals, and water are all necessary to the sexual reproductive process.
The number of roles that a single nutrient plays in carrying out one or more of
the four basic functions of nutrition in no way determines its relative importance
to the body. A deficiency or an excess of a nutrient required in minute amounts may
precipitate more severe consequences to the body than the deficiency or excess of a
nutrient needed in larger amounts.6 Optimum intake is the key to nutrient efficiency.
Too much or too little of a given nutrient has an equally deleterious effect on the
body, as illustrated in Figure 1.2.
If the nutrient deficiency or toxicity is marginal, the health and well-being of the
body and its performance may be impaired. The degree of impairment depends on
the extent of the toxicity or deficiency. Whenever the body suffers an acute defi-
ciency or extreme toxicity of an essential nutrient for a prolonged period, death will
result. When the intake of the nutrient is neither deficient nor toxic but provided in
the optimal range, the responses of the body are health and peak performance.
Optimal
100%
Percent of Body Performance
Marginal
Marginal
Death Death
0%
Deficiency Toxicity
Concentration Nutrient Intake
FIGURE 1.2 A typical dose response to nutrient intake. The shape of the curve can change
depending on the nutrient need as well as the nutrient involved.
The previous discussion, of course, assumes that each nutrient operates in a vac-
uum. Such is not the case. The presence, absence, or even the level of presence of a
specific nutrient in the diet may affect the absorption and metabolism of numerous
other nutrients. For example, the amino acid methionine is reported to be preferen-
tially absorbed in the presence of other amino acids.7 Certain minerals can also be
antagonistic to other minerals during metabolism. To illustrate, calcium and mag-
nesium are mutually antagonistic. Calcium is also antagonistic to manganese, but
manganese has no effect on calcium.8 In another example, Wilson’s disease, a meta-
bolic error resulting in copper toxicity, is treated by high doses of zinc, which tend to
reduce or prevent copper absorption.
Balance becomes extremely important to achieve optimal nutrition. There are
three aspects to the concept of balance. There must be balance between food groups
for optimum nutrition. In human nutrition, for example, there must be a balance
between food groups, such as meat, dairy, fruit, vegetables, and so on. When this
balance is ignored, the consequences can be dramatic.
Second, there must be balance between nutrient groups. A high-protein diet at
the expense of carbohydrates and lipids may not be the most efficient way to obtain
energy. Further, other problems, such as ketosis, may result from a high-protein diet.
A strict vegetarian diet is frequently deficient in iron, vitamin B12, folic acid, and
other nutrients. These nutrients must be supplemented for balance to occur.
Besides balance between food and nutrient groups, a third requirement requires
that balance must exist between individual nutrients within a nutrient group. For
example, the essential amino acids, the building blocks of protein, must be in bal-
ance one with another if efficient use of the food is to be accomplished. Many years
ago, Morrison observed, “A shortage of a simple [essential] amino acid will limit the
use of all others, and therefore reduce the efficiency of the entire ration.”9 Figure 1.3
clearly shows the necessity of an amino acid balance.10 This drawing illustrates that
excesses of any amino acid can interfere with the utilization of those amino acids to
which the arrows emanating from the originating amino acid point. For example, an
Threonine
Histidine Glutamic Acid
Lysine Cystine
Glycine Alanine
Phenylalanine Arginine
Leucine Serine
Valine Methionine
Proline Isoleucine
FIGURE 1.3 The relationships of several amino acids to each other. An excess of one
of the amino acids will affect the absorption/metabolism of those amino acids to which it
points. A deficiency of that amino acid will allow the accumulation of those amino acids to
which it points. (From Graff, D, “Radioactive isotope research with chelated minerals,” in
Ashmead D, ed., Chelated Minerals Nutrition in Plants, Animals and Man (Springfield, IL:
Thomas) 275, 1982.)
excess of threonine can interfere with the utilization of phenylalanine. High levels
of glutamic acid can also affect phenylalanine. If phenylalanine is in excess, it can
interfere with the utilization of glutamic acid, but it has no effect on threonine.
Vitamins also have definite relationships with each other. For example, if the
body has a vitamin B6 deficiency, it cannot utilize vitamin B12 efficiently.11 Vitamin
A and E are synergistic.12 Some of the basic interrelationships between vitamins
are summarized in Figure 1.4 and are based on several published vitamin balance
studies.13,14 This figure emphasizes that when there is a deficiency of one vitamin,
such as B6, it results in less utilization of several other vitamins, including riboflavin,
vitamins B1, A, E, C, niacin, folic acid, biotin, and vitamin B12.
To further complicate the picture, it will be recalled that amino acids are essential
for growth and maintenance of body tissues. To regenerate the protein for body tissues,
these required amino acids must be in balance. Selecting three specific amino acids,
valine, leucine, and isoleucine, as an example, there must be adequate amounts of
biotin and pantothenic acid present for the utilization of those particular amino acids.11
Figure 1.4 emphasizes that both of these vitamins cannot be utilized efficiently unless
there are appropriate amounts of available riboflavin, folic acid, and vitamin B12.
Referring to Figure 1.3, it can be quickly noted that both leucine and isoleucine
will depress the uptake of valine. If the diet were marginally deficient in biotin and
pantothenic acid, they would first be utilized to meet the requirements for isoleucine
uptake followed next by leucine. If any of the vitamins remained after satisfying
the requirements for isoleucine and leucine, they would then be utilized for valine
A (β-carotene)
E
(Tocopherol) B1 (Thiamin)
Pantothenic Acid Riboflavin
C
B6
(Ascorbic Acid)
(Pyridoxine)
Niacin B12
(Cobalamine)
Folic Acid D
(Calciferol)
K Biotin
(Phylloquinone)
FIGURE 1.4 Synergism among several vitamins. An excess or deficiency of any one of
the vitamins in this figure will affect the absorption or metabolism of the other vitamins
connected to it by the lines. (Redrawn from Patrick, H, and Schaible, P, Poultry Feeds and
Nutrition (Westport, CT: AVI) 144, 1980; and Levander, O, and Cheng, L, eds., Micronutrient
Interactions: Vitamins, Minerals and Hazardous Elements (New York: New York Academy
of Sciences) 80–129, 1980.)
absorption and utilization. Thus, the body could potentially suffer from a valine
deficiency due to marginal deficiency of biotin and pantothenic acid. At this point,
the question of balance becomes even more complicated. If riboflavin, folic acid, or
vitamin B12 were marginally deficient in the diet, they may cause a depression in the
biotin and pantothenic acid utilization, resulting in an inadequate utilization of all
three of these amino acids.
If one were to add minerals to the nutritional balance equation, the results become
even more complicated. For optimum nutrition, the minerals must also be in balance.
An excess of any one of them could result in a depression of certain other minerals,
just as excesses or individual amino acids can result in the depression of other amino
acids. Figure 1.5 indicates this.8
The late Professor Eric Underwood said, “Metabolic interactions among trace
elements are so potent and so diverse that no consideration of the current status of
nutrition would be reasonable without some account of their nutritional implica-
tions.”15 Underwood went on to state that the interactions are more common among
metals that share common chemical parameters and compete for common metabolic
sites within the body.
Suttle summarized these interactions and grouped them into six categories16:
1. The formation of insoluble complexes between dissimilar ions

2. Competition for metabolic pathways between similar ions
Ag
Ca Cd
Na
Be
Se
Al
Fe
Cu
N
Mn
Co
K
P
Mo
S I
F Mg
As Zn
FIGURE 1.5 Mineral relationships in the body. The absorption or metabolism of an individ-
ual mineral is affected by the levels of intake of the other minerals pointing to that individual
mineral. (Redrawn from Dyer, IA, “Mineral requirements,” in Hafez, ESE, and Dyer, IA,
eds., Animal Growth and Nutrition (Philadelphia: Lea & Febiger) 313, 1969.)
3. The complexing of ions by metal-binding proteins

4. Changes in the metallic component of metalloenzymes
5. Facilitation of trace mineral transport
6. Codependence of trace element reactions on each other
The first category relates to the formation of insoluble complexes between dissim-
ilar ions.16 In their ionic form, while in the digestive tract, minerals are able to form
insoluble complexes with anionic ligands sourced from the diet, resulting in lower
mineral bioavailability. For example, dietary phosphorus, generally in the form of
phosphates, can reduce the availability of both iron and zinc.17,18 The chemical reac-
tion occurring in the gastrointestinal tract can produce either iron or zinc phosphate,
both of which exhibit very poor solubility. When dealing with inorganic metal salts,
generally, solubility is a major key to their availability.
Digestion can also lead to the formation of insoluble compounds.16 The release of
phytates from grain-based foods is an excellent example. The phytic acid can bond with
a cation and reduces its solubility and thus availability. Another example is illustrated
in a study in which the combination of dietary molybdenum and sulfur along with iron
reduced the absorption of dietary copper.19 The molybdenum-sulfur effect begins with
the substitution of the sulfur in the sulfide ion for oxygen from the MoO42- ion:
MoO42- → MoO3S2- → MoO2S22- → MoOS32- → MoS42-
The tetrathiomolybdate (MoS42-) ion is then able to bind with dietary copper ions and
render them insoluble and unavailable.
In animal diets, if molybdenum intake exceeds 10 mg/kg of dry matter, the

MoS42- formed may also interfere with copper metabolism. The tetrathiomolybdate
ions form in the plasma following the absorption of molybdenum (associated with
albumin) and sulfide ions and subsequently complex with copper ions. Suttle sug-
gested this and other reactions may result in the formation of insoluble inorganic
complexes in the tissues.16
The second group of trace element interactions can occur chemically between sim-
ilar ions. These interactions generally manifest themselves through competition for
transport molecules to carry the minerals into the mucosal tissue from the lumen. The
competition for binding sites on transport molecules can occur between groups of trace
elements or groups of macrominerals or between trace elements and macrominerals.20
To illustrate this, when competing with iron ions, copper ions are preferentially
bound to transferrin, which has been identified as a protein transport molecule in
the intestinal mucosa. Under normal circumstances, the transport mechanism is
not saturated. Thus, there are adequate bonding sites for both iron and copper ions.
However, when both copper and iron are administered in excess, iron absorption is
inhibited because the copper is bound first to the transferrin, and inadequate binding
sites are left for all of the iron ions.21
The third group of interactions summarized by Suttle involves the formation of
metal-binding proteins. When metal loading occurs, the normal biological reaction
is to synthesize proteins in the plasma and tissues to complex the increased metal
load. The problem is that these proteins are not specific to the metal that stimulated
the production of the protein molecules in the first place. These protein molecules
can bind other elements as well. For example, the addition of a cadmium or zinc load
to the diet will induce the formation of a soluble cysteine-rich protein in the kidney
or liver. Further, it will bind not only the cadmium or zinc but also mercury and cop-
per.16 The binding of these minerals is preferential depending on the metal and its
valence. As shown in Table 1.1, there is a hierarchy of the minerals. The metal at
the top will replace all of the metals below it in the table. As one moves down the
electromotive series, each element will displace those metals below it. Concurrently,
that element can be removed by any of the minerals above it, which complicates the
potential processes.22
A change in the metal component of a metalloenzyme involves the fourth group
of mineral interactions. As noted, most enzymes require the presence of a mineral
to function. This metal can be part of an apoenzyme, but more often it is part of
the cofactor within the prosthetic group. Other minerals have been noted in cer-
tain enzymes that have integral functions that are not yet elucidated. Further, some
enzymes are activated by a specific mineral, whereas the activities of other enzymes
are blocked by the presence of that same mineral.23,24
Aminopeptidase is an example of this. It contains manganese or magnesium
as active parts of the prosthetic group. Either element will activate the enzyme.
Additional manganese and zinc are also found in the enzyme, but their functions are
not completely understood. The manganese or magnesium in the prosthetic group
can replace each other and the enzyme will continue to function, but if the man-
ganese or magnesium is displaced by iron, lead, mercury, or copper, the enzymatic
activity of aminopeptidase is blocked.23
TABLE 1.1
A Partial List of the Electromotive
Series of Minerals and Oxides
Metal
V+3 Ni+2 Fe+2
Fe+3 Pd+2 Mn+2
In+3 Y+3 V+2
Th+4 Pb+2 Ca+2
Hg+2 TiO+2 Sc+3
Ti+3 Zn+2 Mg+2
Ga+3 Cd+2 Sr +2
Cu+2 Co+2 Ba+2
VO+2 Al+3 Rare Earths
Source: Data from Ashmead, H, “Tissue trans-

portation of organic trace minerals,”
J Appl Nutr 22:42–51, Spring 1970.
A second example is the enzyme carboxypeptidase. This enzyme is activated by

zinc. When activated, the enzyme will split the peptide bonds of certain peptides
and thus liberate the amino acids. Replacing the zinc with cobalt in the enzyme will
retard its activity.25 In the same group of proteolytic enzymes that attack the pep-
tide bonds of proteins and peptides is glycyl-glycine dipeptidase. It requires cobalt
or manganese for its activation.26 On the one hand, cobalt activates one peptidase
enzyme; on the other hand, its presence retards a different peptidase enzyme.
The fifth group of mineral interactions listed by Suttle involves the transport and
excretion of trace elements.16 These relate to specific interrelationships. One example
is the role of copper in ceruloplasmin. Its presence will facilitate the transport of iron
for normal hemopoiesis. The ceruloplasmin functions as a ferroxidase and catalyzes
the conversion of ferrous iron to the ferric state. This allows iron that is stored in the
liver and reticuloendothelial system to be transported in the plasma as ferric iron.27
While this example is somewhat synergistic, the following is exactly the opposite.
As was pointed out above the trace element, molybdenum, can interfere with cop-
per metabolism through the formation of highly stable CuMoS4 molecules in the
plasma.19 In a ruminant study, a group of calves was fed a supplement that contained
20 mg Cu and 10 mg Mo/kg of supplement. Each animal received 0.68 kg of this
supplement daily for 120 days. At 0, 60, and 90 days, liver biopsies and blood serum
samples were obtained and assayed for copper and molybdenum. Table 1.2 summa-
rizes the mean results as a percentage of the initial levels.
This study demonstrated that as the molybdenum concentration increased in the liver
or the serum, the concentration of copper declined. The molybdenum appeared to cause
a mobilization of tissue copper with a consequential increase in copper excretion.19
The final group of mineral interactions involves the codependence of different
reactions on each other.16 Suttle has reported that the involvement of a trace element
TABLE 1.2
Effect of Molybdenum on Copper Concentrations
in Liver and Blood Serum (%)
Initial 60 Days 90 Days
Liver Cu 100 39.37 26.90
Liver Mo 100 140.00 147.00
Serum Cu 100 83.3 78.13
Serum Mo 100 118.50 418.33
Source: Data from Ashmead, HD, and Ashmead, SD, “The

effects of dietary molybdenum, sulfur and iron on
absorption of three organic copper sources,” J Appl Res
Vet Med 2:1–9, 2004.
in the formation of an insoluble complex will limit the capacity of that element to
interfere with the absorption or metabolism of other trace elements. Referring to
Figure 1.5 and considering the previously described copper/molybdenum animal
study, if the molybdenum were tied up with the copper, then it cannot depress or
interfere with phosphorus metabolism.
Not only do the interactions between minerals affect their absorption and metabo-
lism, but these interactions can also influence the metabolic response to other nutri-
ents. To illustrate, in the biochemical utilization of valine, coenzyme A (CoA) is
required. It is produced in adequate quantities provided that a sufficient amount
of available magnesium is present as a cofactor to catalyze the enzyme activity.28
Pantothenic acid is also needed in that same series of reactions.28 Again referring to
Figure 1.5, if calcium, phosphorus, or manganese is too high, utilization of magne-
sium may be reduced or prevented. If that were to occur, then again, there is interfer-
ence with valine utilization by the body.
Thus, in this simple example relating to the utilization of valine, the optimal use
may be prevented by excessive amounts of leucine, isoleucine, calcium, phospho-
rous, or magnesium and deficiencies of riboflavin, folic acid, vitamin B12, panto-
thenic acid, thiamin, or magnesium. For purposes of illustration, this example has
been kept simple. Carbohydrates, fats, and water have not been considered. Neither
have all of the side reactions and the nutrients involved in them that are necessary to
build the molecules needed for the simple primary reaction of converting valine into
usable substance. Nutrient balance is essential for optimum nutrition.
Justus von Liebig (1803–1873) was one of the early investigators of organic, physi-
ological, and agricultural chemistry.29 As a result of his studies, he advanced the law
of the minimum, which states that the nutrient that is the relative minimum deter-
mines the rate of growth.30 This law coupled with Voisin’s law of the maximum (the
nutrient present in the relative maximum determines yield)31 emphasize that both
positive and negative interactions between nutrients exist and that balanced nutrition
can occur at various levels of nutrition.30
The following theoretical example illustrates the possible consequence of ingest-

ing excessive amounts of a specific nutrient. In 1970, Pauling reported that taking
several grams of ascorbic acid on a daily basis prevented the common cold.31 While
several experts disputed the claim of Dr. Pauling,32–34 many laypeople continue to
supplement their diets with large doses of ascorbic acid, which frequently results
in unwanted consequences. Monsen reported that ascorbic acid will enhance non-
heme iron absorption three- to sixfold when consumed concomitantly with the
iron.35 Furthermore, the ascorbic acid will mobilize iron from the ferritin mole-
cule by reducing it from Fe+3 to Fe+2. This becomes significant at concentrations of
50 mM.36 Studies conducted in the United Kingdom have demonstrated that iron is
a very potent antagonist of copper metabolism.37,38 Furthermore, ascorbic acid also
depresses copper bioavailability.39–43 So, the high intake of ascorbic acid could nega-
tively affect copper absorption and metabolism directly through its effect on copper
availability and indirectly by promoting iron uptake and mobilization. Further, even
with the greater uptake of iron, iron deficiency anemia may result due to the role of
copper in ceruloplasmin.
Copper plays many other roles within the body, including formation of bones,
pigmentation of hair, keratinization, prevention of infertility, creation of elasticity in
the cardiovascular system, enhancement of immunity and lipid metabolism. Another
role of copper is facilitating glucose metabolism.44–47 Reduced glucose tolerance is
brought about by reduced lipogenesis and glucose oxidation. Both reductions result
from copper deficiency.48 Under normal conditions, glucose is metabolized at a rate
that maintains a relatively constant concentration of glucose in the blood. Excess,
or unmetabolized, glucose is stored as glycogen and ultimately as fat.49 Thus, when
carbohydrate intake remains constant but glucose metabolism is impaired, fat deposi-
tion increases. Once deposited in the tissue, it becomes more difficult for the body to
metabolize that fat in a copper-deficient state.50,51 Further, in a copper-deficient state
there is elevated serum cholesterol because the cholesterol cannot be degraded.52,53 The
net result of this discussion is that, while the excessive intake of ascorbic acid is not
directly related to weight gain, it could potentially be one of the root causes. Besides
its direct effect on copper, ascorbic acid also enhances iron absorption/metabolism,
which can negatively affect copper absorption and metabolism. That in turn could
potentially have an impact on glucogenesis and result in increased fat deposition in the
tissues. All of these relationships demonstrate Voisin’s law of the maximum.
While this example is a little extreme, a deficiency of a mineral can have a direct
impact on overall metabolic health. If, for example, zinc is deficient in what is other
wise a reasonably balanced diet, it can affect the utilization of the other nutrients
in growth. In 1963, Prasad et al. published their findings relating to zinc deficiency
and its effect on growth or sexual maturation.54,55 Oral zinc treatment over a period
of months corrected the growth retardation and delayed puberty. The other nutrients
necessary for normal growth and sexual maturity were previously present in the diet,
but the deficiency (not a complete absence) of a single essential nutrient significantly
reduced the efficacies of the other nutrients. This clearly demonstrates Liebig’s law
of the minimum.
In the case of minerals, it is extremely difficult to predict with any certainty the per-
centage of absorption following ingestion. The normal absorption of calcium has been
reported to range between 20% and 50% of the dose. Magnesium has an even wider
range: 25% to 75%. Normal absorption of iron salts is reported to be between 2%
and 10%. Manganese absorption fluctuates between 3% and 20% of the dose. Copper
absorption may be as low as 10% or as high as 97% according to the study consulted.56
Much of this controversy focuses on differing environmental conditions that may
influence the absorption of a specific ion at a specific time. While there is some jus-
tification for that position, even under controlled conditions intestinal absorption of
metal ions can vary depending on the source of the mineral.57,58
Brise and Hallberg conducted a study in 80 human volunteers in which they com-
pared the absorption of nonheme iron from 12 sources to ferrous sulfate absorption.
They dissolved 30 mg of iron from one of the 12 salts tagged with 55Fe in 25 mL of
distilled water. They then added 10 mg of ascorbic acid to each solution to prevent
oxidation and to enhance absorption of the iron. A ferrous sulfate solution was simi-
larly prepared, except the iron was labeled with 59Fe. The volunteers consumed the
iron salt solutions assigned to them on the first day. The next day, they all took the
ferrous sulfate solution. The following day, they took the iron salt solutions assigned
to them. Treatments continued for 10 days and alternated daily between the two iron
solutions. Each solution was administered in the morning. Following the last dose
on the 10th day, blood samples were obtained from each individual and assayed for
59Fe and 55Fe. Figure 1.6 summarizes the results. The absorption of each iron salt
was compared to ferrous sulfate, which was arbitrarily set at 100%.58 As can be
seen, both the valence of the iron and the anion attached to the iron ion influenced
the absorption of the iron.
When considering the law of the minimum and applying it to the six basic nutri-
ent groups, minerals tend to be the most limiting. As seen in the previous examples,
140
120
Compared to FeSO4
Percent Absorption
100
80
60
40
20
0
e
te
e
e
e
e
e
ph ate
ou s f u t e
te
ou ous ate
te
TA
at
at
at
at
at
at
at
fa
ta
fa
lfa
tr
ph
on
ar
m
n
yr artr
tr
itr
D
l
ul
u
i
su
ci
ci
a
cc
ta
eE
sc
os
s
c
ss
sl
iso
lu
t
lu
Fe cine
ic
su
ic
aF
ou
u
ou
sg
sg
rr
rr
in
ro
N
us
o
rr
ou
Fe
rr
Fe
ol
ly
ou
ou
r
r
ro
Fe
Fe
er
Fe
sg
ch
rr
sp
rr
rr
r
F
Fe
Fe
Fe
ic
rr
rr
rr
Fe
Fe
Fe
FIGURE 1.6 The percentage of iron absorption in human subjects from different sources
compared to ferrous sulfate. (Redrawn from Brise, H, and Hallberg, L, “Absorbability of dif-
ferent iron compounds,” Acta Med Scan Suppl. 358–366, 23–37, 1960.)
bioavailability can change depending on the source of the metal. There are numerous
other factors that can also affect bioavailability. The fact that a mineral is present in
the diet does not guarantee it is bioavailable. Intrinsic, extrinsic, and luminal factors
all influence mineral bioavailability. Table 1.3 summarizes these factors in mam-
mals, including humans.59
TABLE 1.3
Factors Affecting Mineral Bioavailability
Intrinsic Factors
1. Animal species and its genetic makeup
2. Age and sex
3. Monogastric or ruminant (intestinal microflora)
4. Physiological function: growth, maintenance, reproduction
5. Environmental stress and general health
6. Food habits and nutrition status
7. Endogenous ligands to complex metals (chelates)
Extrinsic Factors
1. Mineral status of the soil on which the plants are grown
2. Transfer of minerals from soil to food supply
3. Bioavailability of mineral elements from food to animal
a. Chemical form of the mineral (inorganic salt or chelate)
b. Solubility of the mineral complex
c. Absorption on silicates, calcium phosphates, dietary fiber
d. Electronic configuration of the element and competitive antagonism
e. Coordination number
f. Route of administration, oral or injection
g. Presence of complexing agents such as chelates
h. Theoretical (in vitro) and effective (in vivo) metal binding capacity of the chelate for the
element under consideration
i. Relative amounts of other mineral elements
In the Lumen
1. Interactions with naturally occurring ligands
a. Proteins, peptides, amino acids
b. Carbohydrates
c. Lipids
d. Anionic molecules
e. Other metals
2. At and across the intestinal membrane
a. Competition with metal-transporting ligands
b. Endogenously mediating ligands
c. Release to the target cell
Source: From Kratzer, F, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 35, 1986.
Some of the factors are outside the ability of the organism to change, such as
genetics, age, or sex. Others, like endogenous ligands in the diet that can complex
minerals, can be modified by changes in the diet. There are millions of ligands in the
body. The ingested metals can be attached through covalent or ionic bonding to one,
two, or many ligands within the biological system. Changes in diet can change the
makeup of the ligands in the gastrointestinal tract, which in turn affects the binding
of the metal ions also present in the lumen.
To illustrate this concept, one needs only to consider the absorption of ferrous
(Fe+2) ion resulting from the ingestion of ferrous sulfate. When the diet is rich in phy-
tates, the intestinal absorption of the iron is reduced. The ferrous ion formed in the
lumen is complexed by the phytic acid from the phytates, resulting in creation of an
insoluble iron complex.60,61 On the other hand, if the diet is rich in animal proteins,
such as beef, pork, lamb, chicken, or fish, there is pronounced enhancement of the
intestinal absorption of the iron.62,63
This discussion focused on absorption variability resulting from ingestion of
different sources of the same metal. Consuming different foods can cause signifi-
cant differences in the absorption of the same mineral from that food. For example,
Layrisse and Martinez-Torres determined the percentage of iron absorption from
various foodstuffs.63 As shown in Figure 1.7, iron from vegetable origins was not as
bioavailable as was iron from animal origins. Even so, there were wide variations
within the two food groups. Availability of iron from rice was less than 1%, whereas
iron from soybean was more than seven times greater. Iron from ferritin and hemo-
globin had lower absorptive values than did veal liver, but all of them were eclipsed
by the absorption of iron from veal muscle.
These data suggest that the chemical presence of a mineral in a food is no guar-
antee of its availability. What the mineral is bound to affects its absorption. For
Vegetable Origin Animal Origin

Veal Hemo- Fish Veal
Rice Spinach Beans Corn Lettuce Wheat Soybean Ferritin
liver globin muscle muscle
20
15
Percent of Iron Absorption*
10
1
* All values are expressed as means plus or minus one standard deviation
FIGURE 1.7 Absorption of iron from food. (Redrawn from Layrisse, M, Martinez-Torres,
C, and Roche, M, “Effects of interaction of various foods on iron absorption,” Am J Clin Nutr
21:1175–1183, 1968.)
example, calcium that is located in the wall of a plant cell is much less bioavailable
compared to calcium in the cytoplasm of the plant cell. Further, as noted in Table 1.2,
precipitating ligands in the food can, and often do, bind with ions that are freed from
the food during digestion even if those metals were not originally attached to the
precipitating ligands in the food.64
The health of the individual will also affect his or her ability to absorb and, more
particularly, utilize absorbed minerals in a metabolic process. Besel published a
schematic representation of the sequence of nutritional responses resulting from con-
tracting an infectious disease.65 The very first response following exposure is depres-
sion of plasma amino acids, zinc, and iron. This is followed by retention of urinary
zinc. Ultimately, at the height of the illness, there is a negative balance of zinc,
magnesium, sodium, and potassium (Figure 1.8).
There have been numerous books and countless articles written on mineral
absorption. Most try to explain why there are variations and how to maximize
absorption through selection of specific foodstuffs or use of certain salts. This work
is another attempt. It will focus on ingestion of metal amino acid chelates as a way
to optimize mineral absorption.
Phagocytic activity
Depression of plasma amino acids, Fe and Zn
Saluresis retention of urinary PO4 and Zn
Increased secretion of glucocorticoids and growth hormone
Increased deiodination of thyroxine
Increased synthesis of hepatic enzymes
Secretion of “acute phase” serum proteins
Carbohydrate intolerance
Increased dependence on lipids for fuel
Increased secretion of aldosterone and ADH
Negative Balances Begin – N, K, Mg, PO4, Zn, and SO4
Retention of body salt and water
Increased secretion of thyroxine
Diuresis Return to positive balances
Fever
Incubation Illness Convalescent

Period Period
Moment of Exposure
FIGURE 1.8 Schematic representation of the sequence of nutritional responses that evolve
during the course of a “typical” generalized infectious illness. (Redrawn from Besel, WR,
“Magnitude of the host nutritional responses to infection,” Am J Clin Nutr 30:1236–1247, 1977.)
REFERENCES
1. Wilson, ED, Fisher, KH, and Garcia, PA, Principles of Nutrition (New York: Wiley)
4, 1979.
2. Ashmead, HD, Conversations on Chelation and Mineral Nutrition (New Canaan, CT:
Keats) 26, 1989.
3. Brody, T, Nutritional Biochemistry (San Diego, CA: Academic Press) 35, 1994.
4. Schutte, KH, The Biology of the Trace Elements (Philadelphia: Lippincott) 15, 1964.
5. Guthrie, HA, Introductory Nutrition (St. Louis, MO: Mosby) 200, 1975.
6. Ibid., 13.
7. MacInnis, A, and Graff, DJ, “Specificity of amino acid transport in the tapeworm,
Hymenolepis diminuta, and its rat host,” Rice Universities Studies 62:183, 1976.
8. Dyer, IA, “Mineral requirements,” in Hafez, ESE, and Dyer, IA, eds., Animal Growth
and Nutrition (Philadelphia: Lea & Febiger) 313, 1969.
9. Morrison, FB, Feeds and Feeding, Abridged (Clinton, IA: Morrison) 49, 1961.
10. Graff, D, “Radioactive isotope research with chelated minerals,” in Ashmead, D, ed.,
Chelated Minerals Nutrition in Plants, Animals and Man (Springfield, IL: Thomas)
275, 1982.
11. Sauberlich, H, “Interactions of thiamine, riboflavin and other B-vitamins,” in Levander,
O, and Cheng, L, eds., Micronutrient Interactions: Vitamins, Minerals and Hazardous
Elements (New York: New York Academy of Sciences) 80, 1980.
12. Arnich, L, and Arthur, V, “Interaction of fat soluble vitamins in hypervilamenases,” in
Levander, O, and Cheng, L, eds., Vitamins, Minerals and Hazardous Elements (New
York: New York Academy of Sciences) 109, 1980.
13. Patrick, H, and Schaible, P, Poultry Feeds and Nutrition (Westport, CT: AVI) 144, 1980.
14. Levander, O, and Cheng, L, eds., Vitamins, Minerals and Hazardous Elements (New
York: New York Academy of Sciences) 80–129, 1980.
15. Underwood, E, “The current status of trace elements: an overview,” paper presented at
International Minerals Conference, St. Petersburg Beach, Florida, January 17, 1978.
16. Suttle, NF, “Trace element interactions in animals,” in Nicholas, DJD, and Egan, AR,
eds., Trace Elements in Soil-Plant-Animal Systems (New York: Academic Press)
278–285, 1975.
17. Waddell, DG, and Sell, JL, “Effects of dietary calcium and phosphorus on the utilization
of dietary iron by the chick,” Poultry Sci 43:1249–1257, 1964.
18. Vohra, P, and Kratzer, FH, “Influence of various phosphates and other complexing
agents on the availability of zinc for turkey poults,” J Nutr 89:106–112, 1966.
19. Ashmead, HD, and Ashmead, SD, “The effects of dietary molybdenum, sulfur and iron
on absorption of three organic copper sources,” J Appl Res Vet Med 2:1–9, 2004.
20. Starcher, B, “Studies on the mechanism of copper absorption in the chick,” J Nutr
97:321–326, 1969.
21. EI-Shobaki, F, and Rummel, W, “Binding of copper to mucosal transferrin and inhibi-
tion of intestinal iron absorption in rats,” Res Exp Med 174:187–195, 1989.
22. Ashmead, H, “Tissue transportation of organic trace minerals,” J Appl Nutr 22:42–51,
Spring 1970.
23. Herrick, JB, “Minerals in animal health,” in Ashmead, HD, ed., The Roles of Amino Acid
Chelates in Animal Nutrition (Park Ridge, NJ: Noyes Publications) 3–20, 1993.
24. Schutte, KH, The Biology of Trace Elements (Philadelphia: Lippincott) 17–23, 1964.
25. DeLuca, H, “Vitamin D and calcium transport,” Ann NY Acad Sci 307:356–376, 1978.
26. White, A, Handler, P, and Smith, EL, Principles of Biochemistry (New York: McGraw-
Hill) 219, 1973.
27. Frieden, E, “Ceruloplasmin, a link between copper and iron metabolism,” Adv Chem
100:292–321, 1971.
Hill) 339, 556, 668, and 731, 1973.
29. McCullum, EV, A History of Nutrition (Boston: Houghton Mifflin) 92, 1957.
30. Schutte, KH, The Biology of Trace Elements (Philadelphia: Lippincott) 43, 1964.
31. Voisin, A, Soil Grass and Cancer (London: Crosby Lockwood) 1959.
32. Pauling, L, Vitamin C and the Common Cold (San Francisco: Freeman) 1970.
33. Hodges, RE, “What’s new about scurvy?” Am J Clin Nutr 24:383–384, 1971.
34. Wilson, CWM, “Ascorbic acid function and metabolism during colds,” Br Med J,
1:669–677, 1971.
35. Diehl, HS, “Vitamin C for colds,” Am J Public Health 61:649–651, 1971.
36. Monsen, ER, “Ascorbic Acid: An enhancing factor in iron absorption,” in Kies, C, ed.,
Nutritional Bioavailability of Iron (Washington, DC: American Chemical Society)
85–95, 1982.
37. Beinfait, HF, and van Den Briel, ML, “Rapid mobilization of ferritin iron by ascorbate
in the presence of oxygen,” Biochem Biophys Acta 631:507–510, 1980.
38. Humphries, WR, Bremner, I, and Phillippo, M, “The influence of dietary iron on copper
metabolism in the calf,” in Mills, CF, Bremner, I, and Chesters, JK, eds., Trace Elements
in Man and Animals—TEMA 5 (Aberdeen, Scotland: Commonwealth Agricultural
Bureaux) 371–374, 1985.
39. Bremner, I, and Price J, “Effects of dietary iron supplements on copper metabolism
in rats,” in Mills, CF, Bremner, I, and Chesters, JK, eds., Trace Elements in Man
and Animals—TEMA 5 (Aberdeen, Scotland: Commonwealth Agricultural Bureaux)
374–376, 1985.
40. Carlton, WW, and Henderson, W, “Studies in chickens fed a copper deficient diet sup-
plemented with ascorbic acid, resperine and diethylstilbestrol,” J Nutr 85:67–72, 1965.
41. Finley, EB, and Cerklewski, FL, “Influence of ascorbic acid supplementation on copper
status in young adult men,” Am J Clin Nutr 37:553–556, 1983.
42. Hill, CH, and Starcher B, “Effects of reducing agents on copper deficiency in the chick,”
J Nutr 85:271–274, 1965.
43. Howell, J, Mc Edington, N, and Ewbank, R, “Observations on copper and caeruloplas-
min levels in the blood of pregnant ewes and lambs,” Res Vet Sci 5:160–164, 1968.
44. Van Camper, DR, and Gross, E, “Influence of ascorbic acid on the absorption of copper
by rats,” J Nutr 95:617–622, 1968.
45. Cohen, AM, Tetiebaum, A, Miller, E, Ben-Tor, V, Hirt, R, and Fields M, “Effect of cop-
per on carbohydrate metabolism in rats,” Isr J Med Sci 19:840–842, 1982.
46. Hassel, CA, Marchello, JA, and Lei, KY, “Impaired glucose tolerance in copper-defi-
cient rats,” J Nutr 113:1081–1083, 1983.
47. Klevay, LM, “An increase in glycosylated hemoglobin in rats deficient in copper,” Nutr
Rep Int 26:329–334, 1982.
48. Kelvay, LM, Canfield, WK, Gallagher, SK, Hendriksen, RD, Lukaski, HC, Bolonchuk,
W, Johnson, LK, Miline, DB, and Sandstead, HH, “Decreased glucose tolerance in two
men during experimental copper depletion,” Nutr Rep Int 33:371–382, 1986.
49. Fields, M, Ferretti, RJ, Smith, JC, and Reisser, S, “Impairment of glucose tolerance
in copper-deficient rats: Dependency on the type of dietary carbohydrate,” J Nutr
114:393–397, 1984.
50. Szepesi, B, “Carbohydrates,” in Zingler, EE, and Filer, LJ, eds., Present Knowledge in
Nutrition (Washington, DC: ILSI Press) 36–38, 1996.
51. Cunnane, SC, Horrobin, DF, and Manku, MS, “Contrasting effects of low or high
copper intake on rat tissue lipid essential fatty acid composition,” Ann Nutr Metab
29:103–110, 1985.
52. Wahle, KWJ, and Davies, NT, “Effect of dietary copper deficiency in the rat on fatty acid
composition of adipose tissue and desaturase activity of liver microsomes,” Br J Nutr
34:105–112, 1975.
53. Lei, KY, “Alterations on plasma lipid, lipoprotein and apoliprotein concentrations in
copper-deficient rats,” J Nutr 113:2178–2183, 1983.
54. Prasad, AS, Miale, A, Farid, Z, Sanstead, HH, Schulert, AR, and Darby, “Biochemical
studies on dwarfism, hypogonadism,” Arch Intern Med 111:407–428, 1963.
55. Prasad, AS, Schulert, AR, Miale, A, Farid, Z, and Sanstead, HH, “Zinc and iron defi-
ciencies in male subjects with dwarfism but without ancyclostomiasis, schistosmiasis or
severe anemia,” Am J Clin Nutr 12:437–444, 1963.
56. Ashmead, HD, Graff, DJ, and Ashmead, HH, Intestinal Absorption of Metal Ions and
Chelates (Springfield, IL: Thomas) 24–25, 1985.
57. Graff, DJ, Ashmead, H, and Hartley, C, “Absorption of minerals compared with chelates
made from various protein sources into rat jejunal slices in vitro,” Proc Utah Acad Arts
Lett Sci Apr 1970.
58. Brise, H, and Hallberg, L, “Absorbability of different iron compounds,” Acta Med Scan
Suppl 358–366, 23–37, 1960.
59. Kratzer, F, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 35, 1986.
60. Hallberg, L, and Solvell, L, “Absorption of a single dose of iron in man,” Acta Med Scan
19:358, 1960.
61. Manis, J, and Schachter, D, “Active transport of iron by intestine: Features of the two-
step mechanism,” Am J Physiol 203:73–80, 1962.
62. Johnston, FA, Frechman, R, and Burroughs, ED, “The absorption of iron from beef by
women,” J Nutr 35:453–465, 1948.
63. Layrisse, M, Martinez-Torres, C, and Roche, M, “Effects of interaction of various foods
on iron absorption,” Am J Clin Nutr 21:1175–1183, 1968.
64. Kuhn, LC, Schulman, HM, and Ponka, P, “Iron-transferrin requirements and transferring
reception expression in proliferating cells,” in Ponka, P, Schulman, HM, and Woodward,
RC, eds., Iron Transport and Storage (Boca Raton, FL: CRC Press) 149–177, 1990.
65. Besel, WR, “Magnitude of the host nutritional responses to infection,” Am J Clin Nutr
30:1236–1247, 1977.
2 The Chemistry
of Chelation
In 1893, Alfred Werner authored a paper of major chemical significance. Breaking
with traditional chemical thinking, he proposed an entirely new molecular structure
to describe atoms that could exist in more than one valence state and form highly
stable compounds. Werner noted that certain structural entities, which he called
“complexes,” would remain intact through a series of chemical transformations; he
wrote, “If we think of the metal ion as the center of the whole system, then we can
most simply place the molecules bound to it at the corners of an octahedron.”1 He
used platinum bis-ethylenediamine chloride to illustrate his theoretical concept.
In the ensuing years Werner refined his discovery. He concluded that a metal ion
had two kinds of valencies. The first, which Werner termed the “principal valence,”
related to the oxidation state or oxidation number of the metal. He called the second
valence the auxiliary valence, which referred to the number of atoms in the ligand
(chelating agent) that could be associated with the central metal ion.2–7 Today, the
auxiliary valence is referred to as the coordination number. Thus, the platinum ion
illustrated in Figure 2.1 has a principal valence (oxidation number) of 2+ and an aux-
iliary valence (coordination number) of 4+. In this configuration, the platinum metal
lost two electrons in the valence shell, becoming Pt2+, leaving room for four electrons
or four bonds in the valence shell. Thus, the platinum could be bound to the two
ethylenediamine ligands at four points. According to Werner, the ion was attached
to the ligands by principal valencies in the outer sphere of the combination and the
amine groups to the central atom in an inner sphere of combination.8 As a result of
his pioneering research, Alfred Werner received the Noble Prize in chemistry in
1913 and has subsequently been called the father of coordination chemistry.9
While Werner is generally credited with the discovery of chelation chemistry, he
did not use the word chelation to characterize these complexes. In 1920, Morgan and
Drew coined the word chelate to describe the way the ligand bound a metallic cation.
The word chelate comes from the Greek word chele meaning “claw.” As indicated in
Werner’s example in Figure 2.1, each ligand is able to attach to the metal ion at two
points in a claw-like fashion. Morgan and Drew reasoned that this caliper-like action
of the ligand resembled the closing of a lobster or crab claw and suggested the term
to describe metal complexes in which the metal atom is held at more than one point
of attachment by a single ligand.10 Figure 2.2 illustrates a metal glycinate chelate.
When one looks at the ligand of this chelate, it is easy to see how it could be viewed
as a claw as it chelates the metal ion.
The word chelate was originally used as an adjective, but today it is also employed
as an adverb, verb, or noun. The ligands are the chelating agents, and the metal-
ligand compounds that they form are metal chelates.
19
NH2 NH2
H2C CH2
Pt Cl2
H 2C CH2
NH2 NH2
FIGURE 2.1 Platinum bis-ethylenediamine chloride structure as proposed by Alfred

Werner. (Redrawn from Werner, A, “Beitrage zur konstitution anorganischer verbindungen,”
Z Anorg u Allgem Chem 3:267–330, 1893. Translated by D. P. Mellor.)
O
O NH2 NH2
C CH2 CH2
M M
H2C C C
H 2N O
O O
O
FIGURE 2.2 A metal bisglycine chelate illustration showing the claw-like structure of
the ligand.
The purpose of this chapter is to review, on an elementary level, the chemical

characteristics of a chelate. Any metal that is chelated must meet specific criteria.
If it does not meet those criteria, then it is not a chelate but may instead be a com-
plex in which the critical (to chelation) ring structure is not formed.11 Even if the
mineral is chelated, the resulting molecule is not guaranteed to enhance mineral
absorption. The criteria for a nutritionally functional chelate are discussed in the
next chapter.
Prior to the defining work of Morgan and Drew, Ley, another chemist, commenced
the elucidation of the presumed structure of a chelate.12 He was able to synthesize a
copper bisglycinate molecule by bonding two moles of glycine to one mole of copper.
Ley inferred that a chemical reaction had occurred in the solution of copper ions and
glycine due to observing a color change. He also noted that the resulting solution had
very low electrical conductivity compared to the conductivity of copper ions in solu-
tion. When the copper-glycine solution was dried and the precipitate analyzed, it was
determined that the product consisted of a ratio of two moles of glycine to one mole
of copper. Ley had created an amino acid chelate, although he did not call it that.
Instead he gave it the name “inner metallic complex salt” and illustrated the concept
(not copper glycinate) as seen in Figure 2.3.
The nature of bonds between the metal ion and ligand is crucial to comprehend-
ing chelation. The basic principles of the bond formatting are similar for transition
metals (elements with partially filled d or f shells) as they are for others. It is the
d orbital in the electron shells of the transition metals that plays a major role in
The Chemistry of Chelation 21
0
H
O O
H 3C N N CH3
C C
M
C C
H3C N N CH3
O O
H
(x)
FIGURE 2.3 Inner metallic complex salt illustration as conceived by H. Z. Ley. (Redrawn
from Ley, H, “Über Innere Metall-Koomplexsalze. I” Z Elektrochem Angew Physikal Chem
10:954–956, 1904.)
bonding. Further, bonding in the d orbital results in different characteristics com-

pared to elements where the bonding occurs exclusively in the s or p orbitals. The
way the metals are bonded is called ligand field theory.13,14
According to this theory, when transitional metals are oxidized, if they have elec-
trons in their s shells, they generally lose those outer s shell electrons before losing
the d shell electrons. (Some transitional metals have partially filled d orbitals.) It is
the existence of the d subshell electrons and their movement that gives the transi-
tional metals their unique characteristics, such as having more than one oxidation
state, colored compounds, and magnetic properties.14,15 It also allows them to partici-
pate in bonds that are generally unlikely for nontransitional elements.14
While over the ensuing years the following chelation criteria have been refined,
they were recognized10 as early as 1920:
1. In chelation, there are two types of ion ligand bonding: ionic or electrostatic
bonding, in which both atoms participating in the bond each share electrons
to form the bond, and covalent bonding, in which the bonding between the
two atoms occurs when both electrons shared in the bond originate from
the same atom.
2. The same anion ligand may participate in either type of bonding.
3. The metal that participates in the complex has a fixed number of valencies,
one of which is called its coordination number.
4. Coordinate covalent bonds (also known as dative or semipolar bonds) may
be formed with either neutral or ionic entities.
5. Coordinate covalent bonds have definite spatial arrangements.
Concurrent to Ley, Bruni and Fornara were also working with copper glycinate and
arriving at similar conclusions.16 Most chelates produced by early investigators were
insoluble or poorly soluble. Because the copper glycinate had different characteristics
than most of the chelates these investigators were studying, they usually overlooked
it in favor of focusing their research on less-soluble chelates made with synthetic
ligands, which they relied on to describe chelated molecules. Nevertheless, the prin-
ciples resulting from this research with non-amino-acid ligands can be applied to
amino acid chelates. Chelation is a special branch of chemistry, and the principles
governing this chemistry govern all chelates, regardless of the ligand employed. That
does not mean all chelates behave the same. The similarity in chelates only relates to
the type of chemistry required to form the chelate molecule.
A chelate has several distinct characteristics that can be examined based on the
metal atom in the chelate molecule, the ligand used to chelate the metal, or the types
of bonds linking the metal ion and the ligand together.17
Returning to the first criterion, the properties of a chelate are influenced, to a
degree, by the metal in a chelate molecule and its oxidation state. The metals may be
considered as Lewis acids, whereas the ligands are regarded as Lewis bases, which
can share an electron pair with the metal. When bonded together, the result is a
neutralization of the Lewis acid and Lewis base. Depending on the characteristics
of the ligand, it may be attached to a metal through two, three, four, five, or six posi-
tions—creating a bidentate, tridentate, quadridentate, quindentate, or sexadentate
chelate, respectively.18
The bond between a metal atom and the ligand may be electrostatic (ionic) or
covalent. The ionic bonds are a result of attraction between oppositely charged ions.
In a metal chelate bond, the positive charge originates from the metal ion and the
negative charge from a negatively charged atom in the ligand. According to one
theory, after the ionic bond is created through ionic attractions, a sharing of elec-
trons between the metal ion and the reactive moiety of the carboxyl group occurs,
and the bond becomes covalent in nature. The argument to support this view is that
if the bond were purely ionic, it would dissociate in an aqueous solution. Had this
occurred, Ley would have observed more electrical conductivity in the copper gly-
cinate chelate than was created.12 Thus, some chemists believe that no ionic bond
exists in a chelate. Instead, they favor the view that all of the bonds are coordinate.
The initial attraction between the oxygen in the carboxyl moiety is a charged attrac-
tion. This bond will occur in a mostly dry environment or dry mixed with a little
pressure. The amine moiety, which definitely forms a covalent bond, does not bond
to the metal under similar conditions. Further, in a solution the first bond to form is
the carboxyl bond, followed by the amine bond. This suggests that an ionic bond can
potentially exhibit covalent characteristics.14
After the amine moiety in the ligand bonds to the metal, the oxygen is placed in
a spatial orientation that potentially endows it with some covalent characteristics.
Conrad and Nakamoto reported that, following chelation of copper with glycine, the
strength and shift of the amine moiety indicated that the bond between the copper
and the nitrogen in the amino moiety was covalent. On the other hand, the changes in
the carboxyl moiety indicated that the bond between the copper and the oxygen was
more ionic in nature even though the oxygen in the chelate occupied a similar posi-
tion as the nitrogen and still had a covalent nature.14,19 While there is room to argue
on both sides of the issue, the position taken in this chapter is that the bond between
the metal and the atom from the carboxyl moiety is primarily ionic in nature.
In a covalent bond, the metal ion and an atom in the ligand share a pair of elec-
trons. In the past, if both electrons were donated by an atom in the ligand into a
vacant d orbital of the metal ion, the bond between them was referred to as a coordi-
nate covalent bond.20 Today, it is generally called a covalent bond.
In an aqueous solution, the coordination number of a metal ion is generally satis-
fied by water molecules and could be considered coordinate positioning. When the
water molecules are displaced by a soluble ligand, a different metal complex can
result. This complex may be a chelate if certain requirements are met. If the ligand
in the solution has more than one donor atom and the metal ion has a coordination
number of 2 or more, the resulting complex can form a heterocyclic ring. In this case,
the metal will be the closing member of the ring. A heterocyclic ring must be formed
for the resulting compound to be a chelate.18 In fact, a heterocyclic ring is an absolute
requirement for the formation of a chelate. If this ring structure is not produced in
the chemical reaction, the resulting product cannot be a chelate.
The ligand, whether it is synthetic or natural, must possess at least two functional
groups, one of which is capable of donating a pair of electrons to combine with the
metal and the other capable of sharing an electron with the metal.17 If the ligand does
not have these functional groups, chelation cannot occur. In the case of the amino
acid, glycine, for example, the chemical formula is NH2CH2COOH. To chelate a
metal ion, the glycine must lose its carboxyl-group proton (H+) and chelate as a gly-
cine ion via two functional groups, one of which contains the nitrogen atom and the
other, one of the oxygen atoms (Figure 2.4).
The bond between the metal ion and the oxygen from the COOH group is ionic
(electrostatic) because the metal and the amino acid share one electron from the oxy-
gen in the carboxyl group of the amino acid and one electron from the metal ion.21
The second bond between the metal and the nitrogen in the NH2 group is a covalent
bond. As noted, the metal behaves as a Lewis acid and the glycine as a Lewis base.
The glycine donates both electrons from the same atom in the amine group of the
ligand to the metal ion. The donation of electrons will go to the lowest energy orbital
(s, p, and d orbitals) of the metal that is unfilled.21
Since the metal ion in this example has a +2 charge, it is capable of bonding to
more than one amino acid. It shares one electron from the carboxyl group of the
amino acid and accepts two electrons from the amine group. When the metal ion
has a +2 charge, it still has an electron deficiency after chelating to a single ligand.
Thus, to be satisfied, it can accept another electron from another carboxyl moiety of
a second amino acid ligand such as another glycine molecule. This second ligand can
NH2 O O
O
H2C C
2H2N CH2–C OH + M+2 M
C CH2
O H2N
O
FIGURE 2.4 The chelation of a metal with two glycine molecules to form bisglycinate
chelate.
OH
O H C H H
H H
C C N H
O N C
C N H H C
C O M
O C C O
H C H H H2C H H
CH2 C H 3C
N H H HC H
2 H H N
O C H CH2 H3C
N H C C SH
H C H 2N H
H C C C C C H
O C
C C N H
H N H H H
H
C O C C
H H3C S C C C
H 3C C H H
HH N H H H
C O
O C H H
HC C CH3 O
C N 3 H N
C C H H
H 2C H C N C H
H N CH3
O C C C
C C C
H H
C H H C H H
O
C H3C
C
H C C
H H H C N H
N C
M = Metal Atom
H
FIGURE 2.5 The theoretical chelation of a metal ion by ligand composed of 12 amino
acids. This chelate molecule is highly improbable because ligands with reactive moieties
more than six atoms away are generally more likely to form metal complexes with separate
metal ions on each reactive moiety than to chelate back to the original metal ion. (Redrawn
from Jeppsen, RB, “Proteinates vs. amino acid chelates,” paper presented at International
MAAC® Nutritional Conference, Salt Lake City, UT, February 19, 1994.)
also donate a pair of electrons from its amine group. Thus, two ring structures can be
formed, with the single metal ion the closing member of both rings.
The greater the number of rings attached to a metal ion, the more stable the che-
late molecule becomes, up to a point. The electron configuration of the metal will
limit its attachments. Spatial arrangements can be tetrahedral, planar, and so on,
depending on the metal.18 This is referred to as the chelation effect.22 In addition,
steric hindrance is a key consideration. If the ligand is a large molecule, such as
the theoretical molecule illustrated in Figure 2.5, only one ring may be possible.23
Chelating the metal ion with a second or third ligand to form additional rings may
be impossible. If the ligand is small, such as in the case of glycine, it becomes easier
to attach two, three, and sometimes more ligands to the ion and form more than one
ring depending on the valence of the cation involved and the size of the metal atom.
Chelate stability is affected not only by the choice of the metal being chelated but
also by the type of ligand selected for chelating purposes. Chelates can be classified
as multidentate or bidenate depending on the number of donor atoms in the ligand.
Multidentate chelates, chelates in which there are more than two donor atoms in the
ligand, are more stable than bidentate ligands (ligands that have two basic groups,
one acidic and one basic group, or two acidic groups).24 A multidentate chelate is able
to occupy more positions in the coordination shell of the metal ion.25
The coordinating groups in a ligand that are capable of donating electrons to
combine with the metal are18
a. Coordinating groups:
=O –OR –AsR2
–NH2 –NOH –PR2
–NH –OH (alcoholic)
–N= –S– (thioether)
b. After the loss of a proton:

–COOH –NH3 –OH –SH
OH
∕
–SO3 –N–H –N=O –P=O
| | \
R H OH
The atoms within the coordinating groups that generally bond to the metal in a che-
late are N, O, C1, P, S, Br, As, Se, or I.18
Besides having functional groups that are capable of donating a pair of electrons,
the functional groups in the ligand must be located in such a way that they allow
for the formation of a ring structure with the metal as the closing member of that
ring.17 The potential ring formation is greatly influenced by steric hinderances. For
example, the attachment of one functional group may result in too much bulk and
thus prevent the attachment of a second functional group to the metal ion.18 In the
case of the amino acid ligands, the side chains that are attached to the a-carbon may
provide additional hindrance.
The Association of American Feed Control Officials (AAFCO) is an organiza-
tion composed of all 50 U.S. state chemists, the U.S. Food and Drug Administration,
the Canadian Food Inspection Agency, the Costa Rica Ministry of Agriculture
and Livestock, the U.S. Department of Agriculture, and the U.S. Environmental
Protection Agency (EPA) plus nonvoting industry members. The AAFCO publishes
a list of definitions covering all approved feed ingredients, including two chelated
mineral sources: metal proteinates and metal amine acid chelates. Metal proteinates
are “the product resulting from the chelation of a soluble salt with amino acids and/or
hydrolyzed protein.”11 Presumably since that definition requires chelation with amino
acids and/or hydrolyzed protein, more than one amino acid must be employed in the
chelation of a single metal ion. The use of the words and/or presumably requires
that partially hydrolyzed protein be part of the chelating ligands. Under this defini-
tion, the partially hydrolyzed protein could be the source of the amino acids, but
individual amino acids could not be the sources. Such a partially hydrolyzed protein
ligand would result in steric hindrances that could potentially interfere with the che-
lation of the metal with additional ligands, such as illustrated in Figure 2.5. Such a
chelate probably could not be absorbed through the intestinal mucosal membrane
intact due to its molecular size.23 It would require further gastrointestinal digestion
into a smaller molecule before any absorption could occur.
The AAFCO has defined the metal amino acid chelate as
the product resulting from the reaction of a metal ion from a soluble metal salt with
amino acids with a mole ratio of one mole of metal to one to three (preferably two)
moles of amino acids to form coordinate covalent bonds. The average weight of the
hydrolyzed amino acids must be approximately 150 and the resulting molecular weight
of the chelate must not exceed 800.11
In expanding this definition, the AAFCO has added that “one ligand (electron pair
donor) forms two or more bonds to the central metal ion through different atoms
of the ligand. A distinctive feature of a metal is a member of the ring.”11 It is this
complete definition that will be used to discuss amino acid chelates throughout the
remainder of this book except as otherwise noted.
Besides the ligands used to chelate a metal, the choice of the metal atom in the
chelate will also influence the stability of the resulting chelate. The electronic struc-
ture of the metal ion will determine, first, if a chelate can be formed and, second,
the stability of the chelate if it is formed. Monovalent metals generally cannot form
chelates under the definition of binding the ion at two points. A monovalent metal
ion, such as potassium, does not have the electronic orbital structure to bind two sites
of the ligand. If the metal is not bound at two sites, no ring structure can be formed
since, by definition, the metal is the closing member of the ring structure in a chelate,
and at least two bonding sites on the metal ion must be available to close the ring.
The oxidation number of the metal ion can also affect the stability of the chelate.
The higher the valency, the more stable the chelate, assuming there is a sufficient
number of ligands to satisfy all of the charges on the ion and there is adequate space
for all of the ligands to bond to the metal ion. Therefore, the size of the metal ion
is important. Metal ions with smaller radii cannot bond to as many ligands as can
larger cations. Thus, chelates formed from transition metals as well as lanthanides
and actinides are more stable than are alkali metal chelates.26 In the case of transition
elements, the stability increases to a maximum of copper Ni2+ < Co2+ < Fe2+ < Mn2+
< Zn2+ < Cu2+.27 As a general rule, as the atomic number of the metal increases, the
stability of the resulting chelate also increases. It should be remembered, however,
that the smaller the metal ion, the smaller must be the practical coordination number
due to steric hindrance regardless of other considerations.
The total number of ligand atoms that can be bound to the metal ion represents
the coordination number of that metal. The coordination number of the metal will
influence the stability of the chelate. The higher the coordination number, the more
stable the chelate because, in theory, more ligands can be attached to the metal as
long as there are no steric hindrances.
As the coordination number changes, so do the steric considerations of the result-
ing chelate molecule. As additional ligands are bonded to the metal ion, the chelate
will change shape to accommodate the new ligands. This affects the stability of the
TABLE 2.1
Stereochemistry and Oxidation States of Some 3D Elements
Oxidation Coordination Oxidation Coordination
State Number Stereochemistry State Number Stereochemistry
CuI(d10) 2 Linear CoIII(d6) 4 Tetrahedral
3 Planar 5 Square pyramidal
4* Tetrahedral 6* Octahedral
CuII(d9) 4* Square planar CoIV(d6) 6 Octahedral
4 Distorted tetrahedral FeII(d6) 4 Tetrahedral
5 Square pyramidal 5 ?
5 Trigonal bipyramidal 6* Octahedral
6* Distorted octahedral FeIII(d5) 4 Tetrahedral
Cu (d )
III 8 6 Octahedral 6* Octahedral
NiII(d8) 4* Square planar 7 Pentagonal
4* Tetrahedral Bipyramidal
5 Trigonal bipyramidal FeIV(d4) 6 Octahedral
6* Octahedral FeV(d3) 4 Tetrahedral
NiIII(d7) 5 Trigonal bipyramidal MnI(d5) 6 Octahedral
NiIV(d6) 6 Octahedral MnII(d5) 4 Tetrahedral
CoI(d8) 4 Square planar 4 Square planar
5 Trigonal bipyramidal 6* Octahedral
6 Octahedral MnIII(d4) 6* Octahedral
Co (d )
II 7 4* Tetrahedral MnIV(d3) 6 Octahedral
4 Square planar MnV(d2) 4 Tetrahedral
5 Trigonal bipyramidal MnVI(d1) 4 Tetrahedral
5 Square pyramidal MnVII(d6) 3 Planar
6* Octahedral 4 Tetrahedral
Source: Data from Huges, M, The Inorganic Chemistry of Biological Processes (London: Wiley)
25–26, 1972.
Note: * = the most common states.
chelate as illustrated in Table 2.1.28 This table demonstrates that the chelate ring can be
either symmetrical or asymmetrical based on the coordination number. The coordina-
tion sites have well-defined mathematical stereochemical arrangements in space.18
To illustrate, a zinc bisglycinate chelate was formed following the dissolution of
glycine and zinc in water. The ratio of glycine to zinc was two moles of glycine to
every mole of zinc. The resulting product was allowed to crystallize by controlled
evaporation of the water. The pure zinc bisglycinate crystals were subsequently ana-
lyzed by x-ray diffraction spectrometry. As seen in Figure 2.6, each glycine mol-
ecule was attached to the zinc by the carboxyl group (COO -) and the amino group
(NH2) with the ionic bond originating from the oxygen and the coordinate covalent
bond coming from the nitrogen. This structure has two heterocyclic rings, with the
zinc the closing member for each ring.29 Using x-ray crystallography, inductively
O
C
O
N C
C N
N
Zn C
O
C O O
O C Zn
C
O
O
C
N
FIGURE 2.6 Zinc bisglycinate chelate as determined by x-ray diffraction spectrometry.

(Redrawn from Dalley, NK, “Report on x-ray diffraction crystallography of Albion® zinc
amino acid chelate,” unpublished report, Brigham Young University, Provo, UT, 1985.)
coupled plasma spectroscopy, electrospray mass spectroscopy, and combustion ele-

mental analysis, Konar et al. also confirmed this same chelate structure in copper
and zinc bisglycinates.30
Not only are the metal and its characteristics a consideration in forming a che-
late, so also are the characteristics of the ligand. An early proponent of nutritional
chelates, John Miller, grouped ligands into two categories: natural and synthetic.31
He stated that synthetic ligands include molecules such as ethylenediaminetetra
acetic acid (EDTA) or synthesized salicylic acid. Natural ligands, on the other hand,
include carbohydrates, lipids, proteins and their derivatives, some vitamins, and cer-
tain organic acids (amino acids, lactic acid, citric acid, etc.), to name a few.
To illustrate, Table 2.2 lists several ligands. Their stability constants can change,
even when they chelate the same metal.32 When the metal changes, the stability con-
stants also change.28
Each functional group within a chelating ligand (the moieties in the ligand that
actually bond to the metal) must be situated within the molecule in such a way that
it permits the formation of at least one heterocyclic ring with the metal being the
closing member of that ring.17 To form a ring structure, the ligand has to bend, twist,
or both. Thus, while in theory a peptide or even a protein can form a chelate with a
metal ion, for the reasons illustrated in Figure 2.5 the likelihood is extremely small.23
The flat, rigid nature of the peptide bond does not allow much flexibility for the for-
mation of chelates with terminal ends of short peptides.
The size of the ring resulting from creating a chelate will also affect its stability.18
Five or six member rings are the most stable chelate molecules.33 In Figure 2.5, there
are 38 members in the ring. The stability of this chelate, if it existed, would be very
low. A four-member ring results in bonding angles that are too acute to form stable
compounds. These sharp angles encourage breaking of the chelate.17 Larger chelate
rings composed of seven or eight members have been studied, but these chelates
are not generally very stable.17
The Chemistry of Chelation
TABLE 2.2
Logarithms of Formation Constants in Aqueous Solution at 25°C
Ionic
Ligand Strength Constant Mg2+ Ca2+ Mn2+ Fe2+ Fe3+ Co2+ Ni2+ Cu+ Cu2+ Zn2+
Glycine K1 3.44 1.38 3.44 4.30 5.23 5.77 8.62 5.52
K2 3.01 3.50 4.02 4.80 6.97 4.44
Cysteine 0.01 K1 <4 4.10 9.30 19.20 9.86
0.01 K2 7.60 8.84
β2 11.77 32.10 19.30
Serine Varied K1 ~0.50
Varied β2 7.0 8.0 14.54
Arginine 0.15 K1 3.87
0.15 K2 3.20
0.15 K3 2.08
0.01 β2 13.90 7.80
Aspartic acid 0.10 K1 2.43 1.60 3.74 5.90 7.12 8.57 5.84
0.10 K2 4.28 5.27 6.78 4.31
Histidine 0.01 β2 7.74 9.30 13.86 15.90 18.33 12.88
Lysine 0.01 K1 2.0 4.50
0.01 β2 6.80 8.80 13.70 7.60
Glutamic acid Varied K1 1.90 2.05 3.30 4.60 5.06 5.90 7.85 5.45
K2 3.40 4.44 6.55 4.01
Glutamine 0.16 K1 0.18
Asparagine 0.01 β2 ~4.0 ~4.50 6.50 8.13 10.60 14.90 8.70
Continued
29
30
Amino Acid Chelation in Human and Animal Nutrition
TABLE 2.2 (continued)
Logarithms of Formation Constants in Aqueous Solution at 25°C
Ionic
Ligand Strength Constant Mg2+ Ca2+ Mn2+ Fe2+ Fe3+ Co2+ Ni2+ Cu+ Cu2+ Zn2+
Glycylglycine K1 1.06 1.24 2.15 3.49 6.04 3.80
K2 2.39 5.62 2.77
Glycylserine 0.01 K1 3.7
Alanylglycine 0.01 K1 0.66 3.0
Glycylalanine 0.01 K1 3.15 4.1
0.01 K2 2.58
Glycylglycylglycine K1 1.41 2.98 5.41 3.33
K2 2.61 5.15 2.99
Imidazole Varied K1 0.08 2.94 4.20 2.58
K2 2.41 3.47 2.19
K3 1.99 2.84 2.41
β2 10.87
Proline 0.03 β2 <4 5.5 8.3 9.3 16.8 10.2
1.10 Phenanthroline Varied K1 6.30
The Chemistry of Chelation

K2 6.15
K3 5.50
β3 7.35 21.0 14.10
2-Methyl .1.10 phenanthroline β3 10.8
Ammonia K1 0.23 –0.2 2.11 2.79 5.93 4.15 2.37
K2 –0.15 –0.6 1.63 2.24 4.93 3.50 2.44
K3 –0.42 –0.8 1.05 1.73 2.89 2.50
Ethylenediamine 1.0m K1 –0.37 2.73 4.28 5.93 7.66 10.72 5.92
K2 2.06 3.25 4.73 6.40 9.31 5.15
K3 0.88 1.99 3.30 4.55 1.86
1.3 Diaminopropane 0.15 K1 6.98 9.77
K2 4.93 7.17
Ethanolamine β2 6.68
2-Mercapto ethylamine 1.0 K1 9.38 10.96 8.07
Source: Data taken from Huges, M., The Inorganic Chemistry of Biological Processes (London: Wiley) 71–72, 1972.
31
The Lewis base strength of the ligand will also affect the stability of the chelate.
In general, the greater the basicity of the donor ligand, the higher is its tendency
to donate electrons and form stable complexes.18 As noted, bidentate ligands can
be classified as having two basic groups, one acidic and one basic group, or two
acidic groups.24 The ligands with two basic groups, such as two nitrogen donors, will
coordinate as a neutral molecule. The resulting chelate will have the same charge
as was originally on the metal ion. If the ligand contains both an acidic and a basic
group, the coordination of the metal is usually due to the loss of an ionizable proton.
The charge on the metal ion is thus reduced by one unit for each ligand chelated
to the metal as an electron is filled in the s orbital. Amino acids are one class of
bidentate ligands that have both acidic and basic groups, but each group acts as
Lewis bases toward the metal. When the ligand contains two acidic groups, each
ligand anion reduces the positive charge on the metal ion by one unit when the elec-
trons are donated by the ligand to the s orbit of the metal ion. In this case, unless the
charges are balanced, anionic complexes can be formed due to excess ligand. Many
inorganic acids, such as oxalic acid, also fall into this category of ligands.24
In addition to requiring that the ligand possess two functional groups (1) that can
donate electrons, (2) that are located in such a way that they form a ring structure
with the metal ion, and (3) that the potential reaction between the metal cation and
ligand be sterically possible, chelation also requires that the potential reaction be
“energetically possible.”21
In summary, in 1963, Kugelmass wrote that chelation
involves the formation of an organic heterocyclic ring containing a polyvalent metallic
ion. The metal is attached by coordinate links to nonmetal atoms in the same molecule
as a stable, water soluble, non-ionic metallic ring. … Natural chelates bridge the gap
between the inorganic and organic worlds propagating and maintaining life comfort-
ably via chlorophyll, hemoglobin, glutathione, enzymes, hormones and vitamins to
such an extent that the chemistry of life may yet be resolved from the chemistry of
inorganic metal ions uniquely amplified by organic cofactors.34
REFERENCES
1. Werner, A, “Beitrage zur konstitution anorganischer verbindungen,” Z Anorg Allgem
Chem 3:267–330, 1893. (Translated by D.P. Mellor.)
2. Werner, A, and Miolati, A, “Beitrage zur konstitution anorganischer verbindungen,
II abhandlung,” Z Physic Chem (Leipzig), 14:506–521, 1894.
3. Werner, A, and Vilmos, Z “Beitrage zur konstitution anorganischer verbindungen-gen.
Über Oxalatodiathylendiaminkobaltisalze,” Z Anorg Chem 21:145–158, 1899.
4. Werner, A, “Über acetylacetonverbindungen des platins,” Ber deut Chem Ges
34:2584–2593, 1901.
5. Werner, A, “Zer kenntnis des asymmetrischen kobaltatoms V,” Ber deut Chem Ges,
45:121–130, 1912.
6. Werner, A, “Über spiegelbild-isomerie bei chromverbindungen III,” Ber deut Chem Ges,
45:3065, 1912.
7. Werner, A, “Zur kenntris des asummetrischen kobaltatoms XII. Über optische aktivitat
bei kohlenstoffreien verbindugen,” Ber deut Chem Ges, 47:3087–3094, 1914.
8. Werner, A, Neuere Anschauungen auf dem Gebiete der anorganischen Chemic
(Brunswick, Germany: F. Vieweg d John) 1920.
9. Nobelstiftelsen, A, Nobel Lectures Chemistry 1901–1920 (Hackensack, NJ: World

Scientific) 1999.
10. Morgan, GT, and Drew, HDK, “Researches on residual affinity and coordination. II
Acetylacetones of selenium and tellurium,” J Chem Soc 117:1456–1465, 1920.
11. Krebs, S, ed., Official Publication of the Association of American Feed Control Officials
(Oxford, IN: Association of American Feed Control Officials Incorporated) 311, 361, 2008.
12. Ley, H, “Über Innere Metall-Koomplexsalze. I,” Z Elektrochem angew Physik Chem
10:954–956, 1904.
13. Cotton, FA, Wilkinson, G, and Gaus, P, Basic Inorganic Chemistry (New York: Wiley)
503–516, 1995.
14. Hartle, JW, “Comparison of zinc bisglycinate and zinc oxide in vitro,” M.S. thesis,
Lehigh University, Bethlehem, PA, 2009.
15. Brown, T, LeMay, H, and Brusten, B, Chemistry: The Central Science (Englewood
Cliffs, NJ: Prentice Hall) 887–891, 1991.
16. Bruni, G, and Fornara, J, Rend Atti R (Rome: Acad dei Lincei Roma) 13:26, 1904.
17. Mellor, DP, “Historical background and fundamental concepts,” in Dwyer, FP, and
Mellor, DP, eds., Chelating Agents and Metal Chelates (New York: Academic Press)
1–48, 1964.
18. Kratzer, FH, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 1986.
19. Conrad, R, and Nakamoto, K, “Infrared spectra and normal coordinate analysis of metal
glycine complexes,” J Chem Phys 42:2590–2598, 1965.
20. Lower, S, “Bonding in coordination complexes: Introduction to crystal field theory,”
Chemistry I Virtual Textbook, (Simon Fraser University: http//www.chem1.com/acad/
chembound/cb09.html) 2007.
21. Ashmead, SD, “The chemistry of ferrous bis-glycinate,” Arch Latino Am Nutr Suppl
51:7–11, 2001.
22. Cotton, FA, Wilkinson, G, Murillo, CA, and Bochmann, M, Advances in Inorganic
Chemistry (New York: Wiley) 1999.
23. Jeppsen, RB, “Proteinates vs. amino acid chelates,” paper presented at International
MAAC® Nutritional Conference, Salt Lake City, UT, February 19, 1994.
24. Bell, CF, Metal Chelation Principles and Applications (Oxford, UK: Clarendon Press) 11,
1977.
25. Ibid., 27–35.
26. Huges, M, The Inorganic Chemistry of Biological Processes (London: Wiley) 25–26,
1972.
27. Irving, H, and Williams, RJP, “Order of stability of metal complexes,” Nature
162:746–747, 1948.
28. Huges, M, The Inorganic Chemistry of Biological Processes (London: Wiley) 71–72,
1972.
29. Dalley, NK, “Report on X-ray diffraction crystallography of Albion® zinc amino acid
chelate,” unpublished report, Brigham Young University, Provo, UT, 1985.
30. Konar, S, Gagnon, K, Clearfield, A, Thompson, C, Hartle, J, Ericson, C, and Nelson, C,
“Structural determination and characterization of copper and zinc bis-glycinates with
x-ray crystallography and mass spectrometry,” J Coord Chem 63:3335–3347, 2010.
31. Miller, JJ, “Principles of chelation,” paper presented at International College of Applied
Nutrition, Pasadena, CA, March 23, 1962.
32. Eichorn, GL, “Introduction to chelation chemistry” in Soffer, A, ed. Chelation Therapy
(Springfield, IL: Thomas) 133–143, 1964.
33. Frost, AE, “Fundamental aspects of chelation,” The Sci Counselor (Pittsburgh, PA:
Duquesne University Press) June 1956.
34. Furst, A, Chemistry of Chelation in Cancer (Springfield, IL: Thomas) VIII, 1963.
3 The History of
Nutritional Chelates
Although chelation was discovered in 1893 as a chemical phenomenon,1 its employ-
ment in nutrition to protect minerals from entering into gastrointestinal reactions
that would reduce potential mineral absorption did not occur until the 1950s.
Early nutrition investigators tried to chelate metal ions with synthetic ligands such
as ethylenediaminetetraacetic acid (EDTA). These EDTA chelates were fed to sev-
eral species of animals with varying results. One of the pioneering companies that
marketed synthetic chelates to the animal nutrition industry was Erly-Fat Livestock
Company, located in Tucson, Arizona. In 1957, the company was awarded a patent
by the U.S. Patent Office; the patent claimed that when a chelating agent (EDTA)
was mixed with a metal salt of iron, copper, cobalt, manganese, molybdenum, zinc,
or iodine, the digestibility of ruminant feedstuffs was improved because the growth
rates and activity of ruminant bacteria were increased. It was claimed that the che-
lated minerals had enhanced solubility of the minerals, which made these essential
minerals more available to ruminal bacteria.2
Following the granting of Erly-Fat’s patent and the election of former University
of Arizona professor, Bart Cardon, as president of Erly-Fat, the company expanded
its sale of synthetic chelates not only to the cattle industry but also to the equine,
swine, and poultry industries.3 Dr. Cardon’s work led the basic manufacturer of
the EDTA ligand to petition the U.S. Food and Drug Administration to allow diso-
dium EDTA to become an approved feed additive.4
In 1963, Cardon wrote, “Chelation is a term just now coming into general use and
understanding among nutritionists.”5 He claimed that for a mineral to be absorbed
from the intestine, it must remain soluble, but that solubility was compromised due
to chemical reactions with precipitating ligands in the gastrointestinal tract. “EDTA
is one chelating agent which has been used to increase mineral solubility in the
intestinal tract. … Thus, there is a possibility for using chelating agents in the diet to
improve mineral availability.”5
Cardon was not the only investigator working with chelates. Kratzer et al. reported
that the addition of EDTA to soybean protein enhanced zinc bioavailability to turkey
poults.6 Lease et al. also reported improved zinc absorption in chicks when EDTA
was added to the ration.7
Vohra and Kratzer subsequently reported that EDTA had a very high potential
stability constant that allowed it to successfully compete with other diet-sourced
ligands in the intestinal tract, such as phytic acid, for zinc and other trace elements.
They believed the EDTA was absorbed from the intestinal tract into the tissue and
thus concluded that EDTA acts as a carrier of polyvalent transition group trace
35
elements into the body. Once absorbed, they reported that the chick was able to
metabolize the EDTA and thereby gain the needed mineral from the chelate.8
Their argument was strengthened by Rubin and Princiotto, who reported that
young red blood cells will not take up inorganic iron salts from the surrounding
medium nearly as efficiently as they will when iron EDTA chelate is presented
to them. They concluded that EDTA was involved in passing essential metal ions
through cell walls.9
In 1969, Dyer wrote:
There are many natural and synthetic chelating substances. … One of the more promi-
nent ones is ethylenediaminetetra acetic acid [sic] which sequesters divalent cations.
… The science of chelation as it relates to nutrition of domestic animals is new. This
phenomenon, much more than techniques heretofore used, offers possibilities of regu-
lating the amount of a given metal at the cellular level.10
But alarms were sounding. Scott, who had once enthusiastically endorsed EDTA
as a way of improving zinc bioavailability,11 now warned:
We have seen that many chelating agents perform useful and vital functions in the
animal body. Others may cause drastic interference with metabolism. … It remains,
however, for us to learn which chelates are useful and beneficial and which are harmful
and toxic. Until this has been worked out very thoroughly, attempts to make practical
use of chelates in animal nutrition may produce unexpected results.12
Miller tested the theoretical work of Vohra and Kratzer8 in poultry and demon-
strated that a chelate with a high stability constant, such as that formed by EDTA,
does indeed protect the metal from chemical reactions in the gastrointestinal tract,
but it does not optimize mineral metabolism following absorption. The bond between
the metal and the EDTA ligand was too strong to be degraded by the body, and the
entire chelate was frequently excreted into the urine and lower bowel still intact
as the original molecule.13 In fact, earlier metabolic studies had demonstrated that
88.32% of orally ingested EDTA was excreted in the feces within 24 hours follow-
ing ingestion. In the same 24-hour period, 10.32% of the dose was eliminated in the
urine.14 Miller concluded, “It was hoped that evidence could be found that chelation
of trace minerals for poultry would enhance performance. No evidence to support
this premise has been uncovered which supports such an application to the type of
diets presently in use.”13
By the middle 1960s, nutritionists were starting to turn away from synthetic che-
lates in favor of natural amino acid ligands. Scott wrote, “Simple amino acids …
may have important chelating functions in the body.”15 Cardon added, “Glycine is
another chemical [amino acid] which can and has been used to enhance trace min-
eral solubility. It readily forms chelated complexes with a variety of trace minerals.
Of interest also is the ability of lysine, an essential amino acid, to chelate certain of
the minerals.”5 He was referring to the work of Wasserman et al.,16 which was pub-
lished in 1957 but went unrecognized for over a decade.
In the early 1960s, Harvey Ashmead, who had been working with synthetic che-
lates for a drug application, turned his attention to the nutritional possibilities of this
form of complexed minerals. As a result of his previous experience, during which
The History of Nutritional Chelates 37
time he and coworkers had developed a drug that resulted in minerals in the urine
being sequestered and removed following ingestion of synthetic chelates, including
EDTA,17,18 he rejected the synthetic chelates for nutritional purposes. Like Scott,12
Ashmead reasoned that chelated mineral bioavailability was, to a large extent, a
function of the stability constant of the chelate. The amino acids formed chelates
with stability constants that were similar to the stability constants of chelates formed
naturally in the body.19
In 1963, Ashmead contracted with a laboratory to study the effects of an amino
acid chelate on gestating rats. Amino acid chelates were made by hydrolyzing soy
flour and chelating each mineral separately with the resultant amino acids. The sepa-
rate chelated minerals were then blended together to contain 8% magnesium, 0.8%
iron, 0.8% zinc, 0.23% copper, 0.15% manganese, and 0.02% cobalt for a total of
10% metal. This chelate mixture was included in the rat chow at the rate of 0.1% of
the total feed and fed to gestating rats. The controls received the same feed but with-
out chelates. Infant mortality was less in the pups born to the dams fed the chelate. In
the control group, 50% of the pups were cannibalized by their mothers. The mothers
receiving the chelates did not kill any of their pups. The mean growth rate of the
babies was 7% greater in the chelate group compared to the surviving controls. The
mothers receiving the chelates also produced approximately 22% more milk than
the control mothers, leading the researcher to conclude that the chelated minerals
definitely contributed to greater lactation and faster growth rates.20
These results motivated Ashmead to expand his studies to include other animals.
Initially, most of his work focused on animal performance. In one of the early stud-
ies, 90 Holstein dairy cows that had recently freshened were selected from a herd of
300 animals. They were subsequently divided into two groups and matched for age,
lactation number, and production. Both groups received the same feed except the diet
of the treated group was supplemented with amino acid chelates (Table 3.1) mixed in
a grain concentrate at a 0.5% level and fed at the rate of 10 pounds (4.54 kg of grain
concentrate) per day containing 227 mg of mineral supplement. The control animals
received the same grain concentration but without supplemental minerals.
Both milk production and butterfat production (Table 3.2) were higher in the cows
assigned to the experimental group. The data pointed to the benefit of feeding the
amino acid chelates to maintain better production, particularly as lactation began its
normal decline during the lactation period.
TABLE 3.1
Dairy Cow Chelate Formula
Chelate Actual Metal Percentage Metal
Magnesium 34.08 lb 309.49 g 6.80
Iron 9.33 lb 84.72 g 1.86
Zinc 6.32 lb 57.41 g 1.86
Copper 0.89 lb 2.27 g 0.05
Cobalt 001 lb 0.11 g 0.0024
Total 50.00 lb 454.00 g 10.5224
TABLE 3.2
Butterfat Production in Cows Receiving Amino Acid
Chelates Compared to Control Cows
Average Experimental Group Controlled Group
Butterfat Production (45 cows) (45 cows)
December 63.0 lb 61.0 lb
January 64.0 lb 62.0 lb
February 54.0 lb 53.0 lb
March 54.0 lb 55.0 lb
April 50.0 lb 50.0 lb
May 52.5 lb 49.5 lb
The number of services per conception was also reduced. At estrus, the mean
services per conception (artificial insemination) were 1.59 for the experimental ani-
mals compared to 2.02 services in the control group. The herd manager stated that
the improvements he observed in the experimental group were more dramatic than
anything he had ever seen.21 Thus, Ashmead was encouraged to further his investiga-
tions with other animal species.
In another study, a formulation of metal amino acid chelates (Ca 45.66%,
Mg 46.66%, Zn 4.56%, Fe 2.28%, Mn 1.14%, Cu 0.45%, and Co 0.23%) was included
in the feed of laying hens at the rate of 0.1% of the total feed ration. Previous reports
had suggested that egg breakage was reduced when the mineral formula, in a non-
chelated form, had been fed to layers.22
The amino acid chelate formula was fed for 60 days to 5,000 birds in their ninth
month of lay and the resulting collected egg data compared to 5,000 control birds
from the same flock. First and foremost, the experimental group laid 18,000 more eggs
than did the control group or about 0.6 more eggs per bird per day. It required 0.77 kg
(1.7 pounds) more pressure to break the eggshell of the experimental group compared
to the control group. Both calcium and magnesium content of the egg shell were sig-
nificantly greater (p < 0.05) than in the control shells. The yokes from the experimental
birds contained 11.4% more zinc, 10.6% more iron, and 6.0% more copper than the
controls.23 The data from this study strongly indicated greater mineral absorption and
subsequent metabolism occurred when the amino acid chelates were ingested.
In a swine study, 58 gestating sows were divided into three groups, which were
labeled control pigs, iron dextran group, and iron amino acid chelate group. At
21 days before expected farrowing, 3,500 ppm of iron amino acid chelate (350 ppm
Fe) were added to the daily feed ration of the amino acid chelate group. Other than
this, all three groups received the same feed and amounts each day. The supple-
mental iron was discontinued at farrowing. Following parturition, the iron dextran
group of piglets was injected with iron dextran at 3 days and 14 days postfarrowing
to prevent baby pig anemia, which generally occurs in pigs raised on concrete floors.
Neither the control group nor the amino acid chelate group received the iron dextran
injections. All of the pigs from all groups were weaned at 28 days. Table 3.3 provides
the results for the three groups of piglets.24
TABLE 3.3
Effects of Iron Amino Acid Chelate on Piglet Performance
Iron Amino Acid
Control Group Iron Dextran Group Chelate Group
Number of pigs farrowed 144 285 168
Average birth weight 2.86 lba 2.82 lba 2.99 lbb
Number of pigs weaned 115 233 157
Pigs weaned (%) 79.9a 81.7a 93.5b
Mortality (%) 20.1a 18.3a 6.5b
Source: Summarized from Ashmead, D, Beck, B, and Hopson, H, “A new prophylactic

approach to reduction of piglet mortality,” Modern Vet Practice 58:509515, 1977.
a,b Difference in superscripts represents difference between treatments for the variable
(p < 0.05).
The original purpose of the study was to demonstrate that the iron amino acid
chelate was absorbed by the pregnant sow and transferred to her fetus in sufficient
quantities to prevent the naturally occurring baby pig anemia. If an inadequate sup-
ply of iron is transferred across the placenta from the maternal blood to the fetus, it
will impair fetal growth and result in increased morbidity and infant mortality fol-
lowing farrowing. Because of the difficulty of obtaining sufficient placental transfer
of dietary iron from the mother, baby pigs are generally born anemic and tradition-
ally require iron dextran therapy. Figure 3.1 shows the mean hemoglobin levels for
the three groups of piglets during the first 21 days of life. As the baby pigs grew,
the demand for iron increased proportionally. Neither the control group nor the iron
Iron dextran
Iron amino
acid chelate
Hemoglobin Level
Danger zone
No treatment
Blood hemoglobin
Birth 1 Week 2 Week 3 Week
FIGURE 3.1 Effect of different iron therapies on piglet hemoglobin levels. (Redrawn from
Ashmead, D, Beck, B, and Hopson, H, “A new prophylactic approach to reduction of piglet
mortality,” Modern Vet Practice 58:509–515, 1977.)
dextran group had sufficient tissue iron reserves to meet the demand of the rapidly
growing piglets. Iron deficiency anemia resulted in both groups. The supplemental
iron dextran injections overcame the lack of iron reserves in the iron dextran group;
thus, baby pig anemia was averted. In the amino acid chelate group, there were suf-
ficient iron reserves created during fetal life to meet the postnatal demands of those
growing pigs without resorting to iron dextran injections.
This study indirectly demonstrated that significant placental transfer of iron
occurred in the sows fed iron amino acid chelate. Hemoglobin analysis from the
blood of the baby pigs taken at 0, 7, 14, and 21 days postfarrowing confirmed this.
In terms of mortality, there was no significant difference between the groups of pigs
receiving no iron therapy and those pigs injected with iron dextran. When the moth-
ers consumed iron amino acid chelate during gestation, their offspring had signifi-
cantly lower mortality (p < 0.05). This led the researchers to examine the role of iron
in reducing infant mortality. They ultimately concluded that the lower mortality was
a function of birth weight. Pigs that are even slightly heavier at farrowing tend to
exhibit lower infant mortality.25 Thus, the birth weight differences could be traced
to differences in fetal nutrition, and in this case the placental transfer of iron from
the amino acid chelate fed the gestating sows was the source of that fetal nutrition.
Investigators at Iowa State University have reported that a heavier pig at birth results
in a heavier pig at weaning.26 Widdowson noticed that the rate of body cell division in
animals that are born small was lower than in larger animals. Furthermore, there was
a reduction in the mature size of the cells composing the organs of the smaller ani-
mals compared to the larger animals. She concluded that the intrauterine environment,
including nutrition, set the stage for postnatal cellular division and mature cellular size.27
This led Ashmead et al. to initiate studies examining the role of manipulating fetal min-
eral nutrition with amino acid chelates in an attempt to influence postnatal growth rates.
In another swine study, gestating sows were fed a ration containing iron amino
acid chelate (0.25% of the ration) beginning 30 days prior to expected farrowing. At
10 days before farrowing, the amount of chelate in the ration was increased to 0.30%
of the total ration. This was continued for 14 days following farrowing and then
stopped. The growth rates of the baby pigs from these sows were compared to herd-
mates who did not receive supplemental iron amino acid chelate. Table 3.4 summa-
rizes these results.28,29 These findings tended to confirm Widdowson’s report.27 The
small difference in birth weight translated into a significant difference 170 days later.
Ashmead and his group were not the only people publishing data relating to amino
acid chelates. John Miller, a former editor of Chemical Abstracts, had organized his
own company and began to publish journal articles and monographs on the use of
chelated minerals for nutritional purposes. In a lecture given at the International
College of Applied Nutrition in 1962, Miller said
Absorption is no doubt a major problem in considering the nourishment of both human
and animal body. … When minerals are properly complexed or chelated before inges-
tion, the positive charges are absorbed or neutralized by the organic chelating sub-
stance resulting in a slightly negatively charged molecule, which in unaltered form can
then be transported through the intestinal wall into the blood stream.30
Miller expanded his theoretical views in subsequent publications.31–33

TABLE 3.4
Effect of Heavier Birth Weights from Iron Amino Acid
Chelates on Weights of Growing Pigs
Sow Treatment
Amino Acid- From Amino Acid-
Weight Chelated Iron (lb) Control (lb) Chelated Iron (lb)
Birth 3.15 2.95 +0.20
28 days 15.00 14.40 +0.60
63 days 48.70 46.20 +2.50
170 days 220 208 +12.00
Source: Summarized from Svajgr, AJ, “Getting more iron into the nursing pig,”
Feedstuffs 48:34, March 8, 1976; and Wayne Feeds, Boost’n Iron: The
Different Chelated Iron, Allied Mills Monograph, Chicago, IL, 1976.
Miller’s editorial training led him to abstract the various publications that inter-
ested him. He would then merge those ideas into his own theories. Miller was not
a bench chemist. His views of chelation relied on the findings of others to make his
case. That was not to say that his views were incorrect, but they were simply based
on piecing together literature reviews instead of his own primary investigations.
For example, he quoted pertinent statements from the Hinze and Ashmead stud-
ies (described previously)22,23 in his Miller Pharmacal Company monthly bulletin,
“Miller News and Comments.”34 The abstracted quotes focused on greater calcium
uptake by laying hens and was followed by the announcement, “Our newest product
is chelated/complexed calcium—namely Ca-Plus.”
Ultimately, Miller was awarded two patents in which he described how to pre-
cipitate minerals from brines.35,36 He claimed that these patents taught how to make
organic mineral chelates that were the results of the precipitation.
Miller invited Ashmead to speak on the research being conducted by Ashmead
et. al. at Miller Pharmacal-sponsored events. In 1965, Miller entered into a contract
with Ashmead to have Ashmead’s company, Albion Laboratories, Incorporated, pro-
duce human-grade metal amino acid chelates for Miller Pharmacal.37
Miller also licensed another company, Key Minerals, Incorporated, to produce
animal-grade chelates using the Miller technology. The license did not include bio-
availability data. Thus, Ashmead was also asked to be the research arm for Key
Minerals. Ultimately, Key Minerals secured its own patent describing the manufac-
ture of chelated minerals.38
In 1968, Ashmead filed for a patent that focused on mineral bioavailability.39 His
patent described how chelating minerals with amino acids would raise the levels
of essential bivalent metals in the tissues of animals ingesting those chelates. In
preparation for filing the patent application, Ashmead felt he did not have sufficient
data. All of his early research had concentrated on animal performance as a result
of feeding amino acid chelates. Neither he nor anyone else had measured the actual
absorption of these minerals. However probable the concept might be, greater min-
eral absorption was still an unproven theory.
This lack of bioavailability data prompted a new series of research studies financed
and directed by Dr. Ashmead. The first of these studies was conducted in the late
1960s. It was an in vitro investigation comparing the absorption of several inorganic
trace mineral salts to amino acid chelates derived from various ligands. At this point,
all of the companies offering commercial “amino acid” chelates were hydrolyzing
protein more or less to the peptide/amino acid state as a source for amino acid ligands.
Until 1986, purified, single amino acid chelates were not commercially available.40
Using three protein sources—fish meal, soybean meal/corn meal (50:50), or
casein—amino acid chelates of copper, magnesium, iron, and zinc were produced.
The object of the study was to compare the absorption of these three chelated
sources to the same minerals in the form of carbonates, sulfates, and oxides. Adult
male Sprague-Dawley rats that were mineral sufficient were sacrificed on the day of
experimentation. Their intestines were removed and washed three times in Krebs-
Ringer solution that had been buffered to pH 7.4 with bicarbonate. The intestinal
segment from the pylorus to the distal end of the jejunum was removed, severed
longitudinally along the mesenteric line, and then cut into 2-cm segments. Following
randomization among several animals, they were preincubated in the Krebs-Ringer
bicarbonate buffered solution for 60 seconds at 37°C. A mixture of 95% O2 and 5%
CO2 was bubbled through the solution. Each amino acid-chelated mineral source
was presolubilized in a solution containing 5% thermolysine. The metal carbonate,
sulfate, and oxide sources were also solubilized. Several intestinal segments were
incubated in a beaker to which one of the metal-containing solutions was added.
The intestinal segments in the beaker were left exposed to the mineral solution for
120 seconds. During that exposure time, the segments continued to be bathed in
Krebs-Ringer buffered solution and the O2:CO2 gas bubbled through the solution.
Following exposure to the metal source, the intestinal segments were removed,
washed six times in the Krebs-Ringer buffer solution, and dried. Each intestinal
segment was then digested and analyzed for metal content by atomic absorption
spectrophotometry. Table 3.5 summarizes the results.22,41
For the first time, a controlled, clinical study proved that when minerals were
chelated to amino acids, the intestinal uptake of those minerals was greater than
absorption of the same minerals in an inorganic salt form. To the surprise of the
investigators, however, the study also showed that the different hydrolyzed protein
sources being employed as ligands resulted in different mineral absorption rates,
although generally, the mineral absorptions from the chelated sources were greater
than from the salt sources.
In trying to explain the reason for these differences, it was realized that different
amino acids form chelates with differing stability constants as well as different bond
strengths and play important roles in the availability of the chelated mineral.22,42
In simple terms, a stability constant describes the strength of the bonds between
the metal ion and ligand and is expressed in logarithmic terms.43 The three protein
sources used in the study had different amino acid compositions, which presumably
resulted in different overall stability constants. That realization led to radioactive
TABLE 3.5
Absorption of Minerals (ppm) from Different Sources in Rat Intestine In Vitro
Organic Organic
Fish Soybean Organic
Meal Meal Whey Inorganic Inorganic Inorganic
Mineral Control Chelate Chelate Chelate Carbonate Sulfate Oxide
Copper Trace 33 35 17 12 8 11
Magnesium 7 94 57 52 77 36 23
Iron 23 298 130 80 171 78 61
Zinc 14 191 191 126 87 84 66
Source: Summarized from Ashmead, H, “Tissue transportation of organic trace minerals,” J Appl Nutr
22:42–51, Spring 1970; and Graff, DJ, Ashmead, H, and Hartley C, “Absorption of minerals
compared with chelates made from various protein sources into rat jejunal slices in vitro,” paper
presented Utah Academy of Arts, Letters, and Science, Salt Lake City, April 1970.
isotope studies in which the absorption of amino acid chelates of the same metal, but
with different stability constants, was compared.
One such study involved three groups of individually housed adult Sprague-
Dawley rats. All animals were mineral sufficient. The rats were fasted for 24 hours
prior to administering a single dose of 32 mg of 54Mn to each animal. The rats in the
first group received the manganese as 54MnCl2 dissolved in 50 mL of distilled water.
The second group received the same amount of 54Mn as amino acid chelate. The
ligand consisted of 14% methionine, 55% aspartic acid, and 31% glycine with the sta-
bility constant adjusted to 107. This chelate was also dissolved in 50 mL of distilled
water. The third group of rats received the 54Mn chelate with the same ligand, but the
stability constant was adjusted to 1010. All of the rats were mildly sedated with ether,
and each animal received a single oral dose of 54Mn from one of the three sources
via a 250-mL automatic pipette.
Two weeks following dosing, all of the rats in all three groups were sacrificed.
During the 2-week period prior to sacrifice, the urine and feces of each animal had
been collected and assayed for 54Mn. At the time of sacrifice the whole body, blood,
and various tissues were assayed for 54Mn. Table 3.6 summarizes the results.44
It was concluded that, “The stability constant formation is a very critical factor in
the assimilation and metabolism of manganese amino acid chelate. How the chelate
is made is extremely important as it relates to maximizing its absorption.”44
Additional radioactive stability constant studies were conducted on several
other biologically essential minerals. All of them used a similar protocol as the
one described above. One such study looked at 45Ca absorption. The control group
received 1 mg 45Ca as CaCl2 dissolved in 40 mL of distilled water. The second
group also received 1 mg 45Ca but in the form of an amino acid chelate. The ligand
had a ratio of 40% aspartic acid, 40% glycine, and 20% methionine with a stability
constant of 107. The third group received the same amount of 45Ca chelated to the
same ligands but the stability constant was adjusted to 1010.
TABLE 3.6
54Mn Absorption and Metabolism in Rats Receiving Different
Manganese Sourcesa
Manganese54 Manganese54 Amino Manganese54 Amino
Chloride Acid Chelate (107) Acid Chelate (1010)
Whole body 3,124 10,600 4,540
200 mL whole blood 85 217 165
Blood serum 54 136 104
Blood plasma 31 81 61
Liver 52 106 126
Kidney 80 97 109
Spleen 190 397 93
Lung 54 56 35
Testes 63 62 59
Small intestine 89 141 79
Gastroc (muscle) 28 22 17
Frontal bone 112 94 266
Masseter 145 110 110
Urine 1,250 1,570 1,910
Feces 4,221,888 3,168,144 4,106,880
Source: Summarized from Ashmead, H, Jensen, N, and Graff, D, “Stability constants of

manganese 54 chelates and their effects on manganese metabolism,” unpublished
research report, Albion Laboratories, Clearfield, UT, 1974.
a Data are presented as corrected counts per minute per gram dry tissue.
At the end of 7 days, all of the animals were sacrificed. Their tissues and blood
(plasma and serum) were assayed for 45Ca. Table 3.7 summarizes the results.19
As with the manganese isotope study, when the metal was chelated to an amino
acid ligand, absorption and subsequent tissue deposition increased, but uptake by the
various tissues changed depending on the stability constant of the chelate. While no
explanations for these observations were offered at the time the studies were con-
ducted, subsequent research has provided those answers. That research is discussed
in the following chapters.
Giroux and Prakash also evaluated zinc availability in rats that ingested differ-
ent zinc chelates. They found that when zinc was chelated to phytic acid, EDTA, or
penicillamine, the availability of the zinc was significantly reduced. However, when
the zinc was chelated to lysine, cysteine, glycine, or histidine, zinc absorption was
much higher in comparison to the chelates with higher stability constants or even
zinc sulfate.45
Not all of the research conducted by Ashmead et al. used multiple amino acid
ligands. They also employed single amino acid ligands. For example, in one study
TABLE 3.7
45Ca Absorption and Metabolism Study in Ratsa
Control 2 Control 3
Stability Constant Stability Constant
Control 1 107 1010
Frontal bone 3,682 5,878 5,772
Masseter 602 844 904
Gastroc (muscle) 614 620 1,206
Heart 642 598 932
Liver 664 546 742
Right cerebrum 698 726 804
Kidney 686 656 730
Lung 676 672 648
Blood serum 8.4 39.6 31.0
Blood cells 18.6 0 13.2
Whole blood 27.0 39.6 44.2
Source: Summarized from Ashmead, H, Jensen, N, and Graff, D, “The influ-

ence of stability constants on calcium metabolism in vivo,” unpublished
a Data are presented as corrected counts per minute per gram dry tissue.
two groups of adult male Sprague-Dawley rats were used. One group received 59Fe
as ferrous sulfate, whereas the other group was administered 59Fe as iron amino
acid chelate with the ligand being methionine. Each sedated animal received a
single oral dose of 36.7 mg of iron dissolved in 40 mL of distilled water. Each
animal was sacrificed 72 hours postdosing and certain of their tissues and blood
assayed for 59Fe. Feces and urine were also collected during the period between
dosing and sacrifice and were assayed for 59Fe. Table 3.8 reports the results.46 The
amino acid-chelated iron was absorbed in greater quantities, but less iron was
subsequently excreted into the urine. The reason for this becomes clear in the fol-
lowing chapters.
The studies discussed are not all inclusive. They are simply a limited represen-
tation of the developmental work on amino acid chelate absorption primarily con-
ducted by or directed by Harvey Ashmead during the 1960s to 1970s. Ashmead
estimated that by 1970, he personally had experimented on over 200,000 animals.22
These included horses, cattle, sheep, and fish. After 1970, he could also add labora-
tory rats. At the time that these original bioavailability studies were initiated, little or
no work was being conducted on amino acid chelates by other investigators. (Their
contributions would come later.) Consequently, it was through the exclusive and
direct efforts of Dr. Harvey Ashmead that it was initially proven that when minerals
were chelated to amino acids, their absorption was increased.
TABLE 3.8
Comparison of 59Fe Absorption and Metabolism
from Iron Sulfate and Iron Methionate in the Rata
Iron Sulfate
59 59Iron Methionate
Body Part (cc/min/g) (cc/min/g)
Heart 63 151
Liver 136 243
Leg muscle 2 54
Jaw muscle 14 138
Brain 31 130
Kidney 2 327
Testes 20 109
Blood serum 700 1,797
Red blood cells 724 2,076
Whole blood 1,355 4,215
Feces 302,400 214,000
Urine 490 370
Source: Summarized from Ashmead, H, Ashmead, D, and Jensen,

N, “Chelation does not guarantee mineral metabolism,”
J Appl Nutr 26:5–12, Summer 1974.
a Data are presented as corrected counts per minute per gram of
body portion being analyzed (cc/min/g) average of rats.
REFERENCES
1. Werner, A, “Beitrag zur konstifution anaorganischer verbindungen,” Z Anorg U Allgem
Chem 3:267, 1893.
2. Cardon, BP, “Chelated metals in feedstuffs for ruminants,” U.S. Patent 2,960,406,
November 15, 1960.
3. Berglund, R, “Arizona feed firm gears products and services to individual livestock
feeders,” Feedstuffs, 23, December 14, 1963.
4. Federal Register, p7895, Friday, June 18, 1965.
5. Cardon, BP, “The importance of chelation in trace mineral nutrition,” paper presented at
Montana Nutritional Conference, Billings, February 11, 1963.
6. Kratzer, FH, Allred, JB, Davis, PN, Marshall, BJ, and Vohra, P, “The effect of autoclav-
ing soybean protein and the addition of EDTA on biological availability of dieting zinc
for turkey poults,” J Nutr 68:313–322, 1959.
7. Lease, JG, Barnett, BD, Lease, EJ, and Turk, DE, “The biological unavailability to the
chick of zinc in a sesame meal ration,” J Nutr 72:66–70, 1959.
8. Vohra, P, and Kratzer, V, “Influence of various chelating agents on the availability of
zinc,” J Nutr 82:249–256, 1964.
9. Rubin, M, and Princiotto, JV, “Chelation in nutrition, chelation as a basic biological
mechanism,” J Agric Food Chem 11:98–103, 1963.
10. Dyer, IA, “Mineral requirements,” in Hafez, ESE, and Dyer, IA, eds., Animal Growth
and Nutrition (Philadelphia: Lea & Febiger) 313, 1969.
11. Scott, ML, “Increasing utilization of zinc,” Western Feed and Seed, 123–125, January
1962.
12. Scott, ML, “Chelates in animal nutrition,” Proceeding of Maryland Nutrition Conference
54–60, 1965.
13. Miller, RF, “Chelating agents in poultry nutrition,” paper presented at Delmarva
Nutrition Short Course, Delmarva, MD, March 1968.
14. Foreman, N, Vier, M, and Magee, M, “Metabolism of C14 ethylenediaminetetraacetic
acid in the rat,” J Biol Chem 203:1045–1053, 1953.
15. Scott, ML, “Some practical aspects of chelates in animal nutrition,” Feedstuffs 37:31–32,
February 13, 1965.
16. Wasserman, RH, Comar, CL, Schooley, JC, and Lengemann, OR, “Interrelated effects
of L-lysine and other dietary factors on the gastrointestinal absorption of calcium 45 in
the rat and chick,” J Nutr 62:367–376, 1957.
17. Ashmead, H, “Urinary calculi treating compositions and methods of using same,” U.S.
Patent 3,184,381, May 18, 1965.
18. Phillips, LR, “Clinical studies on the control of urolithiasis in the feline,” Animal
Hospital 2:15–19, 1966.
19. Ashmead, H, Jensen, N, and Graff, D, “The influence of stability constants on calcium
metabolism in vivo,” unpublished research report, Albion Laboratories, Clearfield, UT,
1974.
20. MacGregor, D, personal communications, December 7, 1963.
21. Hinze, P, and Ashmead, H, “The use of Metalosates™ in bovine nutrition,” unpublished
Spring 1970.
23. Hinze, PM, “Metalosates: Their electromotive potential in laying hens,” Feed
Management 20:28–30, March 1969.
24. Ashmead, D, Beck, B, and Hopson H, “A new prophylactic approach to reduction of
piglet mortality,” Modern Vet Practice 58:509–515, 1977.
25. Speer, VC, “Iowa swine nutrition herd performance data,” Monograph, Iowa State
University Ames, Iowa State University, 1970.
26. Smith, AL, Stalder, KJ, Serenius, TV, Baas, TJ, and Mabry, JW, “Effects of piglet
birth weight on weights at weaning and 42 days post weaning,” J Swine Health Prod
15:213–218, 2007.
27. Widdowson, E, “Malnutrition during pregnancy and early neonatal life,” paper pre-
sented at Symposium on Fetal Malnutrition, New York, February, 1970.
28. Svajgr, AJ, “Getting more iron into the nursing pig,” Feedstuffs 48:34, March 8, 1976.
29. Wayne Feeds, Boost’n Iron: The Different Chelated Iron, Allied Mills Monograph,
Allied Mills, Chicago, IL, 1976.
30. Miller, JJ, “Principles of Chelation,” paper presented to International College of
Nutrition, Pasadena, CA, March 23, 1962.
31. Miller, JJ, “The story of chelation: Past, present and future,” J Appl Nutr 24: 40–45, Spring
1972.
32. Miller, JJ, Chelation and Its Role in the Life Processes of Plants, Animals and Man,
Miller Pharmacal Monograph, 1959.
33. Miller, JJ, Chelation, Miller Pharmacal Monograph, Miller Pharmacal Company,
Chicago, IL, 1960.
34. Miller, JJ, “Miller News and Comments,” Miller Pharmacal Newsletter, Miller
Pharmacal Company, Chicago, IL, June 10, 1969.
35. Agreement between Albion Laboratories, Inc., and Miller Pharmacal Company,
December 12, 1965.
36. Miller, JJ, “Precipitation of minerals from brines,” U.S. Patent 3,396,104, August 6,
1968.
37. Miller, JJ, “Process of treating saline water,” U.S. Patent 3,396,104, August 6, 1968.
38. Richards, AZ, “Precipitation of metal proteinates from brine by base-acid-base hydroly-
sis,” U.S. Patent 3,775,132, November 27, 1973.
39. Ashmead, HH, and Little, PA, “Increasing metals in biological tissues,” U.S. Patent
4,020,158, April 26, 1977.
40. Ashmead, HH, “Pure amino acid chelates,” U.S. Patent 4,599,152, July 8, 1986.
41. Graff, DJ, Ashmead, H, and Hartley C, “Absorption of minerals compared with chelates
made from various protein sources into rat jejunal slices in vitro,” paper presented Utah
Academy of Arts Letters and Science, April 1970.
42. Oyler, D, Stability Constants as Applied to Chelation in Biological Minerals Absorption,
Albion Laboratories Monograph, Albion Laboratories, Clearfield, UT, 1971.
43. Kragten, J, Atlas of Metal-Ligand Equilibria in Aqueous Solution (Chichester, UK:
Horwood) 1978.
44. Ashmead, H, Jensen, N, and Graff, D, “Stability constants of manganese 54 chelates
and their effects on manganese metabolism,” unpublished research report, Albion
Laboratories, Clearfield, UT, 1974.
45. Giroux, E, and Prakash, N, “Influence of zinc-ligand mixtures on serum zinc levels in
rats,” J Pharm Sci 66:391–395, 1976.
46. Ashmead, H, Ashmead, D, and Jensen, N, “Chelation does not guarantee mineral metab-
olism,” J App Nutr 26:5–12, Summer 1974.
4 The Requirements
for a Nutritionally
Functional Chelate
In Chapter 2, the four basic chemical criteria absolutely essential for chelate forma-
tion were discussed.1 These are, first, the bidentate ligand must possess at least two
functional groups, each of which is capable of donating electrons to the metal ion.
When this occurs, the metal ion acts as a Lewis acid at the time it accepts electrons
from the ligand, and through the process of donating electrons, the ligand acts as a
Lewis base. The resulting bond between the metal and ligand is a covalent bond with
an ionic character. The second bond between the ligand and metal can be covalent
if the metal ion shares an electron with the ligand and the ligand shares its electron
with the cation.2
The second chemical criterion to form a chelate is that the functional groups
within the ligand, that is, the moieties that actually bond to the metal ion, must be
located in such a way to allow a ring structure to be formed between the two func-
tional groups and the metal.1 The metal thus becomes the closing member of the
ring. If a ring cannot, or has not, been formed, then chelation has not occurred. By
definition, the ring must be heterocyclic, which means that the ring must include
atoms of different elements.3 The size of the heterocyclic ring will affect the stability
of the chelate. As the number of atoms in the ring increases above six members or
decreases below five members, the stability of the chelate tends to decrease.4,5
The number of heterocyclic rings formed with a single metal ion will also affect
the stability of the chelate.4 Generally, the more rings that can be formed, the greater
the stability of the chelate will be.4 As pointed out by Glusker, “When a metal ion
coordinates a ligand, it can affect the electron distribution of the ligand and therefore
its reactivity.”2 Thus, with more ligands surrounding the metal ion, there is potential
for lower reactivity of the chelate molecule.6
The third chemical criterion for chelation to occur requires that the potential reac-
tion between the ligand and metal ion be sterically possible.1 The positions of the
atoms in the ligand must be able to rearrange themselves in three-dimensional space
in such a way that they are able to bond to the metal ion.7 Thus, the spatial relation-
ship between those atoms becomes critical to chelation.
The following formula, in which the ligands are two moles of glycine and M2+ is
one mole of a divalent cation, illustrates that steric criterion for a bis-glycinate chelate:
49
O
H H O
O C
C O
O M
2H 2N CH2 C OH + M +2 C N
C
N H
H H H H
H

This formula demonstrates the chelation of a metal ion using the backbone common
to all amino acids. In a more complex amino acid, other reactive moieties on the
amino acid could sterically form a ring. If this were to occur, the chelate ring would
be different from the ring formed with the moieties shown.
The fourth basic chemical criterion for chelation to occur requires that the poten-
tial reaction between the metal ion and the ligand be energetically possible. If the
molar ratio of the ligand to metal is not satisfied, that is, if there are not enough
ligand molecules to balance the charge of the metal ion, then it creates the possibility
of allowing the metal ion to be bonded to some other ligand with a higher potential
stability constant. If this were to occur, this new ligand could displace the original
ligand or compete with the potential chelation by the original ligand.1 This concept
is more complex than the following example would suggest since temperature, con-
centration, pH, and more can all influence stability constants. Nevertheless, for the
sake of simplicity assume glycine has a potential stability constant with Fe2+ of log
K = 3.73 and a stability constant of log K = 9.2 with histidine in an aqueous solution
at 25°C.8,9 If the solution contained Fe2+, glycine, and histidine, the resulting chelate
molecules would probably be a combination of pure iron histidine chelate and if suffi-
cient iron remained, a pure iron glycine chelate. It would be unlikely that the reaction
would create an iron histidine-glycine chelate as the major reactant. The histidine,
having a significantly higher formation constant, would generally bind the Fe2+ first
in a 2:1 molar ratio. The glycine, if it had originally bonded to the iron first, would
probably be mostly displaced by the histidine until all of the histidine ligands had first
bonded to the iron cation. After the histidine was completely reacted with the iron,
the glycine would then be allowed to attach to the remaining unreacted iron ions.
Now, if two moles of ethylenediaminetetraacetic acid (EDTA) (with a stabil-
ity constant of log K = 14.30) were introduced into the solution, it would probably
displace all of the glycine in the iron glycinate product first because its potential
formation constant is more than double that of the glycine.9 It could also poten-
tially displace the histidine, but it is more likely it would first attack the weaker
glycine chelate since histidine and EDTA have closer potential formation constants
(although EDTA is slightly higher).9
Many uninformed manufacturers claim to create “chelates” by mixing the metal and
ligands by physical weight rather than by molar weights, resulting in more metal avail-
able for chelation than ligand to chelate with.10 This leaves the unreacted metal free
to bind with other ligands that lower or prevent optimum mineral absorption from the
gastrointestinal tract following ingestion of the mineral product.
The Requirements for a Nutritionally Functional Chelate 51
Even if a metal is chelated, that does not guarantee that it will be absorbed or
metabolized following ingestion.11 To illustrate, Bates et al. compared the absorp-
tion, tissue retention, and tissue distribution of iron from ferrous sulfate to chelated,
iron citrate, iron nitrilotriacetric (NTA), and iron EDTA. They reported that absorp-
tion was about the same for all of the sources of iron and concluded that, “Chelation
does not, in itself, insure efficient uptake.”12
Some of the iron chelates used in the Bates et al. study had relatively low stabil-
ity constants, which may have been part of the reason they did not observe efficient
uptake of the iron.11 Other chelates like the iron NTA or EDTA have high stabil-
ity constants.11 Vohra and Kratzer demonstrated that the availability of the mineral
from the chelate is dependent on having a stability constant that is higher than the
metal-bonding substances in the food (e.g., phytates, oxalates, etc.).13 If the stability
constant of the ingested chelate were lower than the potential formation constants
of the ligands found in the digested food, then the metal could be exchanged from
the original chelate to a food-sourced ligand instead. If that food ingredient were
subsequently absorbed, then, of course, the metal would be absorbed concurrently.
If, however, that food substance were not absorbed due to a number of reasons, then
the metal would probably also not be absorbed.
It has been suggested11 that the ideal stability constant for zinc availability is
between log K = 13 and log K = 14.5. Once such a chelate is absorbed from the lumen
into the intestinal tissue, another ligand with a potential stability constant of between
log K = 14.5 and log K = 16.5 must exchange the metal from the absorbed chelate and
distribute that metal throughout the body or the cellular environment must change so
that the stability constant of the absorbed chelate is reduced.
Some chelates form stability constants that are higher than the log K = 16.5
value.11 Picolinic acid is such an example.14 Some researchers originally proposed
that the body secreted picolinic acid into the intestine to facilitate zinc absorption,
particularly when the zinc was bound to phytic acid or oxalic acid from the diet.
The zinc picolinate has a higher stability constant than do either the phytates or the
oxalate chelates of zinc and therefore protects the metal ion while in the gastroin-
testinal tract.15–17 Such a chelate can be absorbed. Following its absorption, how-
ever, the once-beneficial protective stability constant can become a detriment to the
metabolism of the chelate. The absorbed picolinate will not release its metal to a
ligand having a lower potential formation constant. Consequently, the metal from
this highly stable chelate cannot generally be distributed and metabolized through-
out the body tissues after absorption.18
Seal reported that when rats were fed diets containing 25, 60, or 120 ppm of zinc
and 0, 20, 40, or 60 moles of picolinic acid per kilogram of diet, zinc excretion in the
urine was directly related to the quantity of picolinic acid in the diet.18 The greater
the concentration of picolinic acid in the diet, the greater the amount of zinc pico-
linate recovered in the urine.
Based on the studies of Ruben et al. it appears that certain chelates, such as
EDTA, are absorbed by diffusion rather than by active transport.19,20 To be absorbed
by active transport, one of the moieties of the ligand of the chelate would have to
release the metal and reattach that functional group to a carrier molecule in the
mucosa via the charge that it carries.21 If the stability constant of the EDTA chelate is
stronger than the surrounding ligands, including the transport molecules, then there
is no inducement for a functional group to release the metal and reattach to a carrier
molecule. Rubin et al. reported that even though the iron EDTA was absorbed by dif-
fusion, due to its high stability constant, this chelated iron later was recovered in the
urine unmetabolized.19,20 This led Kratzer and Vohra to conclude that even though
EDTA effectively protected the iron from competing dietary ligands that would
potentially interfere with mineral absorption from the gastrointestinal tract, and
even though the iron EDTA was absorbed intact, it was not a nutritionally functional
chelate and was not indicated as an effective treatment for iron deficiency anemia.22
There are several criteria that are essential to create a nutritionally functional
chelate. A nutritionally functional chelate is one in which both the ligand and the
metal are available to the body, and both can be utilized by the body following their
absorption. The fact that a mineral is chelated does not necessarily make it nutrition-
ally functional. The fact that a chelate is absorbed does not make it nutritionally
functional. It is only when the ligand and metal are both utilized by the body, after
absorption, that the chelate can be called totally nutritionally functional.
The stability constant is critical in creating a nutritionally functional chelate. It
must be higher than the potential formation constants of the ligands in the chyme of
the stomach and intestines.1 A chelate, even if it is nutritionally functional and has a
low stability constant, will readily give up its metal to a ligand in the chyme that has
a potentially higher formation constant, such as phytate.11
Chelates with stability constants of around 10 are generally formed by ligands in
the body.11 If, as was noted with picolinic acid or EDTA, a chelate is formed with a
stability constant that is higher than that of the potential formation constants of the
ligands found in the body, then the stronger chelate will not readily give its metal
up to a ligand that forms a weaker chelate. The result is that while absorption of the
chelate may occur, mineral utilization will not.
While the absorption and metabolism mechanisms for plants and animals, includ-
ing humans, are different, the following study illustrates the above concept: Three
59Fe solutions were made in the laboratory. The first used ferrous sulfate as the iron
source. The second was an iron EDTA solution. The final solution was an iron amino
acid chelate in which glycine was employed as the ligand. Bean plants, grown in a
greenhouse, were treated with 4 microcuries of one of the three 59Fe sources placed
on one leaf of each bean plant. Previous investigators had reported that if the foliar-
applied iron is absorbed and metabolized, about 25% of the applied dose would be
translocated to the apex leaves of the plant,23 leaving only a small portion of the iron
remaining within the treated leaf.24
Three days postdosing, the plants were harvested. Their leaves and stems were
thoroughly cleaned to remove external contamination and any nonabsorbed 59Fe.
These samples were cut into slices measuring about 2.0 mm in thickness before
being transferred to scintillation vials containing 1.0 mL of Soluene 350, an alkaline
tissue solublizer; they remained in the vials for 48 hours at 65°C under constant agi-
tation to facilitate digestion of the tissue. Each sample was subsequently assayed for
59Fe by ligand scintillation.
At the end of 3 days postdosing, the plants had absorbed 43.09 ccm (corrected
counts per minute per milligram of dry tissue) of the 59Fe from the amino acid che-
late, 26.23 ccm (corrected counts per milligram) of the 59Fe from the EDTA, and
29.24 ccm of the 59Fe from the ferrous sulfate into the plant tissue following the
single topical application. The difference in the absorbed 59Fe concentration between
the iron amino acid chelate and the other two iron sources was significant (p < 0.05).
While both the iron EDTA and ferrous sulfate were relatively immobile within the
treated leaf, the iron amino acid chelate was reasonably mobile, and 59Fe transloca-
tion by phloem and xylem had commenced. The stems of the dosed leaves had a
mean amount of 0.34 ccm of 59Fe from the amino acid chelate compared to 0.03 ccm
from EDTA (p < 0.05) and 0.12 ccm from ferrous sulfate (p < 0.05). The adjacent
leaves were also taking up the translocated iron from the amino acid chelate.25 The
iron amino acid chelate was absorbed and metabolized 63% better than EDTA (p <
0.05) and 47% better than ferrous sulfate (p < 0.05). The absorption and metabolism
of 59Fe from ferrous sulfate was 11% better than from the iron EDTA. Thus, even
though the iron was protected when it was chelated to the EDTA molecule, due to the
high stability constant of that chelate, this source of chelated iron was not absorbed
or metabolized as well as nonchelated iron or iron chelated with glycine. Chelation
does not guarantee absorption or metabolism of the mineral within the organism,
particularly if the stability constant of the chelate is too high or too low in relation-
ship to potential formation constants of the multitude of ligands within the organism.
Stability constants are influenced by many factors, some of which have been dis-
cussed previously.26 These include
1. the size of the chelate ring;

2. the number of rings bonded to the metal ion;
3. the Lewis basic strength of the chelating ligand;
4. the size and charge of the ligand;
5. whether the ligand is multidentate or bidentate;
6. modifications of the p bonding strength;
7. the nature of the donor ligand atom;
8. the resonance effect within the chelate ring; and
9. the steric effect of the ligands.
Even though chelating a mineral does not guarantee its absorption or metabolism,
the way a mineral is chelated and which ligand is employed will influence its ulti-
mate absorption or metabolism. When stability conditions favor the organism, the
chelate can be considered nutritionally functional.
Besides having the correct stability constant, a nutritionally functional chelate
must also have a low molecular weight.1 The purpose of the digestive process is to
render food into smaller absorbable molecules. Both mechanical digestion, such as
chewing or gut movement, and chemical digestion, including enzymatic digestion,
prepare the food constituents for absorption. Chemical digestion starts in the mouth
with the secretion of digestive enzymes into the saliva. In adult humans, over a liter
of saliva is produced daily by salivary glands located under the tongue. These glands
also secrete mucus to lubricate the esophagus and help the food be swallowed. The
saliva is alkaline, and that pH is the requisite environment for initiation of that par-
ticular enzymatic activity in the saliva.27
As the food enters the stomach, it temporarily remains in the upper portion of the
stomach. This allows the enzymes secreted into the saliva to continue their hydrolytic
activity. Ultimately, the mixture becomes sufficiently acidic due to mixing with acid
from the gastric juices. The lower pH arrests the enzymatic action of the salivary
enzymes. The gastric juice secreted by glands in the stomach contains not only hydro-
chloric acid but also other digestive enzymes that function only in an acidic pH.27
In the stomach, the bacteria that may have entered concurrently with the food
are generally destroyed by the acid. Some complex sugars are hydrolyzed. Certain
minerals, such as inorganic zinc and iron, have their solubilities increased due to the
lower acid pH. This leads to their increased absorption. The peptide bonds in the pro-
teins commence being broken by protease enzymes such as pepsin, an enzyme that
is active in an acidic environment. These protease enzymes begin degrading proteins
and peptides by cleaving the bonds after the N-terminal of the aromatic amino acids.
The process of breaking complex fats into simpler lipids also begins.27
Ultimately, the chyme leaves the stomach. Carbohydrates are first, followed by
protein and finally fats. What the minerals are bound to in the chyme will deter-
mine the order in which they exit the stomach or, to a degree, whether they can be
absorbed directly from the stomach.
As the chyme enters the duodenum of the small intestine, bicarbonate ions and
water, which are secreted in the pancreatic juice into the distal end of the duode-
num, commence to neutralize the acidic pH of the chyme. The pancreatic juices,
some intestinal juices, as well as bile are all secreted into the intestine from the
common bile duct. Bile contains no enzymes, but it emulsifies the dietary fats
for subsequent digestion by pancreatic lipase. The secretions of intestinal juices
into the small intestine coupled with secretion of pancreatic juices introduce new
enzymes into the digestive process that complete the digestion of carbohydrates,
fats, and proteins.27
Most basic nutrients—water, amino acids, lipids, carbohydrates, vitamins, and
minerals—are primarily absorbed from the small intestine. There is some absorp-
tion from the stomach and the large intestine, but as a percentage of the whole, that
absorption is low. The purpose of digestion is to break down the complex molecules
that compose the food into small molecules that can be absorbed across the intestinal
mucosa by diffusion or by active transport.
Diffusion occurs when the nutrients in the intestinal tract are in a higher concen-
tration than they are in the blood and lymph systems. They will then diffuse across
the intestinal membrane from the lumen into the absorption cells and ultimately into
the blood and lymph system.
Conversely, if the concentration of a given nutrient is lower in the intestine than it
is in the lymph or blood systems, then its movement, if any, from that lower concen-
tration to the higher concentration across the intestinal membrane requires an active
transport system. An active transport system requires a transport molecule to carry
the nutrient across the membrane and energy to drive that process.
The digestive system is relatively efficient. It is estimated that several hundred

grams of carbohydrates, about 100 g fat, 50 to 100 g amino acids, 5 to 10 g mineral,
and 8 to 9 L water are absorbed from the small intestine each day.28 With the excep-
tion of water, the movement of these nutrients across the mucosal membrane of the
small intestine occurs only if the food containing these nutrients has been hydrolyzed
into small molecules as a result of the digestive process. If nutrients are not digested
into their basic units, their molecular sizes remain too large to be absorbed, and they
are transferred to the large intestine, where they become part of the fecal material.
Both Hofner29 and Tiffin30 have reported that if mineral complexes, including
chelates, are larger than 1,500 daltons, they are not able to cross cell membranes as
intact molecules. Kratzer and Vohra have added that for a ligand from a chelate to
promote metal absorption, the chelate, including the metal, must have a total molecu-
lar weight that is less than 1,000 daltons. Higher-weight metalloproteins, such as fer-
ritin (445,000 daltons) or ceruloplasmin (160,000 daltons), will facilitate storage of
the absorbed metal, but these proteinaceous ligands are too large for intact passage
across cell membranes. The transfer of a mineral from cell to cell requires that these
storage ligands (like ferritin) release the metal to smaller molecular weight ligands
that are capable of carrying the metal across the cell membrane.31
In a comparative study, several commercially available “zinc amino acid che-
lates” were obtained. All were claimed to have the same percentage of zinc in their
products. Each was subjected to a molecular weight analysis using mass spectros-
copy. Table 4.1 summarizes the analytical results.32 Based on molecular weight and
the official Association of American Feed Control Officials (AAFCO) definition
requiring that an amino acid chelate have a molecular weight of fewer than 800 dal-
tons to be classified as such, it is obvious that only one of these products was able to
be defined as an amino acid chelate. In the case of the other “chelates,” the molecular
weights of the ligands were too great to define them as amino acid chelates.
Referring to Table 4.1 and assuming that each of the zinc chelates utilized either
the heaviest amino acid, tryptophan, or the smallest ligand, glycine, as a ligand, the
minimum number of amino acids composing the ligands of each of the chelates in
Table 4.1 may be calculated. Tryptophan has a molecular weight of 204.2 daltons.
Glycine has a molecular weight of 75.1 daltons. For free amino acids to chelate, they
TABLE 4.1
Molecular Weights (daltons) of Five Zinc Amino Acid Chelates
Total Calculated Zn Calculated Amino Acids
Molecular Weight Molecular Weight Molecular Weight
Company A 274.7 65.4 208.3
Company B 9,175.6 65.4 9,109.2
Company C 1,569.8 65.4 1,504.4
Company D 91,746.0 65.4 91,680.6
Company E 2,038.0 65.4 1,972.6
must each lose one hydrogen, which makes the molecular weight contributions of
these ligands 203.2 and 74.5 daltons. Thus, in theory, the ligand of Company A would
contain between 1.0 and 2.8 amino acids. However, for the larger molecular weight
products in Table 4.1, it is presumed that partially hydrolyzed protein molecules
are represented in the products. In addition, each joined amino acid has lost a mol-
ecule of water (18 daltons) in the creation of each peptide linkage. Thus, Company
B’s ligand would contain between 48.8 and 159.2 amino acids; Company C’s ligand
would have between 8.1 and 26.0 amino acids; Company D’s ligand would contain
between 492.8 and 1,605.3 amino acids; and finally, Company E’s ligand would have
between 10.5 and 34.2 amino acids. With the exception of the product produced
by Company A, none of the other products could be considered to be amino acid
chelates. At best, they may be metal proteinates or metal complexes (not chelates).
Only because the companies that manufactured them labeled their zinc products as
“chelates” will they be referred to as chelates in the comments that follow.
Payne has reported that intestinal absorption of a compound composed of amino
acids is related to the number of amino acids in the product. The larger the peptide or
protein molecule, the less likely it will be absorbed.33,34 In practical terms, a peptide
molecule composed of more than three amino acids is not generally absorbed intact
through the intestine.35
From the theoretical calculations discussed, only one company, Company A, pro-
duced a zinc amino acid chelate that could potentially be absorbed intact and would
be classified as a nutritionally functional chelate. The ligands of the other companies
would require further digestion before any absorption could occur. In the process of
digestion, peptide bonds between amino acids would have to be broken, and depend-
ing on the location of the zinc ion in those larger chelates, the metal could be released
from the ligand. Gastric hydrolyzation of a large molecular weight proteinaceous
chelate is like digestion of a protein-rich food. Both processes transform the protein
into the smallest possible absorbable molecule. During the process, integral com-
ponents, such as minerals, can, and often are, removed from their original protein
ligands in the food. For a chelate to be nutritionally functional, it must have a ligand
that is small enough to allow the chelate to be absorbed intact.1,31 Larger molecular
weight chelates are not nutritionally functional because they can potentially release
their metals through hydrolysis. The released metal ion can subsequently be exposed
to other dietary ligands, like phytic acid, and if complexed by that ligand, would
result in a reduction or prevention of the mineral ion absorption.
Even though a nutritionally functional chelate may have the correct stability
constant and be a small molecule, there is a third requirement for a nutritionally
functional chelate. The ligand must be easily metabolized by the body following
absorption of the chelate. In theory, if chelates were absorbed and the ligand subse-
quently cleaved from the metal in the cytoplasm of the absorptive cell, and if that
ligand were subsequently eliminated by the body intact without further metabolism,
then that chelate could not be considered to be nutritionally functional. It is possible
that if the ligand were cleaved from the metal and then not metabolized, by its very
reactive nature, that ligand would be attracted to a replacement metal. If the replace-
ment metal were an essential element, then the presence of the ligand in the cell
could create a new mineral deficiency because the ligand, being foreign to the body,
will ultimately be eliminated and take that newly acquired mineral with it. Prior to
being eliminated, the nonmetabolizable ligand would tie up the replacement mineral
and prevent it from entering into other essential biological uses.18,31 Thus, to be nutri-
tionally functional the ligand must be metabolizable following absorption.
In some instances, the ligands of absorbed chelates do not release their original
minerals even though they are absorbed. For example, Kratzer and Vohra have noted
that iron EDTA has a low molecular weight and is readily absorbed through the
intestine. Once absorbed, however, the body has difficulty metabolizing the EDTA.
Consequently, most of the EDTA is eliminated in the urine. Because the iron, in this
example, was chelated to the EDTA, it will also be eliminated in the urine attached
to the EDTA. In this case, ingestion of iron EDTA will not effectively deliver its
iron to the body because the EDTA ligand cannot be metabolized and the iron, being
strongly bound to that EDTA ligand, is eliminated concurrently.31
If they could be absorbed as intact chelates, both citric acid and ascorbic acid
chelates would be good potential nutritionally functional chelates because the ligands
of both chelates (citric acid and ascorbic acid) can be metabolized by the body.
Admittedly small amounts of citrates or ascorbates may be absorbed intact, but
generally the stability constants of these chelates are so low they generally do not
meet the requirements for a nutritionally functional chelate. They tend to break apart
in the stomach and not be absorbed as intact chelates. Lactic acid also falls into the
same category.
As ligands, amino acids meet all the criteria necessary to create a totally nutri-
tionally functional chelate.1 As noted, their stability constants can be ideal—not too
strong, but not too weak. They form chelates having low molecular weights. Even
with a molar ratio of three amino acid ligands to one metal ion, the molar weight
of the resultant chelate remains well under the 1,000-Da cutoff point proposed by
Kratzer and Vohra.31 And finally, the amino acid ligand will be hydrolyzed from the
metal ion in the mucosal cells and can be metabolized as any other absorbed amino
acid would be metabolized.
To illustrate this concept, an in vitro study was designed that allowed iron che-
lated to lysine to be absorbed into mucosal tissue from rat ileum.36 Based on the
corrected counts per minute, the original ratio of 2 moles of lysine to 1 mole of iron
in the mucosal solution became 3.05:1 in the mucosal tissue and 1.56:1 in the serosal
solution (p < 0.05), which indicated that at least some of the chelate had been hydro-
lyzed into iron and lysine in the mucosal tissue and then transferred to the serosal at
different rates.
The lysine that was dissociated from the chelate during chelate hydrolysis in the
mucosal tissue would have subsequently been metabolized as shown in Figure 4.1 if
this hydrolysis had occurred in a living animal instead of in vitro intestine loops.37
In conclusion, only when the three basic requirements are met (the correct stabil-
ity constant to protect the mineral while still allowing it to be hydrolyzed following
absorption, a molecular weight under 1,000 Da, and metabolizable ligand) can a che-
late be considered to be totally nutritionally functional. If any one of the criteria is not
met, the chelate cannot be considered nutritionally functional, even if it is absorbed.1,31
HI-PROT REGEN NH+3 CH2CH2CH2CH2CHNH+3 COO– HI-PROT (sl.)

GLUCAG CORT LYS
–
—OOCCHCH2CH2COO– OOCCH2CH2COCOO– NH+3 CH2CH2CH2CH2COCOO–
KG Ketoaminocaproate
NH
CH2CH2CH2CH2CHNH+3 COO–
Saccharopine
COO–
N
HCOCH2CH2CH2CHNH+3COO– –
OOCCH2CH2CHNH+3COO– 1-Piperidine-2-carboxylate
Aminoadipate semialdehyde GLU
–
OOCCH2CH2CH2CHNH+3COO–
Aminoadipate
COO–
– N
OOCCH2CH2CH2COCOO– H
Ketoadipate Pipecolate
CO2
–
OOCCH2CH2CH2COCoA
Glutaryl-CoA
COO–
N
– 1-Piperidine-6-carboxylate
OOCCH2CH=CHCOCoA
Glutaconyl-CoA
CO2
CH3CH=CHCOCoA CH3COCH2COCoA
Crotonyl-CoA AcOAcCoA
FIGURE 4.1 The catabolism of lysine. (Redrawn from Schepartz, B, Regulation of Amino
Acid Metabolism in Mammals (Philadelphia: Saunders) 98, 1973.)
REFERENCES
1. Ashmead, SD, “The chemistry of ferrous bis-glycinate chelate,” Arch Latin Am Nutr
Suppl, 51:7–12, 2001.
2. Glusker, JP, “Structural aspects of metal liganding to functional groups in proteins,” in
Anfinsen, CB, Edsall, JT, Richards, FM, and Eisenberg, DS, eds., Advances in Protein
Chemistry (San Diego, CA: Academic Press) 42:1–76, 1991.
3. Martell, AE, and Calvin, M, Chemistry of the Metal Chelate Compounds (New York:
Prentice-Hall) 1952.
4. Martell, AE, “Chemical structure and metal-binding action,” in Seven, MJ, Metal
Binding in Medicine (Philadelphia: Lippincot) 1–18, 1960.
5. Furst, A, Chemistry of Chelation in Cancer (Springfield, IL: Thomas) 12, 1963.
6. Ashmead, HD, Graff, DJ, and Ashmead HH, Intestinal Absorption of Metal Ions and
7. Pettit, LD, and Hefford, RJW, “Stereoselectivity in the metal complexes of amino acids and
dipeptides,” in Sigel, H, ed., Metals in Biological Systems (New York: Dekker) 9:173–212,
1979.
8. Hughes, M, The Inorganic Chemistry of Biological Processes (London: Wiley) 71–72,

1972.
9. Bjerrum, J, Schwarzenbach, G, and Sillen, LG, Stability Constants of Metal Ion
Complexes (London: Chemical Society) 40, 1957.
10. Beck, B, and Ashmead, SA, “An analysis of Zn, Fe, Cu, Mg, Ca, Mn and Co chelates
from seven commercial manufacturers,” unpublished research study, Snow College,
Ephraim, UT, 1990.
11. Ashmead, H, Ashmead, D, and Jensen, N, “Chelation does not guarantee mineral metab-
olism,” J Appl Nutr 26:5–21, Summer 1974.
12. Bates, GW, Boyer, J, Hegenauer, JC, and Saltman, P, “Facilitation of iron absorption by
ferric fructose,” Am J Clin Nutr 25:983–986, 1972.
13. Vohra, P, and Kratzer, FH, “Influence of various chelating agents on the availability of
zinc,” J Nutr 82:249–256, 1964.
14. Ashmead, HD, “Picolinic acid chelates,” Nutr Diet Consultant 11:3, November 1989.
15. Evans, G, and Hahn C, “Copper and zinc binding components in rat intestine,” Adv Exp
Med Biol 48:285–297, 1974.
16. Evans, G, Grace, CJ, and Votava, HJ, “A proposed mechanism for zinc absorption in the
rat,” Am J Physiol 228:501–505, 1975.
17. Evans, G, “Zinc absorption and transport,” in Prasad, A, ed., Trace Elements in Human
Health and Disease (New York: Academic Press) 1:181, 1976.
18. Seal, C, “Influence of dietary picolinic acid on mineral metabolism in the rat,” Ann Nutr
Metab 32:186–191, 1988.
19. Rubin, M, Houlihan, J, and Princiotto, JV, “Chelation and iron metabolism I: Relative
iron binding of chelating agents and siderophilin in serum,” Proc Soc Exp Biol Med
103:663–666, 1960.
20. Rubin, M, and Princiotto, JV, “Synthetic amino acid chelating agents and iron metabo-
lism,” Ann NY Acad Sci 88:450–459, 1960.
Chelates (Springfield, IL: Thomas) 217, 1985.
22. Kratzer, FH, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 42–44,
1986.
23. Doney, RC, Smith, RL, and Wiebe, HH, “Effects of various levels of bicarbonate, phos-
phorus and pH on the translocation of foliar-applied iron in plants,” Soil Sci 89:269–275,
1960.
24. Bukovac, MJ, and Wiltwer, JH, “Absorption and mobility of foliar applied nutrients,”
Plant Physiol 32:428–435, 1957.
25. Graff, DJ, Ashmead, HD, and Hsu, HH, “Radioactive isotope studies in plants,” unpub-
lished report, Albion Laboratories, Clearfield, UT, 1980.
1986.
27. Johnson, LR, ed., Physiology of the Gastrointestinal Tract (New York: Raven Press)
vol. 2, 1981.
28. Guyton, AC, Textbook of Medical Physiology (Philadelphia: Saunders) 1976.
29. Hofner, W, “Eisen and manganhaltige. Verbindungen im blutungassafit von helianthus
annaus,” Physiol Plant 23:673–677, 1970.
30. Tiffin, LO, “Translocation of micronutrients in plants,” in Dinauer, RC, ed., Micro
nutrients in Agriculture (Madison, WI: Soil Science Society of America) 207, 1972.
1986.
32. Johnson, B, “Molecular weights of several commercial chelates,” unpublished report,

Albion Laboratories, Clearfield, UT, 1974.
33. Payne, JW, “Transport of peptides in microorganisms,” in Matthews, DM, and Payne, JW,
eds., Peptide Transport in Protein Nutrition (Amsterdam: North-Holland) 283–364, 1975.
34. Payne, JW, “Peptides and microorganisms,” in Rose, AH, and Tempest, OW, eds.,
Advances in Microbial Physiology (London: Academic Press) 55–113, 1976.
35. Burston, D, Marrs, TC, Sleisenger MH, Sopanen, T, and Matthews, DM, “Mechanisms
of peptide transport,” in Elliott, K, and O’Connor, M, eds., Peptide Transport and
Hydrolysis (Amsterdam: North-Holland) 79–98, 1977.
37. Schepartz, B, Regulation of Amino Acid Metabolism in Mammals (Philadelphia:
Saunders) 98, 1973.
5 The Development
of Analytical Methods
to Prove Amino
Acid Chelation
Since the discovery of amino acid chelates and their application to mineral nutrition,
much has been written about their benefits. Nevertheless, little attention has been
paid to the requirements necessary to actually produce a nutritionally functional
chelate. The reacting of metals with amino acids to form chelate molecules is a very
exacting science. Frequently, due to incomplete understanding of chelation chemis-
try by certain manufacturers, the requisite heterocyclic ring structure is not the end
result. Instead, a type of amino acid complex often results in which the metal may
be attached to the end of an amino acid, or a metal salt could be formed, or a simple
admixture of protein/amino acids and metal salts could be made.
Beck and Ashmead1 conducted a study in which they purchased what were
reported to be metal amino acid chelates from several commercial manufacturers.
These chelates included zinc, iron, copper, magnesium, calcium, manganese, and
cobalt. Samples from each product were first analyzed for metal content by atomic
absorption spectroscopy. A second sample of each product was assayed for nitrogen
by the Kjeldahl method as an indication of its amino acid content. Based on the
label claim that each product was an amino acid chelate, it was assumed that the
ligands were sourced from amino acids and not partially hydrolyzed protein. (If the
chelates were produced from partially hydrolyzed protein, the amount of available
ligand to chelate the metal would have been significantly less.) Using the amino
acid ligand assumption, the molar ratio of each of the metals to amino acids was
subsequently calculated. A third sample of each of the products was put in distilled/
deionized water and stirred for 30 minutes before being filtered. Both the soluble
and insoluble fractions were subsequently analyzed for metal and nitrogen contents,
using the same procedures employed in analyzing the original product. The molar
ratios of metal to amino acids were then calculated for the products in the soluble and
insoluble fractions and compared to the original product analysis and ratios.
With the exception of one company, which maintained a consistent molar ratio of
metal to amino acid in the original product as well as the soluble and insoluble frac-
tions, in all of the other companies’ products a majority of the amino acid ligand was
found in the insoluble fraction and the metals in the soluble fraction. For most of the
analyzed product, there was an insufficient amount of amino acids to bond the amino
acids to the metals and create chelates. Most of the commercial products appeared to
61
be admixtures due to the ease of separating the mineral salts from the amino acids
in a solution of water.
These investigators concluded:
It has become evident that many manufacturers are confusing the terms, admixture
and complex, with the term, chelate. They are not synonymous in any sense. The data
indicate that many companies claim to have chelation but can only offer a mixture that
may have occurred in an aqueous [or even a dry] environment. They do not understand
that chelation means chemical reaction, forming and breaking bonds and [creating] a
new, more bioavailable product.1
Parks and Harmston secured commercial samples of copper amino acid chelates/
proteinates from eight different manufacturers.2,3 Their initial premise was that by
putting each “chelate” in solution and analyzing the filtered solution, it would indi-
cate the association between the copper and the ligand. They believed that the con-
centration of unbound (free mineral), free ligand and complex mineral recovered in
the soluble fraction could be used to calculate the stability of a chelate.
As they subsequently discovered, the basic problem with their analytical pro-
cedure was the assumption that all of the products that they were analyzing were
actual chelates. As was previously demonstrated by Beck and Ashmead, calling a
product an amino acid chelate on the label did not guarantee that it was actually
chelated.1 Thus, when Parks and Harmston based their analytical conclusions on
results obtained from amino acid complexes rather than true chelates, their conclu-
sions could not produce reliable data for analysis of amino acid chelates.
Ligands have certain requirements that dictate how they will be bound to metals.
Assuming that these investigators were analyzing actual chelates, there are many
factors that will influence bonding affinity of the ligand, including, but not limited
to, ring size of the chelate, the number of ligand attachment points to a single metal,
the Lewis basicity of the donor ligand, the size and charge of the metal, the nature
of the donor ligand, whether the resonance within the chelate ring is interfered with,
and finally the steric hindrance of the members forming the ligand.4 Thus, a ligand
with the potential to form a very strong bond may not always bond or bond optimally
due to interference from one or more of these factors, and the chelate assumptions
based on the potential of the ligand to bind the metal are not absolute.
As a result of Parks and Harmston’s erroneous assumptions (1) that all of the
products labeled as chelates were actually chelates and (2) that the potential stability
constant between the metal ion and the ligand is the sole criterion for determination
of bound metal concentration in the product, they concluded that their analytical
procedure could not determine if the bound copper they recovered in the soluble frac-
tion was actually chelated. They reported, “This could be copper sulfate, soluble pro-
tein, copper complex, organic acid copper complex, dipeptide copper complex and/or
amino acid copper complex.”2
Leach and Patton later revived the Parks and Harmston technique and, after some
slight modifications, published an article describing their analytical procedure in
an industry trade magazine.5 They advocated assaying for metal content by atomic
absorption spectroscopy and nitrogen by the Kjeldahl method. Their basic assumption
The Development of Analytical Methods to Prove Amino Acid Chelation 63
was if the initial product contained, for example, 15% metal and 30% protein and if
they recovered that same percentage of metal and protein in the residue after filtra-
tion of the “chelate,” then the product must be chelated. Their procedure, however,
contained three basic assumptions that could potentially result in an erroneous con-
clusion. First, as with the Parks and Harmston procedure, they assumed that if the
metal and nitrogen assays met the label claim, then the product was a chelate. They
did not consider that the metal and nitrogen (protein) could create other molecules
with the same metal-to-nitrogen ratio. Second, they assumed that an amino acid
chelate could never be soluble, and thus they erroneously concluded that the filtrate
solution would not contain any chelated mineral. Third, their procedure focused on
the physical weights of the products being analyzed instead of the molar ratios of the
individual components that composed the products. For example, it was assumed
that the actual physical weights of the ligand and metal reflected a 2:1 (amino acid-
to-metal) chelate, when in fact the 2:1 ratio can only be calculated using the molar
weights of the constituents.
Using the logic expressed in the Leach and Patton’s article, if a small molecular
weight protein ligand, such as insulin, which has a molecular weight of 5,808, were
used as a ligand, an iron chelate containing 15.0 g iron would require 30.0 g insulin to
form a 2:1 chelate based on physical weight. If 30.0 g insulin were employed as a ligand
and if a 2:1 ligand-to-metal molar ratio chelate were being sought assuming a 100%
reaction, on a molar ratio basis, only 0.75% of the iron in the product could potentially
be chelated.6 The major problems with the analytical procedure of Leach and Patton as
it appeared in print in the magazine was substituting physical weights for molar require-
ments in producing a theoretical chelate and the broad assumptions that all insoluble
mineral protein complexes are chelates without proving the chelate structure first.
Presumably, based on the Association of American Feed Control Officials
(AAFCO) definition, a metal amino acid chelate must have a molecular weight of
under 800 Da, while proteinates can be much larger molecules.7 Accordingly, Hynes
et al. proposed an analytical method that employed ultrafiltration.8 They theorized
that a small molecular weight mineral would pass through the filter, whereas larger
chelated molecules would not. To filter the “chelate,” it first had to be solubilized.
They mixed insoluble metal proteinates (metals chelated to amino acids and/or par-
tially hydrolyzed protein7) with water, elevated the pH of the solution, and added eth-
ylenediaminetetraacetic acid (EDTA). This caused the insoluble metal proteinates
they were working with to go into solution. These solutions were then run through
an ultrafiltration process in which only molecules under 500 Da were allowed to pass
through the filter. Since, by definition, metal proteinates have molecular weights that
are higher than 800 Da,7 it was presumed that the insoluble fractions contained all of
the metal proteinates. The recovered residues were then assayed for mineral contents
and presumed to be chelated since they did not pass through the filter, and the origi-
nal products were claimed to be metal proteinates.
There are several problems with this procedure. The first and most obvious is that
assaying the residue for a particular metal does not prove the metal is chelated. It
simply proves that the metal is present and attached to some sort of insoluble ligand
that was too large to pass through the filter. It does not describe what the metal is
attached to or what, if any, type of bond exists between the ligand and the metal.
A second problem may potentially result from attempting to solubilize the metal
proteinate. A metal proteinate, by definition, has a structure that does not have a
very high stability constant because of the size of its ligand.4 Such molecules are
easily broken apart. If the pH of the aqueous environment in which a metal protein-
ate is placed is elevated above 9, it will cause the amine portion of the proteinaceous
ligand to detach from the metal and result in the loss of the heterocyclic chelate
ring.9 Subsequent exposure of this molecule to a solution containing EDTA with its
higher potential formation constant may result in tearing the metal away from the
protein ligand and creating a soluble metal EDTA chelate that has a much smaller
molecular weight and a higher stability constant than the original metal proteinate
product.10 The comment of Hynes et al. that their analytical results correlated very
closely to studies conducted with EDTA chelates alone suggests that at least part
of the metal was chelated by the EDTA as they attempted to solubilize the metal
proteinate. In other words, if the metal proteinate being analyzed was subjected to
a high pH and a solution of EDTA, much of the original product would have been
destroyed, which would have produced erroneous results when the product was sub-
jected to ultrafiltration. Thus, a method that requires that the product be modified
(i.e., solubilized before analysis) runs the risk of destroying the very product that the
analysis is designed to assess.
As early as 1976, paper chromatography was suggested as another method to prove
chelation.11 Like the problem faced by Hynes et al., this analytical procedure did not
provide the hoped-for data due to lack of solubility of some amino acid chelates. The
separation methodology, which chromatography requires, can only occur with prod-
ucts that are in solution.12 When insoluble chelates are subjected to chemical methods
to enhance their solubility for chromatographic analysis, the chelate structures run
the risk of being destroyed, similar to what appeared to have occurred in the tech-
nique of Hynes et al.8 Other investigators have also rejected chromatography for the
same reasons.5 With the development of soluble amino acid chelates,13 some inves-
tigators have suggested that this analytical procedure should be reexamined.14 Most
amino acids do not have a chromophore moiety to absorb light and thus are colorless.
Therefore, chromatography will not work unless a chromophore is derivatized, and
that may also affect the structure of the “amino acid chelate” being analyzed.
When Alfred Werner published his first paper on chelation in 1893, the molecule
he envisioned was theoretical.15 He had no instrumental means to prove that the
structure of the molecule that he described on paper actually existed. His theoretical
concept, however, better explained the chemical phenomena that he was observing
than did any of the other chemical structural theories of the day.
Following Werner’s discoveries, other investigators also studied the chelate phe-
nomenon, but, like Werner, none of them had the analytical methods at their disposal
that were necessary to prove directly that the molecules that they were working
with were actually the same chelate structures that were being drawn on paper.
Consequently, most of these investigators simply focused on the characteristics asso-
ciated with what they believed were chelates, characteristics that were explained by
the theoretical structural drawings.
Ley, whose observations further elucidated the theoretical structure of the chelate,
used copper and glycine as the metal and ligand, respectively, to make his chelate.16
0
H
O O
H3C N N CH3
C C
M
C C
H 3C N N CH3
O O
H
(x)
FIGURE 5.1 The inner metallic complex salt as envisioned by H. Z. Ley. (Redrawn from
Werner, A., “Beitrag zur Kostitution Anaorganischer Verbindungen,” Z Anorg u Allem Chem
3:267, 1893.)
As described previously, he reportedly produced the chelate in solution using a ratio

of 1 mole of copper ion to 2 moles of glycine. When the product was dried, it had
the same molar ratio of copper to glycine. Its electrical conductivity was very low,
leading Ley to conclude that a new copper compound had been created in which the
copper was nonionic. Figure 5.1 illustrates the type of compound that Ley believed
he had created, which he called an “inner metallic complex salt.” The figure he
drew was based on his observations related to the behavior of the compound and not
on any structural chemistry. As was the case with Werner, there was no analytical
equipment at his disposal that could actually prove the compound he had created was
truly a chelate. It would take decades to develop the requisite equipment.
Ashmead and Little noted a similar nonionic characteristic with the mixed amino
acid ligand chelates with which they were working. They reasoned that a loss of
electrical conductivity could be used as a simple proof of chelation.17 They inserted
electrodes into a solution containing an ionized metal or into a solution containing
the same mineral chelated to hydrolyzed protein. They connected the electrodes to
a voltmeter that measured the electrical current in millivolts to ascertain the current
between the electrodes in each of the two solutions. They noted that when the amino
acid chelates were put into the solution, there was little or no electrical conductiv-
ity. Initially, it was believed that this lack of electrical conductivity was a simple
but accurate proof that the mineral in question was chelated.18 It was only later,
when these investigators realized that insoluble metal salts, such as ferric phosphate,
exhibited similar characteristics that they abandoned this electrical conductivity
technique as a proof of chelation.
As a result of this early research, however, a decision was made to look at polar-
ography as a potential way of proving chelation.19 Polarography measures the elec-
tron potential that an electrochemically active atom achieves between the oxidized
and reduced states in an equilibrium condition. The instrument used in this analysis
is designed to sweep solutions of compounds with progressively increasing amounts
of electromotive force (emf). When the oxidized species of the atom being measured
begins to accept electrons and become reduced, the residual current applied to the
system begins to change. As more emf is applied and the oxidation-reduction poten-
tial surpasses equilibrium, the current through the system will again reach a limiting
plateau. The result, on current output, is a sigmoid curve, and the point of inflection
is equal to the equilibrium state in the oxidation-reduction equation. This is referred
to as the half-wave potential E1/2. Each atom has a characteristic half-wave potential,
but if other atoms are bound to that atom, such as in the case of a chelate, then more
emf is required to ionize and reduce the atom. The number of bonds on the atom can
thus be assessed by the degree of E1/2 shift.20 It was believed that an E1/2 shift would
demonstrate that a metal was chelated.
An analysis of zinc glycine solution was carried out using differential pulse polar-
ography. Solutions were prepared in which the zinc concentration was maintained
at 1.0 × 10 –4. The concentration of glycine was varied in the solutions from 0.09
to 2.7 moles. The control solution contained an ionized zinc solution without gly-
cine. As seen in Figure 5.2, as the concentration of glycine increased, there was a
definite negative shift to the right in the curves, indicating that a reaction between
0.8 1.00 1.30
1.02
1 H 2O
KNo3 (0.1M)
+2
Zn (0.0001)
100UA
pH = 7
2 Gly (2.7M)
KNo3 (0.1M)
+2
1 Zn (0.0001)
1.26 pH = 7
1.27 3 Gly (0.90M)
Signal Amplitude
4 1.29 KNo3 (0.1M)

1.25 Zn+2 (0.0001M)
pH = 7
4 Gly (0.30)
KNo3 (0.1M)
3 2 Zn+2 (0.0001M)
pH = 7
5
5 Gly (0.09)
KNo3 (0.1M)
Zn+2 (0.0001M)
1.04 pH = 7
1.07
1.11
1.14
Applied Potential
FIGURE 5.2 Polarographic spectra of zinc amino acid chelate. (Redrawn from James, H,
“A differential pulse polarography study of zinc glycinate chelate,” unpublished research
report, Weber State University, Ogden, UT, 1987.)
the zinc and glycine had occurred.19 One could speculate that the reaction being
observed was proof that a chelate had been formed since the applied potential had to
be increased. Nevertheless, there was nothing in the work that actually proved that
the observed reaction was a chelate reaction because it did not prove the molecular
structure included a heterocyclic ring. The zinc could just as easily have formed a
complex with the glycine.
Ashmead and Little were not the only investigators who tried to use polarog-
raphy to prove chelation. Other researchers have also explored this analytical tool
as a way to prove chelation. Holwerda et al. published a study in which they sub-
jected zinc, which they believed was chelated to amino acids or partially hydro-
lyzed protein (zinc proteinate), to polarographic analysis.21 They stated that when
the zinc was chelated, the resulting molecule enhanced “the difficulty of reducing
Zn+2 to a state in which all chelation is lost.” They subsequently compared the
polarographic curves of a solution of zinc sulfate mixed with glycine to a solu-
tion of zinc proteinate and, based on the similarity of the curves, erroneously
concluded that both were chelated. In spite of this conclusion, they advocated that
polarography be employed as a standard method for testing the degree of chela-
tion of a metal complex.
There are several other concerns related to the Holwerda et al. analytical method.
The first, of course, is that, as demonstrated by James,19 there is no proof that the
chemical reaction occurring in the solution between the glycine and zinc is actually
a chelate reaction. A different zinc molecule may have been formed. The product
resulting from the mixing of glycine and zinc ions in solution may not be a chelate,
but instead a zinc complex. If that occurred, the conclusion that the product was
chelated, based on the polarographic observations, would be faulty. Holwerda et al.
provided no proof that the molecule formed in these studies was truly a chelate and
not something else. They just assumed it was a chelate.
Holwerda et al. also claimed that measuring the half-wave potential, as a function
of excess ligand concentrations, allowed the calculation of both the formation con-
stant and the average number of ligands bonded per metal ion.21 The problem with
this assertion is determining at what point excess ligand exists. Even using saturated
solutions of the products would not ensure excess ligand concentration because the
proportions of ligand to metal in an intact molecule are not changed by saturation.
Further, such an observation would be nullified if the ligands were not attached to
the metal and the two products had different solubilities in water. If that were the
case, neither a chelate nor even a complex would be assessed by their method.
Cao et al. attempted to validate this polarographic analytical method by assay-
ing eight commercial sources of what were reported to be chelated zinc.22 Four of
the products they tested were not chelated zinc. According to the manufacturers
of the tested products, these four products were, instead, amino acid complexes of
zinc. Nevertheless, these investigators tested the complexes as if they were all che-
lated zinc sources and concluded, “The different organic products examined in the
present study showed different degrees of chelation effectiveness as determined by
polarography.”22 As stated previously, four of the products tested were not chelated,
but according to the results of Cao et al., these amino acid complexes still fit the
polarographic criteria necessary for them to be called chelates.
Professor Robert Hancock, an internationally known author, lecturer, and expert

on the behavior and bonding characteristics of elements in aqueous solutions,
reviewed the Holwerda et al. data and concluded that the method is
extremely crude … [and not] sophisticated analysis made to determine the presence of
chelation. … The method of Holwerda et al. could obtain a crude relationship, but this
is far short of what would be required to predict anything significant about the pres-
ence of a chelate. … [They] made too many presumptions on what the possible ligands
could be and what the characteristics of their breakage would be.23
Ashmead hypothesized that nuclear magnetic resonance (NMR) spectroscopy

could potentially provide sufficient structural analysis to prove chelation of the met-
als by amino acids. Ligand coordination of various diamagnetic metal ions leads to
definite changes in electronic distributions.24 Ashmead believed that these changes
could be observed through NMR analysis.
NMR spectroscopy depends on the properties of certain charged, spinning nuclei
to generate a magnetic field with an axis directed along the spin axis of the nucleus.
When solutions of these atoms are placed in a strong magnetic field and spun, the
nuclei will align themselves nearly parallel to the axis of the larger magnet. The
slight displacements of the axis of the nuclei cause wobbles, which result in a proces-
sional migration of the poles of the axis. When a resonating frequency from the loops
of a radio-frequency (RF) generator wrapped around the sample tube equal the fre-
quency of the processional on the magnetic poles, the nuclei absorb energy and flip to
a higher spin state, which processes against the larger magnetic field. The most com-
mon isotope to examine by NMR is the hydrogen atom (1H) followed by carbon-13
(13C). NMR data can assess the number and positions of hydrogens (or other atoms)
in a molecule and structural relationships to near neighbors. Interactions between the
resonant nuclei and the rest of the molecule, different relaxations, peak splitting, and
position shifting can provide more information regarding the structure and configu-
ration of the molecules in the sample.25
Under Ashmead’s direction, Olson and Schweizer26 dissolved 25-mg samples of
glycine, zinc glycinate, and calcium glycinate in 1 mL of heavy water (D2O) and
then added dimethylsulfoxide (DMSO)-d6 for the 1H work. Using a high-resolution
IBM AF 200 spectrometer, spectra were collected at 23°C and the chemical shift
measured from an internal standard of 2,2-dimethyl-2-silapentane sulfinate for the
13C work and tetramethyl silane for the proton data.
There was a substantial shift in the spectra representing the carboxyl carbon
in both the calcium and zinc amino acid chelates at between 5 and 5.7 ppm. The
researchers reported that the shift was reproducible and reflected electronic charge,
shifting from the carboxyl function into the valence shell of the metal. Another shift
of electronic charge occurred in the area of 0.5 to 1.5 ppm. This reflected the point at
which the carbon was attached to the nitrogen. The metal nitrogen bond was further
indicated by the proton chemical shift differences. This was particularly evident
when either a zinc or a calcium glycine chelate was compared to the same metal as a
complex. The 1.9-ppm shift to NH2 protons from the complex to the chelate reflected
the shielding produced by the bond formed between the electropositive zinc and
nitrogen atoms in the chelate.26
The investigators concluded from their study that the NH2 peak of the chelate
samples being analyzed was very broad compared to free glycine. This indicated
“a change in exchange or relaxation effect probably due to a positively charged proton
or cation.”26,27 The shift was much greater in the chelate compared to the complex.
In spite of these results, the basic problem with NMR was that of solubility. At
the time that this study was conducted, NMR required that the product be in a liq-
uid state to be analyzed. When solubility did not occur, the product could not be
analyzed. Not all amino acid chelates are soluble.28 Modifying an insoluble chelate
to put it in solution could potentially affect the integrity of the analytical results,
as observed previously. Thus, the search continued for a more universal analytical
procedure. Since this initial work with NMR, solid-state NMR analysis has been
developed and investigation of insoluble amino acid chelates commenced. To date,
no conclusive data have been published, although investigators are optimistic.
Another analytical procedure that was examined by Ashmead et al. was electron
paramagnetic resonance (EPR) spectroscopy. It focuses on atoms that have electrons
with unpaired spins. If compounds containing these atoms are placed in a strong,
static, magnetic field and are then swept with different frequencies of microwave
energy, magnetic moment resonance and energy absorption will occur at particular
frequencies, resulting in a spin-state change in the unpaired electron. The position of
resonance is indicated by the g-value, which is constant for unbound, unpaired elec-
trons but variable when an unpaired electron is involved in a bond between atoms. The
g-value is thus an indication of the bonding environment. EPR spectra are generally
recorded as the derivative of the actual resonant peaks to increase the sensitivity of
the observation. The resulting derivative curves can be further split up by interactions
of the resonating electron with the magnetic moments of nuclei. This is referred to
as hyperfine splitting and can be used to further identify the bonding environment.29
Transition metals such as iron, copper, and manganese have unpaired electrons in
their orbitals. This unpaired spin generates a magnetic moment that can be lined up
with, or against, an external magnetic field. Since the environment of the transition
metal determines the energy required to flip the spin, this can provide a means for
identifying a complex or chelate.29 For example, in a study that compared the EPR
spectra of an iron amino acid chelate with a known iron chelate (Ferrichrome A),
comparable absorptions were observed for both Kramer’s lower and upper doublet
(see Figure 5.3).30
Using cryostatic EPR, chelates from mixed amino acid ligands or single amino
acid ligands were analyzed. The spectra demonstrated the presence of two bonds
from each amino acid ligand to the metal, which were splayed out at tetrahedral
angles.30 This is exactly what one would have predicted for an amino acid chelate. As
exciting as these analytical data were to the research team, EPR still did not show the
total structure of the amino acid chelate. It only strongly suggested that the product
being analyzed was a chelate. Further, the technique was applicable only to certain
transitional metals. Chelates of calcium, magnesium, or zinc could not be analyzed
by this method.
While most of these procedures elucidated chelation characteristics, none of
them proved that the organic metallic products being analyzed were truly amino
acid chelates. Certainly, the analytical results from NMR and EPR suggested
9 4.3 2.0
Ferrichrome A
9.6 4.2 2.0
Iron Amino Acid Chelate
FIGURE 5.3 Comparison of the EPR spectra between Ferrichrome A and iron amino acid
chelates. (Redrawn from Rogers, K, and Landcaster, J, “Report on electron paramagnetic
resonance spectra of Albion® metal amino acid chelates,” unpublished research report, Utah
State University, Salt Lake City, 1985.)
with convincing clarity that chelates were probably being formed, but none of
these procedures could actually prove it. “Physical characteristics alone (solu-
bility, percent nitrogen and approximate molecular weights) do not confirm the
existence of chelation.”31 Jeppsen and Ashmead added that the interest in proving
chelation has
led to a variety of tests which make subjective measurement of a quality other than
chelation and then attempt to correlate the results with the presence of chelation. All of
these methods have deficiencies and loopholes which would allow some compounds to
be identified as “chelates” which were not really chelated.32
The only way to verify that the product being examined is an amino acid chelate is
to look at the bonds that exist between the metal and the ligand.31
By the early 1970s, Ashmead et al. began to focus on an analytical procedure
that examined chelation bonds as an absolute way to prove chelation by employing
infrared (IR) spectroscopy for analysis of chelates.33 They reported that:
Each metal chelate in the body has a different molecular arrangement. This molecular
configuration, or fingerprint, can be demonstrated by an infrared spectrophotometer trac-
ing. Just as each individual’s fingerprints are different from anyone else’s, so does every
molecular configuration of the same metal have a different infrared spectrophotometry

tracing fingerprint. This is a means of identification of that particular metal compound.34
These researchers initially tried to build amino acid chelates that would gener-
ally match known chelate-containing materials. For example, in 1974, they reported
having built an iron amino acid chelate in the laboratory that was almost identical to
the presumed iron chelate in veal liver. Figure 5.4 reproduces a comparison of the IR
spectra published at that time. The IR spectrum of ferrous sulfate or iron amino acid
chelate is compared to the spectrum created from analyzing iron found in liver.34
A Comparison of the Infrared Spectrophotometry Tracings of

Iron Sulfate and Iron Amino Acid Chelate Found in the Liver
4000 3000 2000 1500 CM–1 1000 900 800 700
100 100
80 Iron in the liver 80

Transmittance (%)
60 60
40 Iron sulfate (FeSO4) 40
20 20
0 0
3 4 5 6 7 8 9 10 11 12 13 14 15
Wavelength (Microns)
A Comparison Between Iron Amino Acid Chelate and

the Iron Amino Acid Chelate in the Liver
4000 3000 2000 1500 CM–1 1000 900 800 700
100 100
Iron Amino Acid Chelate
80 80
Transmittance (%)
60 60
Liver
40 40
20 20
0 0
3 4 5 6 7 8 9 10 11 12 13 14 15
Wavelength (Microns)
FIGURE 5.4 IR spectra comparing iron amino acid chelate or iron sulfate to iron found
in veal liver. (Redrawn from Ashmead, H, Ashmead, D, and Jensen, N, “Chelation does not
guarantee mineral metabolism,” J Appl Nutr 26:5–21, Summer 1974.)
They explained that the rationale for this analytical procedure was that the closer
one builds a chelate to what the body produced, the greater would be the intesti-
nal absorption of the chelate.34 What this rationale overlooked was that the “chelate
standard,” used to compare against, in this case chelated or complexed iron in veal
liver, was not a pure amino acid chelate. While the iron in the liver may have been
chelated, the chelate molecule was part of a larger molecule. The IR spectrum of the
iron in the liver included many other substances that interfered with or hid the bonds
between the iron and its ligands in the liver, thus making an accurate IR spectrum of
a pure iron amino acid chelate impossible.
Another problem was the question of absorption. Layrisse and Martinez-Torres
published data showing that absorption of iron from veal liver was only about 21%.35
The reported absorption of the iron amino acid chelate was significantly higher.33
Thus, the naturally occurring organic minerals in food proved not to be good refer-
ence standards for proving chelation.
In spite of the initial approach to IR analysis, Ashmead et al. realized that the IR
spectra did reflect the stretching and bending of the bonds between atoms. When the
bonds forming the amino acids are bent and stretched to create the five-member che-
late rings, the resonance frequencies are forced to shift, thus suggesting chelation has
occurred. For these reasons, the authors continued to work with infrared spectroscopy.
Other investigators picked up on the work of Ashmead and his team and expanded
on it. Using x-ray powder diffraction followed by infrared spectroscopy, Fujita et al.
presented a method by which they analyzed iron chelated to amino acids that were
derived from hydrolyzed soybean protein.36 They could not recover iron crystals in
the product being analyzed, indicating that the iron in the product was bound to the
protein ligand. In the infrared spectrum of the iron amino acid chelate, the bond
describing the stretching of the COO- groups shifted to a higher frequency site when
compared to free hydrolyzed soybean protein. They concluded that the iron was
bound to the amino acids from the protein by one or more carboxyl groups. The IR
spectra proved that the product was not a mixture of amino acids and iron salts but
instead a coordination compound.
One of the references consulted by Fujita et al. was Nakamoto’s work, Infrared
and Raman Spectra of Inorganic and Coordination Compounds,37 He had pub-
lished, among other things, the infrared frequencies and bond assignments of gly-
cine in the crystalline state. He compared this to the infrared frequencies and bond
assignments of copper bisglycinate chelate. From this and other research, it became
obvious to Ashmead et al. that crystalline amino acid chelates were necessary to
create reference products to expand on Nakamoto’s work. These reference samples
were necessary before the IR spectra could be employed to prove absolute chelation.
From the multitude of chemical studies conducted over decades of time, the char-
acteristics associated with the structure of the amino acid chelate had become well
understood but still not seen. X-ray crystallography seemed to hold the promise of
actually seeing the chelate molecule. If the chelate could be seen visually, then that
visual molecule could become the reference standard to develop analytical methods
that proved chelation. Using x-ray crystallography, structural studies of amino acid
chelate crystals commenced in the 1970s.38
X-ray crystallography is a method of determining the arrangement of the atoms

that compose a crystal. An x-ray beam strikes the crystal and is scattered into many
different directions. Based on the angles of the scattered beams and their intensities,
it is possible to produce a three-dimensional picture of the density of the electrons
within a crystal. From this electron density, the mean position of atoms in a crys-
tal associated with the electrons can be determined along with the chemical bonds
between the atoms.39
The first structure to be ascertained using x-ray crystallography was sodium chlo-
ride—table salt—which occurred in 1914.40 Bragg et al. were subsequently able to
show that by the distribution of the electrons in the table salt, the crystals were not
comprised of covalently bonded molecules. These investigators proved the existence
of ionic bonding.41,42
While 1923 brought the beginning of the elucidation of organic structures,40
structural analysis of amino acid chelates began much later. The first chelates to
be examined were primarily chelates created by the body. For example, x-ray dif-
fraction crystallography revealed that the copper in the Type I proteins azurin and
plastocyanin was bound in a tetrahedral arrangement to two histidine molecules,
a cysteine molecule and a methionine molecule. The cysteine was the amino acid
ligand responsible for chelating the copper.38
When zinc avian pancreatic polypeptide hormone was crystallized and structur-
ally analyzed, it was discovered that the zinc was bonded with glycine to form a
five-member chelate ring. The bonds originated from the carboxyl and amino groups
of the ligand to the metal. The metal ion also cross-linked to three other individual
molecules whereby the crystal lattice was stabilized. Figure 5.5 illustrates the chela-
tion of the zinc in the avian pancreatic polypeptide hormone.43
While he had worked with chelates that utilized a single amino acid as a ligand
(as opposed to hydrolyzed protein that contained a multitude of different amino
acids), it was not until 1986 that Ashmead received a patent describing commer-
cial production of pure amino acid chelates.44,45 This new technology opened the
door to the possibility of growing pure amino acid chelate crystals from a com-
mercially produced product. If pure amino acid chelate crystals could be grown and
proven to be true chelates by x-ray crystallography analysis, then these crystals could
CH2 H
H2N NR
C
Zn2+
FIGURE 5.5 The structure of zinc chelated to glycine in crystallized avian pancreatic poly-
peptide hormone. (Redrawn from Anfinsen, CB, Edsall, JT, Richards FM, and Eisenberg,
DS, eds., Advances in Protein Chemistry Metalloproteins: Structural Aspects (San Diego,
CA: Academic Press) 296, 1991.)
subsequently be used as reference amino acid chelate samples for creating simpler
analytical procedures.
Using commercially prepared amino acid chelate sources, pure crystals from cal-
cium, magnesium, zinc, and copper glycine chelates were grown in the laboratory
and subsequently subjected to x-ray crystallography analysis using x-ray diffraction
spectrometry.46 This analysis determined the bonding angles, atom identifications,
and orientations of the ligand atoms that surrounded the metal atom in the chelate.
Taking zinc chelated to two glycine molecules as an example, Figure 5.6 pres-
ents the results of the x-ray crystallography diffraction analysis of this amino acid
chelate.46 The figure shows the chelate molecules and their crystalline alignment
to each other. It is particularly important to note the presence of four bonds between
each zinc atom and the two glycine molecules of the 2:1 molar (glycine-to-zinc)
molecules. The oxidation number, or valence, of the zinc is two. The bonds entering
each zinc atom demonstrate the presence of coordinate covalent bonds in which the
amine moiety has donated both electrons from the nitrogen atoms of the two glycine
molecules into receptive p-orbitals of the metal atom. (The d-orbitals are filled.)
It is also significant to note when comparing Figures 5.5 and 5.6 that the struc-
ture of this pure zinc glycine chelate is similar to zinc glycine chelate recovered
from avian pancreatic polypeptide hormone. The two bonds from the amino acid
to the zinc originated from the nitrogen and oxygen moieties of the ligand in both
chelate molecules. The angles of the bonds are also identical in both molecules. The
work reported by Dalley46 was the first time that structural analysis had conclusively
proven that a company claiming to make an amino acid chelate was actually produc-
ing such a molecule.
Using the data from the x-ray crystallography diffraction analysis generated by
Dally as reference standards, spectra were subsequently created by analyzing the
same chelate samples by IR spectroscopy. The net result was the development of
O
C
O
N C
C N
N C
Zn
O
C O O
O C Zn
C
O
O
C N
FIGURE 5.6 X-ray diffraction lot of 2:1 molar (glycine:zinc) Albion® Metal Amino Acid
Chelate showing participating elements, bonds, and spatial orientation of the molecules in the
crystal. (Redrawn from Dalley, NK, “Report on x-ray diffraction crystallography of Albion®
zinc amino acid chelate,” unpublished research report, Brigham Young University, Provo,
UT, 1991.)
reference spectra from proven amino acid chelates.47 The shifts in the IR spectra
indicated that the carboxyl bonding from the amino acid ligand to the metal matched
the published data of Nakamoto.37 There was also evidence of the amino bond
forming with the metal. IR spectroscopy was the first relatively simple, but absolute,
analytical method to prove that a particular product was truly an amino acid chelate.
Leach and Patton criticized the use of x-ray crystallography or x-ray diffraction
as impractical and, in effect, impossible.5 They stated that this technique can only
be used on pure, carefully grown crystals, and because they were unable to grow
crystals in their laboratory using their own proprietary commercially prepared
metal proteinate product, they concluded x-ray crystallography “has no applica-
tion on production samples.”5 The chelate produced by the company that employed
these investigators was presumably prepared by a patented method that required
reacting metal salts with a partially hydrolyzed protein slurry that was created by
a base-acid-base hydrolysis.48 Because their starting material utilized a partially
hydrolyzed protein slurry as a ligand to make their product, it was never hydrolyzed
to the amino acid state, and since the starting protein source also contained nonpro-
teinaceous material from plant sources,48 it would probably have been impossible
to grow crystals from such a commercial product. Nevertheless, these investigators
admitted that if amino acid chelate crystals could be grown, they could be used to
prove chelation.5
Using a proven amino chelate as a reference sample, IR spectra were generated
by analyzing commercial samples of amino acid chelates. These spectra indicated
whether the product being analyzed was chelated. It can be seen in an IR spectrum
that certain aggregates of atoms, called functional groups, can be associated with
definite characteristic absorption bands, such as the absorption of infrared radiation
over certain frequency intervals. Thus, the infrared spectrum of a mineral product
can be interpreted by the use of those known group frequencies. These group fre-
quencies occur within narrow limits. When there is interference, or perturbation, it
will cause a shift of the characteristic bond to occur. These shifts can be brought
about by (a) the electronegativity of neighboring groups or atoms; (b) the spatial
geometry of the molecule, such as the changes that occur in chelation; or (c) the
mechanical mixing of vibrational modes.49
Absorption of infrared radiation is confined largely to molecules, like amino acid
chelates, that have small energy differences between various vibrational and rota-
tional states. To absorb infrared radiation, a chelate molecule must undergo a net
change in dipole moment as a consequence of its vibrational or rotational motion.
In this way, the field of radiation interacts with the chelate molecule and causes
changes in the amplitude of one of its motions. The relative positions of the atoms
in a chelate molecule are not fixed, but instead they fluctuate constantly as a conse-
quence of a multitude of different types of vibrations and rotations about the bonds
in the molecule. These vibrations are either stretching (where there are continuous
changes in the interatomic distance along the axis of the bond between atoms) or
bending (where the angle between two bonds in the chelate molecule changes).50 As
the chelate molecule is subjected to infrared radiation, it will highlight the types of
vibration occurring in the molecule. As a consequence of the bending and stretching
that occurs in forming a chelate, the IR spectrum changes from the spectrum of the
original ligand prior to chelation to a different spectrum.
The original work of Ashmead et al. in developing infrared spectroscopy for
chelate identification employed a dispersive grating spectrophotometer. As instru-
mental technology improved, a multiplex instrument, a Fourier transform infrared
(FT-IR) spectrometer, became the standard. Fourier was the French mathematician
who resolved infinite series. Instruments that have been Fourier transformed utilized
multiple scans and averaging of outputs to obtain very precise spectra. Since electri-
cal noise is random, parts of the signal due to noise drop out of additive spectra. The
resulting spectra are true spectra. The addition of a computer allows very precise
identifications and measurements of the transformed data.51 With the change in
instrumentation from IR to FT-IR, new reference spectra standards were necessary.
Crystals derived from pure amino acid chelates were again assayed by x-ray crys-
tallography to provide standard references that allowed subsequent FT-IR reference
spectra to be created. These FT-IR spectra focused on the significant energies associ-
ated with the bonds in the amino acid chelate.52 The two bonds of particular interest
in chelation with an amino acid, such as glycine, are the oxygen (from the carboxyl
group) and the nitrogen (from the amino group) and how they bond to the metal ion.
When these two atoms bind to the mineral during chelation, their energies change
and, in the case of the carboxyl moiety, vibration is restricted. When bombarded with
IR radiation, there is increased excitement and vibration within the chelate molecule.
Then, as the molecule relaxes, the energy from these vibrations is given off. The energy
transmission is measured, and an infrared spectrum that depicts the energy transmis-
sion at various frequencies is obtained.53 Figure 5.7 shows an FT-IR spectrum for zinc
2.0 N C
O
O
C Zn
C
1.5 O 1643
O
C Evidence of ring 1395
N
Absorbance
formation Symmetric COO -

shift indicating
1.0 bonding of carboxyl
No 504 cm–1 peak
present indicating
NH2 broad band
the carboxyl is
3549.96
3087.09
stretch indicating
462.86
0.5 bound to the Zn

amine bond to Zn
0.0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers
FIGURE 5.7 FT-IR spectrum of zinc bisglycinate crystal. (Redrawn from Hartle, JW, and
Ashmead, HD, “Bonds important for amino acid chelates,” Feedstuffs, 78:16–17, September 11,
2006.)
bisglycinate.31 This procedure, which employs FT-IR, has been validated through the
Association of Analytical Chemists (AOAC) single-laboratory methodology.52
Having developed a validated method proving a mineral is truly chelated with gly-
cine, the next step was to quantify the amount of metal that was actually chelated.
Fujita et al. estimated that approximately 99.5% of the iron in the product that they ana-
lyzed was in the chelated state. From their tests, only 0.5% of the iron was in the ionic
state or, at least, not the same as the iron that was bonded to the protein/amino acids.36
Hartle et al. subsequently developed a more elegant method to quantify the per-
centage of chelation in a particular sample.31,52,54 The spectral energy change of the
nitrogen bonding from the amino moiety causes a shift in the frequency between
3,500 and 3,000 cm–1. Furthermore, when chelation occurs, diminishing freely rock-
ing carbonyl peaks can be observed at 504 cm–1 as the oxygen atom from the car-
bonyl moiety bonds with the metal ion during chelation. These changes are a result
of the carboxyl moiety joining to the metal and thus restraining the atoms directly
involved in the bond. The rocking and deformation vibrations that occur in unbound
zwitterionic amino acids are inhibited. As it relates to a chelate formed with a glycine
ligand, the peak at 504 cm–1 diminishes linearly (R2 = 0.99) as increasing glycine is
bound to the metal. In fact, this peak is dependent solely on the amount of unbound
glycine in the product. As illustrated in Figure 5.8, the peak at 504 cm–1 disappears
when 100% of the metal, in this case glycine, is chelated.14,54
Glycine
50/50 Cu Bisglycinate chelate/Glycine

Absorbance
95/5 Cu Bisglycinate chelate/

Glycine
Cu Bisglycinate chelate
3500 3000 2500 2000 1500 1000 500

Wavenumbers
FIGURE 5.8 FT-IR spectra representation of diminishing 504-cm-1 peaks in chelated

samples. (Redrawn from Hartle, JW, and Ashmead, SD, “Single laboratory validation of the
quantification of chelation in metal glycine chelates through the use of FT-IR analysis,” poster
presentation at 118th AOAC International Meeting, St. Louis, MO, September, 2004.)
As shown in Figure 5.7, other peak assignments in the spectrum denote that the
metal was chelated with the amino acid as opposed to being complexed. The absor-
bance at the 504-cm–1 absorption peak, from Figure 5.8, allows a standard curve
to be employed. For example, using the standard curve mentioned, the formula
y = 768.83x + 0.0071 can be developed. The symbol y represents the absorbance at
504 cm–1, and x represents the unbound glycine in the product. Thus, if a zinc glyci-
nate exhibited an absorbance of y = 0.1847, the determination of free glycine in the
product would be as follows55:
y = 768.83x + 0.0071
x = (y – 0.0071)/768.83
x = (0.1847 – 0.0071)/768.83
x = 0.0002310 = 0.2310 mg
If the zinc being analyzed weighed 150 mg and the sample concentration was
1.0% in KBr, the grams of unbound glycine in the sample can be calculated as55
(0.2310 mg unbound glycine/150 mg sample) × (150 mg sample/1.5 mg Zn) =

0.1540 mg unbound glycine
The purpose of this chapter is to demonstrate that there are now proven analyti-
cal procedures to validate amino acid chelation. If one does not have an elucidated
mineral product, it is difficult, if not impossible, to ascribe the observed in vitro and
in vivo results to the chelation phenomenon. If, however, the product is proven to
actually to be an amino acid chelate, then if its absorption or metabolism is differ-
ent from that of a metal ion, one can conclude that the difference is due to the metal
being chelated and not to some other phenomenon. In the subsequent chapters, all
of the amino acid chelate data generated from the numerous cited studies are from
proven metal amino acid chelates unless a particular study states the contrary.
REFERENCES
1. Beck, B, and Ashmead, SD, “An analysis of Zn, Fe, Cu, Mg, Ca, Mn and Co chelates
from seven commercial manufacturers,” unpublished research report, Snow College,
Ephraim, UT, 1990.
2. Parks, FP, and Harmston, KJ, “A field study of trace mineral complexes,” study pre-
sented to AAFCO for use as a quality control assay, Indianapolis, IN, 1995.
3. Parks, FP, and Harmston, KJ, “Judging organic trace minerals,” Feed Manage 45:35–36,
October 1994.
4. Kratzer, FH, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press)
25–27, 1986.
5. Leach, GA, and Patton, RS, “Analysis techniques for chelated minerals evaluated,”
Feedstuffs Mag 69:13–15, March 31, 1997.
6. Jeppsen, RB, “Critique of the method proposed by Graham A. Leach and Richard S.
Patton,” unpublished report, Albion Laboratories, Clearfield, UT, 1997.
7. Krebs, S, ed., Official Publication of Association of American Feed Control Officials
(Oxford, IN: AAFCO) 386, 2009.
8. Hynes, MJ, Colleran, D, Finn, J, Headon, DR, and Lyons, TP, “Quantitative chemi-
cal assessment of some mineral proteinates used in animal feeds,” J Animal Sci 68
(Suppl):408, 1997.
9. Ericson, C, and Ashmead, SD, “Characteristics of amino acid chelates in different pH
environments,” unpublished research study, Albion Laboratories, Clearfield, UT, 1999.
10. Kragten, J, Atlas of Metal-Ligand Equilibria in Aqueous Solution (Chichester, UK:
Horwood) 1978.
11. Little, P, and Ashmead, H, “Metal ionization chromatography,” unpublished research
paper, Albion Laboratories, Clearfield, UT, 1976.
12. Brown, TF, and Zeringue, LK, “Laboratory evaluations of solubility and struc-
tural integrity of complexed and chelated trace mineral supplements,” J Dairy Sci
77:181–189, 1994.
13. Ashmead, HH, “Soluble iron proteinates,” Washington, DC, U.S. Patent 4,216,144,
August 5, 1980.
14. Hartle, JW, and Ashmead, SD, “Single laboratory validation of the quantification of che-
lation in metal glycine chelates through the use of FT-IR analysis,” poster presentation
at 118th AOAC International Meeting, St. Louis, MO, September, 2004.
15. Werner, A., “Beitrag zur Kostitution Anaorganischer Verbindungen,” Z Anorg u Allem
Chem 3:267, 1893.
16. Ley, H., “Uber Innere Metall-Komplexsalze. I,” Z Elektrochem Angew Physik Chem
10:954–956, 1904.
17. Ashmead, H, and Little, P, “Electrical conductivity of amino acid chelates,” unpublished
18. Ashmead, H, and Little, P, “Albion bioelectronics system,” Proceedings of Albion
Nutrition Conference, Salt Lake City, UT, November 11, 1979.
19. James, H, “A differential pulse polarography study of zinc glycinate chelate,” unpub-
lished research report, Weber State University, Ogden, UT, 1987.
20. Cox, JA, and O’Reilly, JE, “Polarography and Voltammetry,” in Christian, GD, and
O’Reilly, JE, eds., Instrumental Analysis (Boston: Allyn and Bacon) 52–96, 1978.
21. Holwerda, RA, Albin, RC, and Madsen, FC, “Chelation effectiveness of zinc proteinates
demonstrated,” Feedstuffs 67:12–13, 23, June 19, 1995.
22. Cao, J, Henry, RP, Guo, R, Holwerda, RA, Toth, JP, Littell, RC, Miles, RD, and
Ammerman, CB, “Chemical characteristics and relative bioavailability of supplemental
organic zinc sources for poultry and ruminants,” J Anim Sci 78:2039–2054, 2000.
23. Hancock, R, “Review and comments on the Holwerda et al. paper published in Feedstuffs
Magazine June 19, 1995,” personal communication, July 2003.
24. Laurent, JP, ed., Coordination Chemistry (Oxford, UK: Pergamon Press) v21,
171–185, 1981.
25. Sternhell, S, “Nuclear magnetic resonance spectroscopy,” in Christian, GD, and
O’Reilly, JE, eds., Instrumental Analysis (Boston: Allyn and Bacon) 357–394, 1978.
26. Olsen, JI, and Schweizer, MP, “NMR analysis of Chelazome® products,” unpublished
research report, University of Utah, Salt Lake City, 1987.
27. Beck, BR, “The structure of amino acid chelates produced by Albion Laboratories,”
unpublished monograph, Snow College, Ephraim, UT, 1987.
28. Heaney, RP, Recker, RR, and Weaver, CM, “Absorbability of calcium sources: The lim-
ited role of solubility,” Calcif Tissue Int 46:300–304, 1990.
29. Wasson, JR, “Electron spin resonance spectroscopy,” in Christian, GD, and O’Reilly, JE,
eds., Instrumental Analysis (Boston: Allyn and Bacon) 396–410, 1978.
30. Rogers, K, and Landcaster, J, “Report on electron paramagnetic resonance spectra of
Albion® metal amino acid chelates,” unpublished research report, Utah State University,
Salt Lake City, 1985.
31. Hartle, JW, and Ashmead, HD, “Bonds important for amino acid chelates,” Feedstuffs
Mag 78:16–17, September 11, 2006.
32. Jeppsen, RB, and Ashmead, SD, “The need for methods capable of measuring true che-
lation,” paper presented at AAFCO Symposium, St. Louis, MO, May 19, 1997.
33. Ashmead, H, “Minerals and stress,” paper presented American Society of Bariatric
Physicians Seminar, Las Vegas, NV, September 1973.
34. Ashmead, H, Ashmead, D, and Jensen N, “Chelation does not guarantee mineral metab-
35. Layrisse, M, and Martinez-Torres, C, “Iron absorption from veal muscle,” J Am Clin
Nutr 24:531–540, 1971.
36. Fujita, T, Sato, T, Uchiyama, M, Hoshida, H, Kawabe, K, Yashiro, J, and Wakabayashi, T,
“Binding of iron ions with soybean protein,” Hygienic Chem 28:106–110, 1982.
37. Nakamoto, K, Infrared and Raman Spectra of Inorganic and Coordination Compounds
(New York: Wiley) 1978.
38. Anfinsen, CB, Edsall, JT, Richards FM, and Eisenberg, DS, eds., Advances in Protein
Chemistry Metalloproteins: Structural Aspects (San Diego, CA: Academic Press) 1991.
39. Bloss, FD, Crystallography and Crystal Chemistry (New York: Holt, Rinehart and
Winston) 1971.
40. Dickinson, RG, and Raymond, AL, “The crystal structure of hexamethylene-tetramine,”
J Am Chem Soc 45:22–29, 1923.
41. Bragg, WL, James, RW, and Bosanquet, CH, “The distribution of electrons around the
nucleus in the sodium and chlorine atoms,” Philosophical Mag 44:433–449, 1922.
42. Bragg, WL, “The structures of some crystals as indicated by their diffraction of x-ray,”
Proc R Soc London A 89:248–277, 1914.
43. Anfinson et al., Advances in Protein Chemistry Metalloproteins: Structural Aspects (San
Diego, CA: Academic Press) 296, 1991.
44. Ashmead, HH, “Preparation of pharmaceutical grade amino acid chelates,” U.S. Patent
4,830,716, May 16, 1989.
45. Ashmead, H, “Pure amino acid chelates,” U.S. Patent 4,599,152, July 8, 1986.
46. Dalley, NK, “Report on x-ray diffraction crystallography of Albion® Zinc amino acid
chelate,” unpublished research report, Brigham Young University, Provo, UT, 1991.
47. Knutson, K, “IR proofs of chelation,” unpublished research report, University of Utah,
Salt Lake City, 1987.
48. Richards, AZ, “Precipitation of metal proteinates from brines by base-acid-base hydro-
lysis,” U.S. Patent 3,775, 132, November 27, 1973.
49. Socrates, G, Infrared Characteristic Group Frequencies (Chichester, UK: Wiley) 1994.
50. Skoog, DA, Holler, FJ, and Nieman, TA, Principles of Instrumental Analysis
(Philadelphia: Harcourt Bruce College) 380–401, 1998.
51. Colthup, NB, Daly, LH, and Wiberley, SE, Introduction to Infrared and Raman
Spectroscopy (San Diego, CA: Academic Press) 1990.
52. Hartle, J, and Thompson, RC, “Development of a novel method to quantitate the amount
of unbound metal in copper glycinate solutions,” poster presentation at 121st AOAC
Annual Meeting, Minneapolis, MN, September, 2007.
54. Hartle, J, Ericson, C, and Ashmead, S, “Process for determining the percent of chelation
in a dry mixture,” U.S. Patent 7,144,737, December 5, 2007.
55. Hartle, J, “Quantitation of chelation in Albion advanced nutrition’s human products,”
unpublished research report, Albion Laboratories, Clearfield, UT, 2004.
6 Absorption of Amino
Acid Chelates from
the Alimentary Canal
The digestive system of the body starts with the mouth and ends with the anus.
Commonly termed “the digestive tract,” it is about 9 m (29.5 ft) long in humans.
Obviously, it is shorter in some animals and longer in others. Its primary purpose
is the same in all species: convert complex food products into simple nutrients or
monomers, absorb those nutrients, and expel the undigested or unabsorbed residual.
This entire process from the mouth to the anus is commonly referred to as digestion.
The initial step in providing nutrition to the body is ingestion of the nutrient-
containing food, which stimulates the production of saliva in the mouth. This mucus-
like fluid performs two major roles. First, it mixes with aliments to lubricate the dry
foods and dilute thicker ones. The second purpose of saliva is to initiate digestion
by mixing the food with salivary digestive enzymes. As the food is chewed, large
particles are reduced to smaller ones for greater enzymatic breakdown efficiencies.
For example, 1 cubic inch of food presents 6 square inches of surface space for the
salivary enzymes to attack. If by chewing, the 1-cubic inch piece of food is reduced
to ⅛-inch cubes, there will be 512 of these ⅛-inch cubes. This increases the total
surface space available for enzymatic activity from the original 6 square inches to
48 square inches. Further, to completely break down the original 1 cubic inch of
food, the salivary enzymes must penetrate ½ inch into the cubic inch on all sides. If,
however, the cubic inch of food is reduced to 512 individual ⅛-cubic inch pieces, the
enzymes need only penetrate the food 1∕ 16 of an inch on all sides to initiate digestion
on all of the food.1
One of the first enzymes to commence hydrolyzation of ingested food is a-amylase.
It is secreted into the mouth via the saliva and is most efficient in a neutral or slightly
alkaline pH. When the food bolus mixes with saliva, this enzyme immediately com-
mences digestion of the starch in the bolus into carbohydrates. The a-amylase splits
the a-1,4-glycosidic bonds of polysaccharide molecules in a random fashion, giving
rise to a mixture of glucose and maltose.2 The element, chlorine, is part of the pros-
thetic group of a-amylase. Calcium and magnesium as well as several monovalent
elements are also involved with the enzyme.3 These elements are incorporated intra-
cellularly into the enzyme prior to the enzyme being secreted into the mouth through
the saliva. The enzyme is not synthesized in the saliva directly from the nutrients in
a food bolus. Thus, if a calcium or magnesium amino acid chelate were ingested with
the food, neither mineral would have an immediate or direct effect on a-amylase
production or activity until after the minerals were absorbed.4
81
Because amino acid chelates contain no starch or sugar as part of their molecular
makeup, the a-amylase in the saliva has no hydrolytic effect on them. The chelates
ingested concurrently with the food pass through the esophagus intact and enter the
stomach as part of the food bolus.
The food bolus and the amino acid chelate travel through the esophagus as a result
of peristalsis. The churning action of the esophageal peristalsis mixes the swallowed
bolus and chelate with the secreted saliva, thus increasing the a-amylase activity.
That slightly alkaline pH, while stimulating a-amylase activity, also encourages the
ingested amino acid chelate to remain in the chelated form.5 As the bolus enters the
stomach, it has a slightly alkaline pH. In the stomach, the bolus is converted to chyme.1
The chyme forms in the cardiac region of the stomach in monogastric mammals,
where it becomes semifluid and acidic in nature. In polygastric animals, such as
mature bovines, the stomach is divided into four compartments, commonly thought
of as individual stomachs. One of the purposes of the first three stomachs is to provide
a fermentation environment to further hydrolyze complex starches from grass, hay,
and the like into molecules that can ultimately be digested in the fourth stomach. The
ruminant’s fourth stomach produces the same gastric juices as does the stomach of a
monogastric mammal and has the same basic function.6 It is from the fourth stomach
that the chyme will enter the intestine. With that in mind, this discussion focuses on
the role of the stomach in monogastric animals, principally the human. Suffice it to
say that the pH values in the three stomachs of the ruminant are also ideal to keep
the amino acid chelates from hydrolyzing as they move through the three stomachs
toward the fourth.5,7,8
The cardiac region of the stomach consists of about two-thirds of the upper region
of the stomach. It is subdivided into the upper fundus and the lower body. The most
distal portion of the stomach is the pyloric region. There are long folds on the sur-
face of the stomach called rugae. The surface of the rugae contains a layer of cells
known as the gastric mucosa. There are several different types of cells in the gas-
tric mucosa, some of which form the gastric glands. The goblet cells secrete mucus
simultaneously with the secretion of hydrochloric acid and pepsinogen by the pari-
etal cells. Hydrochloric acid, which is discussed in greater detail in this chapter,
is essential for digestion of some foods. The mucus that is secreted by the goblet
cells coats the surface of the stomach and protects it from erosion by the acid and the
activated form of pepsinogen.9
Pepsinogen is an inactive form of pepsin. By the time of its conversion to pepsin,
the stomach surface is coated by the mucus that has been secreted by the goblet
cells. If active pepsin were excreted into the stomach without this coating, it would
attack the protein molecules composing the stomach tissues and gastric glands
before even being excreted. Instead, pepsinogen is converted to pepsin after it is
mixed with hydrochloric acid in the gastric juices. The acid removes 42 amino acid
residues from the N-terminal portion of pepsinogen. Some of the peptides that are
removed from pepsinogen act as pepsin inhibitors. Once they are cleaved from the
pepsinogen, the resulting pepsin is able to commence hydrolyzation of the proteins
in the chyme by breaking their peptide bonds.10 Pepsin activity is influenced by the
intracellular incorporation of calcium and manganese into the enzyme.11 Since
the amino acid chelate does not contain peptide bonds, the pepsin has no effect on
Absorption of Amino Acid Chelates from the Alimentary Canal 83
it. A metal proteinate, however, can potentially be hydrolyzed by pepsin since, by

definition, it is composed of either amino acids plus partially hydrolyzed protein or
completely of hydrolyzed protein without individual amino acids.12
Another type of cell found in the gastric mucosa of the stomach is the argentaffin
cell, which secretes serotonin and histamine.9 Serotonin affects the release of fatty
acids from fatty tissue and aids in lipolysis.13 Histamine is a powerful vasodilator
that stimulates the increased secretion of both pepsin, in the form of pepsinogen, and
hydrochloric acid by the stomach.14 Amino acid chelates contain no lipids, so sero-
tonin has no effect on them. The histamine does not have a direct effect on amino
acid chelates, although its effect on the release of additional acid into the stomach
does.5 This effect is discussed further in this chapter.
The stomach mucosa also contains G cells, which secrete gastrin.9 Gastrin is a
hormone that works in conjunction with histamine to increase secretion of gastric
acid, among other substances. The increased secretion of acid into the stomach, as
noted, will affect the integrity of an amino acid chelate. Most G cells are located in
the pyloric-antral mucosa of the stomach.
The mixing of chyme with gastric juices in the stomach decreases the pH of
the chyme from slightly alkaline to an acidic pH. As the pH is lowered, the starch-
splitting activity of a-amylase is arrested, followed by a commencement of peptidase
activity. The chyme churns in the stomach and is acted on by the stomach enzymes
as they prepare the nutrients for absorption. It is at this point in the digestion process
that the amino acid chelate is first affected by its environment.
With the introduction of hydrochloric acid to the chyme, the ingested amino acid
chelates begin to be modified by the change in the pH environment. Whereas the
mild alkaline pH encouraged the continuation of the amino bond between the nitro-
gen in the amino group of the ligand and the metal ion, in an acid environment,
where the pH can approach 2 near the pylorus of the stomach, that bond can be
broken. The metal remains attached to the amino acid via the carboxyl bond, but
at this pH, it is not in the form of an amino acid chelate. Instead, it is probably an
amino acid complex, which is referred to as an amino acid chelate/complex to dis-
tinguish it from an amino acid complex as defined by the Association of American
Feed Control Officials (AAFCO).5,12 Depending on the amino acid ligand selected,
the strength of the amino bond can be modified.15 Thus, in a chelate having a low
stability constant, this bond will be broken at a higher acid pH, or conversely, with
a chelate with a higher stability constant, it will remain intact until the amino acid
chelate is subjected to a much lower acid pH environment.5,16 Ultimately, the amino
bond may be broken at some point in the acid environment of the stomach.
The significance of this is that when the amino bond is broken, this results in the
amino acid chelate/complex having a positive charge on the amine moiety (NH3+) at
the terminus of the ligand molecule. This allows the molecule to potentially bond to
a negative charge on an absorptive transport molecule in the stomach and initiate its
absorption from the stomach tissue by active transport, similar to the absorption of
free amino acids from the stomach.10 This same phenomenon occurs with the active
transport absorption of free amino acids. Prior to their absorption, they must be
subjected to an acid pH to add a hydrogen ion (+ charge) to their N-termini.17 They
can then be absorbed from the stomach.10 Absorption of free metal ions from the
stomach does not occur with any degree of significance, but several investigators
have reported that, like amino acids, measurable absorption of metal amino acid
chelates from the stomach does occur.18,19
In one study, conducted at Virginia Polytechnic Institute and State University,
a group of 20 weanling pigs (23 days of age) were assigned a diet containing only
17 ppm of zinc for 32 days. They were then randomly divided into four groups of
five animals each. Two groups received 15 ppm of zinc as zinc sulfate or zinc amino
acid chelate, and two groups received 45 ppm of zinc in either the sulfate or chelate
form. Chromic oxide was also included in all of the diets as a marker (0.25%). The
zinc supplements and chromic oxide were mixed with the feed provided to the pigs.
The pigs were fed daily at 0600 and 1800 hours. All of the pigs were individually
housed in stainless steel cages. This allowed for total feces and urine to be collected
for a 7-day collection period.
The zinc supplementation in the feed continued for a total of 24 days. On days
0, 3, 6, 12, and 24, one pig from each group was randomly selected and sacrificed.
The gastrointestinal tract was ligated into the stomach; proximal, medial, and distal
small intestine; cecum; and proximal and distal colon. The digesta were expressed
from each segment, followed by a thorough rinsing with saline solution. After a
48-hour drying period at 70°C, samples from each segment were wet digested and
then assayed for chromium and zinc using flame atomic absorption spectrophotom-
etry. Apparent absorption coefficients of zinc for digesta were determined by the
indirect method.
The zinc absorption was lower in all segments of the gastrointestinal tract from
pigs receiving zinc sulfate compared to those pigs fed zinc amino acid chelates
(p < 0.01).18 The apparent absorption coefficients in the stomach were negative for
zinc sulfate (–18.3%) and positive for zinc amino acid chelate (15.7%).18 This was the
first published study that demonstrated that when zinc is chelated to amino acids, it,
and presumably other metal amino acid chelates, can be absorbed directly into the
blood from the mucosal tissue of the stomach.
A second study conducted at Texas Tech University also demonstrated that the
zinc from the amino acid chelate can be absorbed into the mucosal tissue of the
stomach and subsequently transported directly to portal plasma.18 In this study,
absorption of zinc from zinc sulfate, zinc methionate, and a mixed chelate composed
of zinc glycinate and zinc methionate were compared using a 3 × 3 Latin square
design with three periods to compare the three zinc sources.
Each pig in the study was individually housed in a stainless steel cage and had a
mean weight of 21.5 ± 0.7 kg. While under anesthesia, each pig had four catheters
surgically placed in it to collect blood from the carotid artery, the portal vein, and the
mesenteric vein and to have access to the pyloric region of the stomach. Following
a 2-day recovery period during which each pig received 28 ppm of zinc daily in its
food, the pigs were fasted for 19.5 hours.
At the conclusion of the fasting period, plasma samples were collected from both
the portal vein and carotid artery and were collected again 30 minutes later to estab-
lish a baseline. A zinc infusion (230 mg Zn) from one of the three zinc sources was
placed into the pyloric region of the stomach. At 0, 15, 30, 45, 60, 90, 120, 150, and
210 minutes postdosing, blood samples were obtained from the portal vein and the
3.5 ZS ZM ZGM
3.0
2.5
2.0
mg Zn/min
1.5
1.0
0.5
0.0
–0.5
–1.0
0 30 60 90 120 150 180 210
Min
FIGURE 6.1 Net portal zinc absorption from three zinc sources in blood flow: 1.38 ± 0.23 L/
min. zs = zinc sulfate; zm = zinc methonate; zgm = zinc glycine methionine. (Redrawn from
Kim, SW, “Zinc amino acid chelates as upgraded zinc sources for monogastric animals,”
paper presented at Albion Animal Nutrition Conference, Midway, UT, 2007.)
carotid artery and assayed for zinc. Blood flow was determined by infusing para
amino hippuric acid into the blood via the mesenteric vein catheter. The blood flow
was calculated by measuring the concentrations between the portal vein and the
carotid artery.
Having determined the zinc concentration in the plasma from the portal vein
catheter and the carotid artery catheter and the ratio of blood flow, the net portal zinc
absorption could be determined. This is seen in Figure 6.1.19
The chelates of both methionine and glycine methionine mixture showed an
initial absorption peak in the portal blood at about 31 to 36 minutes postdosing.
These first peaks correspond to the absorption from the stomach into the portal
blood of zinc from the two zinc amino acid chelate sources. The zinc sulfate
source did not exhibit this first zinc absorption peak and zinc was not appreciably
absorbed from the stomach, whereas the zinc from both zinc amino acid chelate
sources was. Due to its lower stability constant,16 more of the amino bonds in the
methionine chelate were broken earlier in the stomach than were the amino bonds
in the glycine methionine chelate.5,16 Thus, more of the zinc from the methionine
chelate was absorbed from the stomach than was zinc from the glycine methio-
nine chelate. This is seen as a higher peak for the zinc methionate source between
0 and 50 minutes postdosing, compared to the zinc glycinate methionine source.
What portion of the absorbed zinc came from the methionine ligand and what
portion came from the glycine ligand in the mixed chelate source was not mea-
sured. The methionate portion was probably absorbed similarly to the methionine
chelate that was provided alone. Thus, the difference between the two chelate
sources was probably due to the glycine chelate. For this reason, the subsequent
discussion refers to the mixed ligand as glycine only and ascribes the different
results to the glycine ligand.
The second peak in Figure 6.1, which starts at approximately 100 minutes post-
dosing, corresponds to zinc absorption from the small intestine. The peak in plasma
zinc from zinc methionate occurred at about 110 minutes postdosing. The zinc peak
from zinc glycinate occurred at about 150 minutes postdosing. The difference in
time again related to stability constants of the two chelates and the pH at which the
amino bonds for each of the chelates was broken.5,15,16 The peak showing plasma
absorption of zinc from zinc sulfate also occurred at about 150 minutes postdosing,
but the height of that peak was significantly lower than either of the zinc peaks from
the two chelate sources (p < 0.05), indicating lower zinc absorption compared to
either chelate source. The secondary plasma zinc peaks from the amino acid chelate
sources and the single plasma zinc peak from the zinc sulfate all occurred in the
small intestine as the three zinc sources left the stomach via the pyloric sphincter and
emptied into the small intestine.
The amino acid chelate changes its molecular shape with changes in pH. At
pH 7, the molecule is a chelated configuration with the amino acid ligands forming
the requisite heterocyclic rings with the metal ion. As the pH rises, the amine bond
from the nitrogen to the metal is once again broken as occurred earlier in the acid
environment (Figure 6.2). These changes in pH result in a chelate/complex molecule
in which the metal continues to remain bound to the amino acid via the carboxyl
bond, but the metal-ligand molecule is no longer a true chelate.5,16
The breaking of the amino bond in amino acid chelates occurs at different acidic
and alkaline pH values depending on the amino acid employed as a ligand. If, as
noted, the amino acid ligand has a higher stability constant, as in the case of glycine
versus methionine,5,16 the acid pH has to be lower before the amino bond of the gly-
cine chelate will be broken compared to that same bond in a methionine chelate.
Thus, in Figure 6.1, more of the zinc from the methionine chelate was available for
absorption from the stomach compared to zinc from the glycine chelate because
the amino bonds on the methionine chelate were broken sooner. This is reflected in
a higher plasma zinc absorption peak from the methionine chelate in the stomach
compared to the stomach absorption of zinc from zinc glycinate absorption. Further,
because the zinc dose was finite, if more of the zinc from methionine chelate was
absorbed from the stomach, there would obviously be less zinc from the methionine
chelate available for subsequent absorption from the intestine. Thus, in the intestine,
one would predict a higher zinc absorption peak from the glycinate chelate com-
pared to the methionate chelate.
O NH2
NH2 O NH3+
O
H2C O
C
O M O M O M O
C CH2
O
H2N O O NH2 +H N O
3
pH ≥ 9 pH 7 ± 2 pH ≤ 5
FIGURE 6.2 The effect of pH changes on the isolated amino acid ligand of an amino
acid chelate. (From Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary
amino acid chelates by the utilization of ninhydrin,” AOAC poster, St. Louis, MO, 2004;
and Ericson, C, “Derivatization of ninhydrin with zinc bismethionate chelate,” unpublished
research report, Albion Laboratories, Clearfield, UT, 2009.)
The small intestine is the longest organ in the body and the major organ of diges-
tion. The reason it is called the small intestine does not relate to its length, which is
approximately 3.7 m in a human adult. It gets its name from its diameter, which
is smaller than the diameter of the large intestine. The first 20 to 30 cm (8 to 12 inches)
of the small intestine, from the pyloric sphincter past the duodenal papilla where the
common bile duct secretes bile and pancreatic juices into the intestine, is called the
duodenum. The next two-fifths of the small intestine are referred to as the jejunum.
The remainder of the small intestine, which empties into the large intestine through
the ileocecal valve, is the ileum.9
If a cross section of the small intestine were examined, one would note that the
structure consists of the mucosa, a submucosa, and two muscular layers. In addition,
where the duodenum is covered by peritoneum, there is a subserosa and serosa.20 As
the chyme leaves the stomach, it is almost completely digested and liquidified and
flows into the lumen of the duodenum. There, it is exposed to the numerous folds
of the mucosa and submucosa. Projecting from mucosa are villi. Some are actually
folds of the mucosal tissue, while others are prominences that project into the lumen
from the mucosal floor (Figure 6.3).21
The villi are designed to increase the surface space for absorption of nutrients.15
Thus, they are most dense in the duodenum and less numerous in the ileum. In the
duodenum and jejunum, the villi are long, broad, and leaf-like in shape, whereas in
the ileum they are more finger-like in shape (Figure 6.4).21
Carbohydrates, lipids, protein, and some metal ions, such as calcium and iron, are
absorbed from the duodenum and jejunum. The bile salts, vitamin B12, water, and
electrolytes are absorbed mainly from the ileum.9
Mu
Vi
Lu
Su
FIGURE 6.3 Cross section of the small intestine from a human subject. Mu = mucosal
tissue; Vi = villi; Su = submucosal tissue; Lu = lumen. (From Kessel, RG, and Kardon,
RH, Tissues and Organs: A Text-Atlas of Scanning Electron Microscopy (San Francisco:
Freeman) 171–180, 1979.)
Vi
MF
FIGURE 6.4 Villi in the jejunum of a human intestine. Vi = villus; MF = mucosal floor.
(From Kessel, RG, and Kardon, RH, Tissues and Organs: A Text-Atlas of Scanning Electron
Microscopy (San Francisco: Freeman) 171–180, 1979.)
The pylorus at the terminus of the stomach regulates the entry and quantity of
chyme entering the duodenum. As the chyme slowly crosses the pyloric valve and
passes through the small intestine, additional digestive enzymes and other digestive
aids are secreted into the lumen. Bile, which is made from acids derived from cho-
lesterol, is manufactured in the liver. The bile is stored in the gallbladder until it is
released into the duodenum via the common bile duct as a result of the stimulation
of the duodenum by the chyme.1,22 It aids in lipid digestion by promoting emulsifi-
cation and solubilization of dietary lipids. Lipids are essentially insoluble in water.
Enzymatic hydrolysis occurs only at the interface between the lipid droplet and the
water. If a lipid is highly emulsified, the lipid droplets are smaller, thus creating more
surface space for enzymatic hydrolysis to take place.22
In addition to the release of bile into the duodenum, hormone secretion stimulates
the pancreas to release digestive juices into the lumen via the common bile duct.
As the pancreas does so, it also secretes sodium bicarbonate into the duodenum.
This alkali substance neutralizes the pH of the acidic chyme and in turn activates the
alkaline-responsive digestive enzymes found in the pancreatic juice.1
When the pH of the chyme changes from acidic to basic, the metal amino acid
chelate/complex resulting from unlinking the amine moiety from the metal ion
undergoes another change. Initially, as the pH approaches neutral, the amine moiety
is reattached to the metal, ion and the amino acid chelate is re-formed.5,10 Prior to this
change in pH, while in the acid environment, the metal amino acid chelate complex
molecule was available for absorption by active transport. In the macro sense, with
the re-formation of the chelate, the resulting electrically neutral molecule can no lon-
ger be absorbed by active transport in the distal portion of the duodenum, where the
pH is approximately neutral. There may be significant absorption by diffusion, but
since the amino acid chelate does not have a positive charge on its ligands, it cannot
be attached to a transport molecule and absorbed by active transport.
As a result of intestinal motility, the amino acid chelate continues to traverse the
lumen of the small intestine in company with the chyme. As this occurs, the pH of
the chyme begins to rise as the pancreatic juices mix with it. It is at this point that
the metal ions resulting from ingestion of inorganic metal salts lose much of their
solubility characteristic and enter into chemical reactions that result in insoluble
metal compounds.23 With the formation of insoluble compounds, these inorganic
metal sources are no longer available for absorption. In the case of an amino acid
chelate, however, when the pH has risen sufficiently, the amino bonds of the che-
late are again broken, and amino acid chelate/complexes re-formed. In the alkaline
pH environment of the jejunum, the amino moieties lose their hydrogen ions, and
the amino terminus becomes NH2. These chelates/complexes cannot be attached to
transporter molecules as long as their amine moieties do not have positive charges
(NH3+) on them. Thus, to be absorbed by active transport, the chelate/complex must
be subjected to an acid environment. As this form of chelate/complex approaches the
mucosal cell membrane, the brush border microenvironment returns to an acidic pH
even though the lumen, as a whole, remains alkaline. The amine moiety is able to
pick up a hydrogen ion in that microacid environment of the brush border and again
becomes NH3+. When this occurs, the chelate/complex is able to bond to a transport
carrier molecule embedded in the mucosal cell membrane and be absorbed into the
jejunum and ileum tissues by active transport.5,24
This breaking and re-forming of the amino bond in the amino acid chelate was
illustrated in a study involving zinc bisglycinate. A solution of 2 mg Zn/mL solu-
tion was prepared and various concentrations mixed with 2 mL of a 1% solution of
dimethylsulfoxide and 20 mg/mL of ninhydrin at various pH values. Ninhydrin has
been used in colorimetric tests because of its ability to react with amines. It will turn
purple (Ruhemann’s purple) at 570 nm on the UV-Vis (ultraviolet/visible) spectrum.
It does this only when a primary or secondary amine is available for reaction with
the ninhydrin. When ninhydrin fails to react with the amine moiety of the amino
acid ligand, the solution containing the amine moiety will not turn purple, thus con-
firming that the amine moiety is bonded to the metal and not available to react with
ninhydrin.25 Figure 6.5 demonstrates that as the pH of the solution containing the zinc
bisglycinate changes, the coordinate covalent amine bonds between the zinc ion and
the amine moieties of its two ligands are broken at a pH of between 4 and 7 and then
again at a pH of 8 to 11. The amine moieties reattach themselves to the zinc ion at the
other pH ranges, and the zinc bisglycinate is re-formed.5 A similar pattern emerges
with other zinc amino acid chelates, although the precise pH points that the amino
bonds break or re-form change with different amino acid ligands.16,25 When Figure 6.5
is compared to Figure 6.1, the double peaks in the two figures are remarkably similar.
This effect of pH on the absorption of zinc from the zinc amino acid chelate
can also be seen in the study discussed next.26,27 A group of 16 adult male Sprague-
Dawley rats (170 ± 10 g bw [body weight]) were housed in individual stainless steel
cages and maintained on a commercial laboratory rat chow ad libitum prior to study
commencement. This commercial feed contained 90 mg Zn/kg as determined by
atomic absorption spectroscopy.
0.3
0.25 1 mg/ml
0.5 mg/ml
Abs @ 570 nm 0.2
0.1 mg/ml
0.15
0.1
0.05
0
2 3 4 5 6 7 8 9 10 11
pH
FIGURE 6.5 The effect of pH on ninhydrin-zinc bisglycinate chelate. (Redrawn from

Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary amino acid chelates
by the utilization of ninhydrin,” AOAC poster, St. Louis, MO, 2004.)
At study initiation, the rats were fasted for 24 hours, during which time they were
randomly assigned to one of two groups. At the conclusion of the fasting period, each
rat was anesthetized with Nembutal and laparotomy performed. Following laparotomy,
and depending on the group assignment, zinc chloride or zinc bismethionate, each
containing 20 mCi of 65Zn, was injected directly into the duodenum of each animal.28
Commencing 15 minutes following dosing, a serial blood sample was taken from
each rat by suborbital bleeding. Additional blood samples were obtained at 30, 45,
60, 120, 150, 180, and finally 240 minutes postdosing.27
Each of the 100-mL blood samples, obtained from the bleedings, was pipetted
onto stainless steel planchets, labeled, and then dried for 12 hours at 100°C. The
assays for 65Zn were conducted using a gamma-ray spectrophotometer with a 2-inch
NaI (Cl) scintillation detector and a strip chart recorder. Each sample was assayed
for 65Zn for a 5-minute period.
The total mean zinc absorption into the blood in the group of rats receiving the
amino acid-chelated source was significantly greater than was the mean uptake of
zinc from the zinc chloride source (p < 0.05). The patterns of absorption, however,
also confirmed the effect of pH on absorption of the zinc from the amino acid chelate
as seen in Figure 6.6.
Both zinc sources were injected into the duodenum, where the pH was acidic. The
absorption of the zinc from zinc chloride, which was already in solution and ionic
at the time of injection, was rapid. As intestinal peristalsis moved the zinc chloride
source further down the small intestine, the pH changed from an acid to neutral and
then to an alkaline pH. As the pH rose, the solubility of that ionic zinc from the zinc
chloride decreased, and by the time the pH reached approximately 7.5, the solubility
of the remaining unabsorbed zinc from the zinc chloride23 was less than half what
it was at pH 5 to 7. With less solubility, there was less absorption. Thus, there was
a corresponding and significant drop in the amount of 65Zn recovered in the blood
from this group of animals at about 90 minutes postdosing, followed by gradual
increases over the next 120 minutes. The subsequent gradual increase was due to not
Test Animals
Zn Methionate
(amino acid chelate)
Corrected Counts/100 ml Blood
ZN CL2
(inorganic metal)
In each group, 10 mcg zinc containing 20 microcuries

radioactive zinc isotope was given
1 2 3 4
Hours After Injection
FIGURE 6.6 Mean intestinal absorption of 65Zn from zinc amino acid chelate and 65Zn from
zinc chloride into blood as a function of time. (Redrawn from Ashmead, H, Ashmead, D,
and Jensen, N, “Chelation does not guarantee mineral metabolism,” J Appl Nutr 26:5–21,
Summer 1974.)
all of the zinc from the chloride source immediately losing its solubility characteris-
tic at the same time. Further, there was also the commutative effect of additional zinc
being added to the zinc already absorbed into the blood over time.
When the curve in Figure 6.6 for the plasma 65Zn from the amino acid chelate is
examined, the first peak also occurred in the acidic environment of the duodenum.
The absorption of the amino acid chelate was initially lower than absorption of zinc
chloride because it took time for the amine bonds of the chelate to be broken and
the chelate/complex molecule to be attached to the transport molecules for active
absorption. The zinc chloride was already ionized and was available for absorption
almost immediately. As intestinal peristalsis moved the chelate/complex zinc source
further down the small intestine, the pH changed, becoming first neutral and then
slightly alkaline. The remainder of the zinc from the chelate/complex, which was
not initially absorbed in the acid pH of the duodenum, re-formed into a chelate as
the pH of the intestine approached neutral. The flat portion of the curve suggests
this is occurring as zinc absorption slows and the resulting chelate commences to be
absorbed primarily by passive diffusion. This is reflected by the 65Zn blood plateau
in the chelate group that commenced at about 45 minutes postdosing and continued
until 120 minutes postdosing. By 120 minutes postdosing, peristalsis had moved
enough of the zinc amino acid chelate from the duodenum into the alkaline pH of
the jejunum, resulting in an increase in zinc absorption from the chelate source. In
the alkaline environment, the pH in the jejunum of the small intestine was alkaline
enough to cause the amine bonds of the chelate to be broken again. As the amine
moieties again became unattached from the zinc ion in the alkaline environment of
the lumen, NH2 molecules at the terminus of the amino moieties were created. When
the chelate/complex molecules with the NH2 moieties moved toward the mucosal
membrane of the lumen, the microacid environment of the brush border near the
surface of the mucosal cell membrane donated a hydrogen ion to each amino moiety,
which changed it from NH2 to NH3+ to recommence active transport of the che-
late. This positive-charged terminus allowed the chelate/complex to be attached to
a transport molecule and be rapidly absorbed into the mucosal cell by active trans-
port. In addition to the active transport, facilitated diffusion was concurrently oper-
ating and taking up zinc amino acid chelate and chelate/complex molecules. This is
seen as the second 65Zn peak in Figure 6.6. The cumulative effect was in operation
here, as it was with the zinc chloride, but because the 65Zn curve for the chelate was
sharper and higher, it also reflected additional zinc absorption compared to the zinc
chloride curve.
At about 240 minutes postdosing, the plasma 65Zn curves from both zinc sources
in Figure 6.6 became relatively flat, indicating that further rapid transfer of zinc into
the plasma from either source was nearly complete, and further uptake of the zinc was
being regulated. The difference between the two curves at 240 minutes postdosing,
however, was significant (p < 0.05). The significant differences between the two
curves also occurred at about 45, 75, and 150 to 240 minutes postdosing (p < 0.05).
This study examined the rates of movement of zinc into the blood following dos-
ing with either zinc chloride or zinc amino acid chelate. As will be discussed in
greater detail in a subsequent chapter, metal absorption into the mucosal cell does
not result in complete and immediate transfer of that same metal to the plasma. Some
of the metal that is absorbed into the mucosal cell is sequestered and stored in that
cell. Its subsequent transfer to the plasma is dictated by the need of the body for
that metal. Thus, more of a particular metal can be absorbed into the mucosa than
will be observed by simply looking for the immediate amount of metal in the plasma.
This is clearly seen in the discussion of the next study.
In this study, two groups, each with 16 male Sprague-Dawley rats, were admin-
istered 65Zn as either zinc chloride or zinc amino acid chelate.29 Both zinc sources
were injected directly into the duodenum as described previously. Approximately
every hour commencing at study initiation and continuing throughout the study,
two rats from each group were sacrificed and their small intestines excised. Their
duodenums were separated from the jejunums and the jejunums from the ileums.
The duodenums and jejunums were cut along the mesenteric line and subsequently
rinsed three times in saline solution to remove external digesta. Following weigh-
ing, samples of each intestinal segment were dried to a constant weight at 105°C.
Each sample was reweighed and then wet digested before being analyzed for zinc
content by atomic absorption spectroscopy. The results, which are plotted against
time, are seen in Figures 6.7 and 6.8.
In the duodenal tissue, the absorption of zinc, from either source, tended to peak
at about 6 hours. By then, intestinal peristalsis had moved most of the zinc, from
either source, into the jejunum. Initially, as shown in Figure 6.8, the zinc absorp-
tion into the jejunal tissue progressed at a similar rate for both sources of zinc. That
occurred near the distal end of the duodenum, where the pH was changing from an
7
6
Moles Zn/gm Tissue

5
4
3
2
1
0
(×10–8) 2 4 6 8
Hours
ZnCl2 Zinc amino acid chelate
FIGURE 6.7 The transfer of 65Zn from 65ZnCl2 and 65Zn amino acid chelate to the duodenal
tissue from mucosal solution. (Redrawn from Ashmead, HD, “Comparative intestinal absorp-
tion and subsequent metabolism of metal amino acid chelates and inorganic metal salts,” in
Subramanian, KS, Iyengar, GK, and Okamoto, K, eds., Biological Trace Element Research
(Washington, DC: American Chemical Society) 306–319, 1991.)
3
Moles Zn/gm Tissue
0
2 4 6 8
(×10–8)
Hours
ZnCl2 Zinc amino acid chelate
FIGURE 6.8 The transfer of 65Zn from 65ZnCl2 and 65Zn amino acid chelate to the jejunal
tissue from mucosal solution. (Redrawn from Ashmead, HD, “Comparative intestinal absorp-
tion and subsequent metabolism of metal amino acid chelates and inorganic metal salts,” in
Subramanian, KS, Iyengar, GK, and Okamoto, K, eds., Biological Trace Element Research
(Washington, DC: American Chemical Society) 306–319, 1991.)
acid to an alkaline pH. As time progressed, however, and as the two zinc sources
moved further down the jejunum, at about 4 hours postdosing the zinc from the zinc
amino acid chelate commenced to be absorbed from the distal part of the duodenum
and the proximal portion of the jejunum in much higher quantities than was the
zinc from the zinc chloride. When the pH started climbing, by the seventh hour, the
solubility of the zinc chloride had dropped, resulting in less zinc being available (and
absorbable) from that source. To the contrary, the zinc chelated to the amino acids
remained available for absorption in spite of the alkaline pH.
The study was terminated at 10 hours postdosing even though the zinc absorp-
tion from the chelated source was still climbing. From other studies, it was believed
that zinc absorption would plateau and decline only when there was no more zinc
available to be absorbed due to intestinal peristalsis or when the zinc supply was
exhausted. Until the point of zinc exhaustion occurred, the difference in absorp-
tion between the two zinc sources would only get larger as time passed. The major
absorption difference between ionic zinc and amino acid chelates in the intestine had
already occurred when the pH changed to an alkaline pH in the small intestine, caus-
ing the ionic zinc to react with anions to form insoluble zinc compounds that were no
longer available for absorption. The amino acid chelate/complex remained available
for absorption in the jejunum or ileum. It was not precipitated as was the ionic zinc.
As suggested in Figure 6.6, most of the absorption of the amino acid chelate/
complex occurred in the duodenum and the jejunum. By comparison, there was
much less absorption in the ileum. That does not mean that the ileum is incapable of
absorbing the amino acid chelate/complex. Although not as numerous as in the duo-
denum and jejunum, there are villi found in the ileum. Further, the villi of the ileum
contain a layer of absorptive cells on their surface.21,30 The lower absorptive rate
in the ileum is due to the fact that most of the nutrients that are capable of being
absorbed have already been absorbed by the time the chyme reaches the ileum.
Thus, by the time the food mass passes through the ileocecal valve at the distal
end of the ileum into the large intestine, most of the nutrients are already absorbed.
All that is primarily left is dietary fiber and water. The water is reabsorbed as the
mass moves through the large intestine. This results in a progressively more solid
mass being expelled as feces through the anus.1
If the metal amino acid chelate has not been absorbed prior to entering the large
intestine, it will probably become part of the fecal material and be eliminated.27
Very little absorption will occur in the large intestine. There are two major sources
for the amino acid chelates that are recovered in the feces. The first is, of course,
the unabsorbed portion of the ingested chelate. The other fecal source of mineral
from ingestion of the amino acid chelate will come from the absorptive cells located
on the intestinal villi. The mucosal cells are formed in the crypt of the villus. They
then migrate toward the tip of the villus, replacing the older absorptive cells as they
move toward the tip. In humans, this migration generally takes 3 to 4 days to com-
plete. Once arriving at the tip of the villus, these absorptive cells are sloughed off
and eliminated as part of the feces.31,32 Any mineral remaining in the sloughed-off
mucosal cells, including amino acid chelates, would also be eliminated in the feces.
REFERENCES
1. Guthrie, HA, Introductory Nutrition (St. Louis: Mosby) 23–34, 1989.
2. Schutte, KH, The Biology of the Trace Elements (Philadelphia: Lippincott) 17–23, 1964.
Hill) 49, 1973.
4. Ashmead, HD, and Samford, RA, “Increasing protein/energy digestion by feeding metal
amino acid chelates,” Int J Appl Res Vet Med 6:38–45, 2008.
5. Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary amino acid
chelates by the utilization of ninhydrin,” AOAC poster, St Louis, MO, September, 2004.
6. Morrison, FB, Feeds and Feeding, Abridged (Clinton, IA: Morrison) 18, 1961.
7. Miller, WJ, Dairy Cattle Feeding and Nutrition (New York: Academic Press) 12, 1979.
8. Lough, DS, Beede, DK, and Wilcox, CJ, “Lactional responses to acid in vitro ruminal
solubility of magnesium oxide or magnesium chelate,” J Dairy Sci 73:413–424, 1990.
9. Fox, SI, Human Physiology (Dubuque, IA: Wm. C. Brown) 526–559, 1990.
Hill) 630–631, 1973.
11. Schutte, KH, Biology of Trace Elements (Philadelphia: Lippincott) 14–15, 1964.
12. Krebs, S, ed., Official Publication of Association of American Feed Control Officials
Incorporated (Oxford, IN: AAFCO) 387, 2009.
Hill) 573, 1973.
14. Ibid., 670.
15. Sillen, LG, and Martell, AE, Stability Constants of Meta-Ion Complexes (London:
Chemical Society) 1964.
16. Ericson, C, “Derivatization of ninhydrin with zinc bismethionate chelate,” unpublished
17. Brody, T, Nutritional Biochemistry (San Diego, CA: Academic Press) 88–89, 1999.
18. Kornegay, ET, Swinkels, JWGM, Webb, KE, and Lindermann, MD, “Absorption of zinc
amino acid chelate and zinc sulfate during repletion of zinc depleted pigs,” in Anke,
M, Meissner, D, and Mills, CF, eds., Trace Elements in Man and Animals-TEMA 8,
(Gersdorf, Germany: Verlag Media Touristik) 398–399, 1993.
19. Kim, SW, “Zinc amino acid chelates as upgraded zinc sources for monogastric animals,”
paper presented at Albion Animal Nutrition Conference, Midway, UT, February, 2007.
20. Netter, FH, The Ciba Collection of Medical Illustrations (Summit, NJ: Ciba
Pharmaceutical) v3, 19, 1994.
21. Kessel, RG, and Kardon, RH, Tissues and Organs: A Text-Atlas of Scanning Electron
Microscopy (San Francisco: Freeman) 171–180, 1979.
Hill) 543, 1973.
23. Ashmead, HD, Graff, DJ, and Ashmead HD, Intestinal Absorption of Metal Ions and
24. Ibid., 213–219.
25. Sheng, S, Kraft, JJ, and Schuster, SM, “A specific quantitative colorimetric assay for
L-asparagine,” Anal Biochem 211:242–249, 1993.
26. Ashmead, H, Jensen, N, and Graff, D, Radioactive Study of Amino Acid Chelated Zinc,
Albion Laboratories Monograph, Albion Laboratories, Clearfield, UT, 1975.
27. Ashmead, H, Ashmead, D and Jensen, N, “Chelation does not guarantee mineral metab-
28. Antonson, D, Barak, AJ, and Vanderhoof, JA, “Determination of the site of zinc absorp-
tion in rat small intestine,” J Nutr 109:142–147, 1979.
29. Ashmead, HD, “Comparative intestinal absorption and subsequent metabolism of metal
amino acid chelates and inorganic metal salts,” in Subramanian, KS, Iyengar, GK, and
Okamoto, K, eds., Biological Trace Element Research (Washington, DC: American
Chemical Society) 306–319, 1991.
30. MacInnis, A, and Graff, DJ, “Specificity of amino acid transport in the tapeworm
Hymenolepis diminuta and its rat host,” Rice Univ Studies 62:183, 1973.
31. Guthrie, HA, Introductory Nutrition (St. Louis: Mosby) 290–293, 1989.
32. Ashmead, HD, “The absorption and metabolism of iron amino acid chelate,” Arch
Latino America Nutr Suppl 51(1):13–21, 2001.
7 The Pathways for
Absorption of an
Amino Acid Chelate
Why does chelating a mineral with amino acids cause its absorption to increase?
The answer lies in comprehending that the intestinal absorption pathway for an
amino acid chelate is different from the pathways employed for intestinal absorption
of metal ions. While greater absorption of amino acid chelates has been well doc
umented, its absorption pathway is not well elucidated. This may be due in part to
most investigators initially thinking of the amino acid chelate in terms of it being a
metallic molecule instead of considering it to be absorbed as a proteinaceous mol
ecule. Comprehension of the absorption pathway of an amino acid chelate ultimately
requires a paradigm shift in thinking.
One of the earliest theories explaining the absorption of amino acid chelates
was postulated by John Miller, who focused on the absorption of the entire che
late molecule rather than looking at the amino acid chelate as a metal ion that
was attached to the amino acid ligands.1,2 He believed that all minerals, whether
chelated or not, were absorbed through pores that were located in the mucosal
cell membrane. Miller wrote that the negative charges on each of the protein mol
ecules forming the membrane caused the pores to be negatively charged at the
openings that were exposed to the lumen. This charge tended to attract positively
charged cations to the surfaces of the pores and “produce considerable fixation
or stagnation.”1 On the other hand, by a process he did not elucidate, negatively
charged anions somehow passed by the “repelling” negative charges on the pore
surfaces and then were pushed into the cell and away from that negatively charged
surface with sufficient force to propel each of these anions into the blood “almost
as though it were lubricated.”1 While Miller did not reference any sources that led
him to formulate his theory, it appears from his writing that he relied heavily on
the cellular membrane model that in 1954 was proposed by Danielli3 and is shown
in Figure 7.1.
Miller wrote:
In order to correct this tending to poor absorption of … metals, it is necessary to com

plex or chelate the [metal] before ingesting it; for by so doing the negative charged
complexing or chelating agent neutralizes the positive charge on the metal and forms
a neutral molecule—thus permitting the physiological absorption of the metal (as a
complex or chelate) by the intestinal wall; hence the metal in combination with organic
substances passes into the blood and through it to the body tissues.1
97
Lipoid molecule
Protein molecule
Polar pore
FIGURE 7.1 The molecular structure of the plasma membrane of the cell showing a charged
pore as proposed by Danielli. (Redrawn from Giese, AC, Cell Physiology (Philadelphia:
Saunders College) 191, 1973.)
Because he believed that the presence of a charged molecule/ion reduced its absorp
tion, he claimed that the increased intestinal absorption of metal amino acid chelates
was due entirely to creating electrically neutral metallic molecules.
Miller acknowledged that amino acid chelates could probably be absorbed by
both diffusion and active transport.2 Active transport, which necessitated energy,
also required that the transport molecule somehow be attracted and bonded to
the chelate molecule. The bonding was probably a charge attraction between the
amino acid chelate and the transport molecule, which, contrary to Miller’s the
ory, suggested that the chelate molecule could potentially have a charge associ
ated with it. Without an attraction between the transport molecule and the amino
acid chelate to form a single molecule, active transport of the chelate across
the mucosal cell membrane would be impossible. Since according to Miller, the
amino acid chelate had no change on it, he could not reconcile how the chelate
molecule could be attached to a transport molecule. Thus, he tended to gravi
tate toward a passive diffusion pathway model to explain the absorption of an
amino acid chelate.2
Miller did not have access to subsequent discoveries4–6 relating to the behavior
of the amino acid chelates in different pH environments when he proposed his neu
tral chelate molecule idea. These findings have tended to outdate his early theories.
Nevertheless, his concept of passive diffusion, as one way to describe a pathway
The Pathways for Absorption of an Amino Acid Chelate 99
by which an amino acid chelate could potentially be absorbed, is probably correct.

Intact amino acid chelates are probably able to diffuse into mucosal cells as a result
of moving from a higher concentration in the lumen to a lower concentration in the
mucosal cell. The diffusion is actually based on amino acid concentrations on both
sides of the cell membrane, not metal concentrations. If an amino acid chelate is
presented to the mucosal cell as a chelate and not as a chelate/complex, the mucosal
cell tends to regard it as an amino acid or small peptide-like molecule.7 The pres
ence of the metal ion in the chelate molecule does not affect the passive diffusion
of an amino acid chelate.8 When introduced into an environment that has a pH that
is approximately neutral, the amino acid chelate retains its chelate molecular con
figuration, and thus, as Miller suggested, it is generally electrically neutral. In that
case, passive diffusion through the membrane can, and probably does, occur. Most
passive diffusion of amino acid chelates probably takes place near the common bile
duct of the small intestine where the pH is approximately 7. As previously discussed,
in an environment where the pH is 7 plus or minus approximately 2, the amino acid
chelate retains its chelate structure.4–6
Simple passive diffusion can also occur through nonlipoid sites on the cell mem
brane (i.e., pores), provided that the electrical gradient across the membrane is zero.9
This can apply to the absorption of free metal ions as well as possibly metal amino
acid chelate/complexes and amino acids/peptides, all of which carry a charge on one
of their moieties. These nonlipoid areas on the mucosal cell membrane are void of a
charge, so there is little, in an electronic sense, to impede the movement of a charged
ion or a charged amino acid chelate/complex molecule across the cell membrane.
If, however, the cation or charged molecule is attached to an anion and becomes
electrically neutral, the resulting molecule can also diffuse through the charged lipid
membrane as described.
Under normal circumstances, when amino acid chelates are ingested, the majority
of the intake is not absorbed by passive diffusion. The total concentration of amino
acids/peptides on the serosal surface of the mucosal tissue can exceed the concentra
tion of amino acid/peptides in the intestinal lumen by a factor of four.10 Since the
amino acid chelate in the lumen is considered by the body to also to be an amino
acid/peptide, its concentration in the lumen would generally be less than the con
centration of amino acids/peptides in the cytoplasm of the mucosal cell. If the metal
ion is chelated to a single amino acid ligand, where total isolation and protection of
the metal ion within the molecule does not occur,11 there may be some absorption
of the amino acid chelate based on the presence of the metal in the chelate. This
would only take place if the metal ion were partially exposed. Most of that molecular
form of an amino acid chelate would, however, be absorbed by facilitated diffusion
or active transport, as described in the following discussion. Passive diffusion of a
metal ion chelated to a single amino acid ligand could occur at nonlipoid sites or
possibly by solvent drag.
Besides passive diffusion, if it is soluble, an amino acid chelate can be absorbed
by solvent drag. Solvent drag, or convection, describes the movement of ionized min
erals or soluble amino acid chelates or amino acid chelates/complexes that are swept
along by the bulk flow of water in and out of the mucosal cell, which in turn is driven
by osmotic or hydrostatic gradients across the mucosal cell membrane.12 Because
the chelates are carried by the water molecules, this is also referred to as facilitated
diffusion. The molecular size of the cation or chelate/complex molecule determines
its rate of flow by solvent drag. Obviously, the smaller the cation or chelate/complex
molecule, the greater will be its rate of flow.8 As mentioned, solubility is a factor in
solvent drag absorption. Some amino acid chelates have greater solubility than oth
ers. If the solubility of an amino acid chelate is low, or nonexistent, very little of the
chelate molecule will be carried into the mucosal cell by solvent drag.
In the duodenum and upper jejunum, there is a rapid and voluminous flow of
water and ionized minerals in and out of the mucosal cells. As a result, there is not
a significant decrease in the mineral content in the lumen in this area of the gastro
intestinal tract due to the two-way solvent drag.13 Since amino acid chelates are
protein-based molecules, this does not appear generally to be the case when they
are taken up by solvent drag, as evidenced by decreases in the metal concentration
(from the amino acid chelate) in the lumen with corresponding increases of that same
metal in the blood.14–16 The greater retention of the amino acid chelates, if they are
taken up by solvent drag, may be due to very little reverse flow of amino acids from
mucosal cells back into the lumen.
If an amino acid chelate/complex is soluble, it could also potentially be absorbed
by diffusion as a solute in the solvent drag. Iron that has been chelated to ethylene
diaminetetraacetic acid (EDTA) is quite soluble17 and serves as an illustration of
solvent drag.18,19 In a study conducted by Rubin and Princiotto,20 they exposed young
red blood cells to a surrounding medium containing either an inorganic iron salt or
iron EDTA. The red blood cells took up the iron EDTA more efficiently than they did
the inorganic iron salt, presumably due to the greater solubility of the iron EDTA.
While these investigators did not elaborate on the exact mechanisms involved in
the absorption of the two iron sources, they concluded that the chelating of the iron
somehow enhanced its diffusion into the cells. A similar concept could be applied
to amino acid chelates.
If an amino acid chelate/complex is formed, it may be absorbed by facilitated
diffusion. When an amino acid chelate is subjected to an environment in which the
pH is approximately 5 or less, the amino acid chelate reconfigures into an amino
acid chelate/complex with a positive charge on a moiety of each of its amino acid
ligands. The positive charge allows the amino acid chelate/complex to be bonded
to a transport molecule prior to uptake by facilitated diffusion or active transport.
Since the transport molecule is reported to be a protein molecule, the bonding of
the amino acid chelate/complex to it could be accomplished at one of the protein’s
hydroxyl (OH), sulfhydryl (SH), or histidyl side chains on the transport molecule.
If the amino acid chelate relies on a transport molecule for its absorption, it is pH
dependent, similar to a metal ion. Numerous investigators have reported that ions of
calcium, copper, magnesium, iron, manganese, and zinc all exhibit higher absorp
tion rates in the acidic portion of the small intestine due to retention of their ionic
characteristic.21 As long as those metals remain soluble, they function as positively
charged cations. As such, they can be absorbed by simple passive diffusion (through
noncharged pores), solvent drag, facilitated diffusion, active transport, or exchange
diffusion.9 As these metals lose their ionic status in the alkaline pH of the proximal
area of the jejunum of the gastrointestinal tract, their solubility is also lost. They tend
to form insoluble precipitates with anionic molecules in the lumen. With the loss of
their charges on the metal ion plus a reduction in solubility, the newly formed metal
compound can be absorbed only by passive diffusion. In the case of the amino acid
chelate/complex, the same changes in pH do not have an impact on its ability to be
absorbed as dramatically, for reasons described next in the discussion.
An amino acid chelate/complex, when subjected to a high alkaline pH, unlike the
metal ion, is not attracted to negatively charged anions in the lumen. Instead, as this
chelate/complex approaches the membrane, it enters an environment where the pH
becomes acidic between the microvilli.10 This change in pH allows the amino acid
chelate/complex to reconfigure and pick up a hydrogen ion, which would result in
a positive charge on each of its ligands. Once the amino acid chelate/complex has a
positive charge on its ligand, it is able to bond to a negative charge on a transport mol
ecule and be absorbed in certain cases by facilitated diffusion or by active transport.
The transport molecule must have (1) substrate and stereo specificity, (2) competi
tion between related substances, and (3) saturability of the transport mechanism with
increasing concentration.9 There are many transport molecules available to carry
metal ions across the mucosal cell membrane. Many of them mimic the transport of an
amino acid chelate/complex, which makes it easy to confuse the absorption of a metal
ion with an amino acid chelate/complex, as can be seen by the discussion that follows.
One transport molecule used for absorption of a metal ion by facilitated diffusion
is a low molecular weight ligand known as prostaglandin.22,23 It is manufactured in the
pancreas and secreted into the intestine through the common bile duct.24 It will bond
to metal ions and create molecules that can be absorbed by facilitated diffusion.22,23
Another low molecular weight protein ligand, which has been reported to facilitate
mineral absorption in infants, is secreted into the milk produced by the mother.25,26
In the case of iron, this molecule has been identified as lactoferrin. The lactoferrin
binds to the ionic iron and in so doing creates an organic iron molecule that exhibits
greater absorption when the breast milk containing this iron complex is consumed
by the offspring.27
Powell et al. have reported that mucins, which are large glycosylated proteins,
may also facilitate the intestinal absorption of cations.28 Because of the negative
charge that is inherent on mucins, they have a high binding capacity for metal ions in
the acid pH environment of the lumen. This capacity increases with higher-valency
metals. Unfortunately, the stronger the bonding constant between the metal and
mucins, the lower will be the absorption of the metal. Thus, it appears that the role
of mucin is to capture metal ions rather than facilitate their absorption. In addressing
this issue, Powell et al. stated that they believed the mucins were more involved in
regulating mineral absorption into the mucosal cell through the strength of the bond
ing constant between the captured cation and the transport molecule than actually
facilitating transport of the mineral into the mucosal cell.28 They postulated that the
mucin would release the metal ion to a transport molecule with a higher potential
binding constant than that of the mucin.
Other investigators have isolated other ligands from the intestine that appear to
be specific to certain metal ions and will facilitate their uptake. For example, there
is a protein found in the intestinal cell membrane that complexes ionic calcium with
a formation constant that is approximately 1,000 times stronger than that of calcium
albumin complex.29 It has the capacity to bind 1 mole of calcium per mole of pro
tein.30 This calcium-binding protein is found in all parts of the small intestine, but it
is most concentrated in the duodenum, (where solubility of calcium ions is greatest),
followed by the jejunum and finally the ileum.31
In 1974, Evans and Hahn discovered that picolinic acid, a bidentate ligand found
in the small intestine, was a strong chelating agent.32 Evans and his colleagues
subsequently proposed a mineral absorption model that employed picolinic acid
as the transporter molecule for the uptake of zinc and presumably other cations.33
Like prostaglandin, picolinic acid is secreted from the common bile duct into the
small intestine. Following its chelation of free metal ions in the chyme, Evans et al.
reported that this picolinate chelate would prevent the metal ions from entering into
other chemical reactions in the lumen and subsequently carry the sequestered met
als into the mucosal cell, presumably by diffusion. According to their model, once
the picolinic acid chelate entered the cytoplasm of the absorptive cell, the picolinic
acid ligand would somehow be induced to release the metal ions to the appropriate
ligands in the cytoplasm at the basolateral plasma membrane. Figure 7.2 summarizes
the model proposed by Evans et al.32–34
While this model may enhance facilitated diffusion of mineral ions in certain
loci of the small intestine, one of the problems with the Evans et al. model, as well
as the other models described, is that when these ligands are secreted into the small
intestine via the common bile duct, they cannot generally sequester cations found in
the duodenum. Intestinal peristalsis moves chyme, including the metal ions, and the
Lumen Epithelial Cell Lamina Propria

Absorptive Cell Basolateral
Membrane Plasma
Membrane
Zn+BF ZnBF ZnBF ZnMP Zn Albumin

Capillary
BF MP Albumin
BF
ZnBF BF
Pancreas
FIGURE 7.2 A proposed model for ionic zinc absorption and transport that employs pic
olinic acid as the transporter molecule. BF = zinc-binding ligand (picolinic acid); MP =
membrane-bound zinc-binding protein. (Redrawn from Evans, GW, Grace, CI, and Votava,
HJ, “A proposed mechanism for zinc absorption in the rat,” Am J Physiol 228:501, 1975.)
secreted ligands toward the anus, not up toward the stomach. Thus, these models do
not generally explain how metal ions can be absorbed by carrier-mediated diffusion
while the cations reside in the duodenum since the common bile duct is located at the
distal portion of the duodenum.
There has to be another ligand that can bind metal ions in the duodenum. None
of the stomach secretions (including digestive enzymes, acid, etc.) qualify as this
cation-binding ligand. Thus, the ligand that binds cations in the stomach and proxi
mal duodenum has to be sourced from chyme.35
Huebers and Rummel36 recognized this problem and proposed a model for ionic
iron absorption that required several steps. They stated that to be absorbed, alimen
tary iron must be bound to ligands sourced from chyme in the gastrointestinal tract.
To accomplish this, the stomach pH liberates the iron as Fe+3. (The formation of fer
ric iron is unlikely since hydrochloric acid reduces iron to a ferrous state rather than
oxidizing it to ferric iron.) According to their model, the ferric iron is released from
its original ligands in the food to other ligands while still in the stomach. It contin
ues to be in this form when it arrives in the duodenum, where the iron complex is
either released or converted into unavailable iron hydroxide or recomplexed to still
different ligands sourced from chyme. The recomplexing of the iron to new ligands
must occur within a very few minutes of the time the iron enters the duodenum.
Otherwise, the released iron will be precipitated due to pH changes in the gastro
intestinal tract, which will render it unavailable for absorption. The portion of the
iron that is recomplexed then migrates to a binding protein located on the mucosal
cell membrane. According to Huebers and Rummel, this ligand with its recomplexed
iron is somehow induced to release the iron at the luminal side of the absorptive cell
membrane in the form of Fe+3. On its release, the metal ion is attached to an ionic
iron transport protein similar to the transport protein utilized by ionized iron for
facilitated absorption. Ultimately, it is carried into the absorptive cell.
Guthrie has proposed a much simpler model that still addresses the problem of
metal ion absorption in the duodenum.37 It utilizes free amino acids to assist in car
rier-mediated diffusion of the cations. Using a nonheme iron as an example, she
stated that, once it has been reduced to the ferrous state in the stomach or duodenum,
it can combine with a free amino acid from chyme. Once attached, the amino acid
ligand will carry the metal ion directly into the absorptive cell. Her model is sup
ported by the observation that the intestinal absorption of cations is increased when
they are ingested concurrently with protein or free amino acids.16
Guthrie’s model is extremely important to this discussion because, by implica
tion, it suggests that an ingested amino acid chelate can function similarly to her
model. Her model also indicates that it is the amino acid that carries the metal into
the mucosal cell, which Payne also asserted.7 Her model is dependent on the absorp
tion of the amino acid and not the metal ion attached to the amino acid. Thus, those
who ascribe protecting the metal ion as the major benefit arising out of chelation
tend to overlook the more significant benefit of chelating with an amino acid: that of
transporting the metal ion into the mucosal cell by its amino acid ligand.
Manis and Schachter38 seemed to confirm Guthrie’s proposal by stating that
amino acids are capable of forming complexes with metal ions in the gastrointesti
nal tract. Once formed, these complexes can function as transport molecules for the
movement of metal ions across the membrane of the mucosal cells. Following uptake
by the mucosal cell, the amino acid and metal ion complex formed in the lumen is
hydrolyzed and allows both the amino acid and the metal to enter into the intracel
lular metabolic roles reserved for them.
Each of these models focuses on capturing metal ions and facilitating their absorp
tion by the presence of the ligand. To apply the absorption of amino acid chelates to
these models, some investigators have proposed that stomach or luminal hydrolysis
of the amino acid chelate must also occur prior to the metal ion being absorbed.
If the chelate is destroyed in the gastrointestinal tract and the metal ion released,
then the cation could potentially be absorbed by any of the models described. That
assumes, of course, that an amino acid chelate must undergo luminal hydrolysis prior
to absorption.
Proponents of luminal hydrolyzation of amino acid chelates point to reports that
they claim prove that amino acid chelates cannot survive intact in the gastrointestinal
environment. Frequently, the references relied on do not report research that is spe
cific to amino acid chelates. For example, in three publications involving complexed
or chelated copper, zinc, and manganese, the metals were reported to dissociate from
their ligands under mildly acidic conditions.39–41 In conducting these three studies,
no chemical proof was provided by the investigators demonstrating that amino acid
chelates were used in these studies. The studies simply assumed that the products
being worked with were actually chelated.
To further complicate the issue, some of the investigators analyzed commercial
metal complexes side by side with the “metal chelates” and then commingled the
attributes and deficiencies of both types of molecules into a single set of criteria,
which was then used to describe amino acid-chelated minerals. For example, Brown
and Zeringue39 used the abbreviation “CCP = complexed, chelated or proteinated
minerals” to describe all of the mineral sources they analyzed without separating the
analytical results from each type of mineral source even though the three mineral
sources are chemically and structurally different from each other.42,43 To add to the
confusion, the definition of a metal proteinate was misquoted by stating that a metal
proteinate is “the product resulting from the chelation of a soluble metal salt with AA
[amino acids] or hydrolyzed protein.”39 This makes a metal proteinate sound as if it
could be the same molecule as an amino acid chelate. The actual definition of a metal
proteinate is “the product resulting from the chelation of a soluble salt with amino
acids and/or partially hydrolyzed protein” (emphasis added).42 Because, by definition,
a metal proteinate must contain partially hydrolyzed protein as part of the molecu
lar structure, its stability is significantly lower than that of an amino acid chelate as
reported by Kratzer and Vohra,44 Mellor,45 and others.16,46,47 Because it is partially
composed of peptide bonds, a metal proteinate may possibly hydrolyze in stomach
acid, but a metal proteinate is not the same molecule as an amino acid chelate.
Implying that amino acid complexes are the same molecule as amino acid che
lates is also a mistake. A metal amino acid complex is, by Association of American
Feed Control Officials (AAFCO) definition, “the product resulting from complexing
of a soluble metal salt (such as potassium or manganese) with an amino acid(s).”42
According to the AAFCO, the bond between the metal ion and the ligand is a coordi
nate bond wherein the metal ion accepts an electron pair from the amino moiety of the
amino acid ligand.43 This definition does not contemplate that the bond between the
metal ion and the amino acid forms a heterocyclic ring, and in fact AAFCO is very
specific in stating that an amino acid chelate must contain a heterocyclic ring as part
of its molecular structure.43 Consequently, an AAFCO-defined amino acid complex,
when subjected to the low-acid pH of the stomach or possibly the alkaline pH of the
majority of the small intestine, would result in the bond between the amine moi
ety of the ligand and the metal ion being broken, with the subsequent release of the
metal ion.4–6 An amino acid chelate/complex is not the same type of molecule as an
AAFCO-defined amino acid complex. Thus, the stability of a metal complex is much
lower than that of an amino acid chelate or even an amino acid chelate/complex.48
In certain pH environments, the amino acid chelate will reconfigure into a metal
amino acid chelate/complex. This creates a molecule in which the metal ion remains
attached to its amino acid ligands via the carboxyl bonds throughout the different pH
environments of the gastrointestinal tract even though the bonds between the amine
moieties and the metal ion have been broken. Thus, an amino acid chelate (amino
acid chelate/complex) is not generally hydrolyzed in the gastrointestinal tract. Its
amino acid components simply change to different molecular configurations. The
metal ion continues to remain attached to its original amino acid ligands.4–6
To illustrate the effects of the lower stability constants on the overall stability of a
zinc methionine complex, an in vitro study was designed.49 The results demonstrated
that the zinc in this complex was released from its methionine ligand in an environ
ment that was equivalent to approximately pH 2 in the lumen. For this to occur, the
zinc ion was probably attached to the methionine molecule via the nitrogen atom
in the amino moiety of the methionine (as in an AAFCO-defined complex) but not to
the carboxyl moiety. Thus, no heterocyclic ring was formed. According to the inves
tigators, had this been an in vivo study, the zinc would have probably been released
in the stomach and subsequently complexed to zinc transporter molecules designed
to carry the ionic zinc into the cell. They also suggested that if the metal were not
released by the time it reached the intestines, a zinc transporter molecule with a
higher potential binding affinity for zinc than the binding constant of the methionine
ligand would ultimately tear the metal away from that methionine ligand and form
a new zinc complex that would then be available for absorption. (This assumes, of
course, that the freed zinc ion is not first complexed by an insoluble and unavail
able ligand.) If the newly formed zinc complex/transporter molecule is taken up by
the mucosal cell, the zinc would then be released from this transporter molecule
inside the absorptive cell, attached to a cytoplasmic ligand (thionein), and form zinc
metallothionein. In the meantime, the original methionine ligand would have been
absorbed separately into the mucosal cells “by diffusion or by a carrier system and
converted to L-methionine.”49 While there is nothing wrong with this model as it
relates to the studied zinc methionine complex, the data do not apply to an amino
acid chelate.
Besides Ericson,4–6 other investigators have studied the effect of pH on amino acid
chelate hydrolysis. For example, Lough et al.50 put magnesium oxide and magnesium
amino acid chelate in gastric solutions and measured magnesium ionization from the
two sources over a 24-hour period. As shown in Figure 7.3, the magnesium ion concen
tration increased hourly in the gastric solution containing the magnesium oxide, whereas
100
90
80
MgO
Mg. ppm
70
60
50 Mg - amino acid chelate

40
30
Control
0 3 6 9 12 15 18 21 24
Time (hours)
FIGURE 7.3 In vitro magnesium ionization from magnesium oxide or magnesium amino
acid chelate in gastric solution as a function of time. (From Lough, DS, Beede, DK, and
Wilcox, CJ, “Lactational response to and in vitro ruminal solubility of magnesium oxide or
magnesium chelate,” J Dairy Sci 73:413–424, 1990.)
during that same period, there was little, or no ionization of the magnesium amino acid
chelate. The magnesium amino acid chelate was resistant to gastric hydrolysis.
Both Ericson and Lough et al. conducted their studies in vitro, as was the pre
viously described zinc methionine complex study. Other investigators have dem
onstrated a lack of gastrointestinal hydrolyzation of amino acid chelate in vivo, as
described in the following discussion. As reported previously, metal ions can be
absorbed into the mucosal cells by active transport. They must first be bonded to a
transport molecule, which then allows them to move from the lumen into the muco
sal cell against their electrochemical gradients, which are independent of water flow.
This requires energy, which is sourced from within the mucosal cell.9 The active
transport mechanism is the result of several enzymatic reactions that occur on one
side of the cell membrane, and while linked to the vectoral transport of the cation,
the transport mechanism is not chemically altered by the process.51 Amino acids and
peptides can also be absorbed by active transport using similar enzyme reactions.
When a metal ion is chelated to two or more amino acids, it also can be absorbed by
active transport, but the absorption is amino acid driven, and the metal is carried into
the cell as if it were part of the amino acid.7
If one were able to examine the membrane of the mucosal cell under a powerful
microscope, one would see enzyme systems or protein molecules that completely
traverse the membrane and connect the lumen to the cytoplasm.52 Known collec
tively as integral proteins, many of these molecules function as transport carriers
for metal ions, amino acid/peptides, or other molecules that are absorbed by energy-
dependent systems. For an amino acid chelate/complex to be absorbed by active
transport, it has to be attached to the transport molecule at a negatively charged
site on that protein transport molecule. The resultant attachment neutralizes the
positive charge on the amino acid chelate/complex and the negative charge on the
transport molecule; in so doing, it initiates the enzymatic process that is necessary
for the commencement of active transport of amino acid chelate/complex across the
cell membrane.16,53
In the case of metal ions, transport molecules in the mucosal cell membrane are
somewhat specific to the cation being transported because they have binding specifi
cations that are different for each metal ion. Some carrier proteins that are involved
in active transport of certain cations have been identified, including those for cal
cium, sodium, and postassium.54 The calcium transport carrier, for example, consists
of an oligomer with two or more large subunits (100,000 Da each). A small proteo
lipid (12,000 Da) and a phospholipid (30 moles) are associated with each subunit.
There is also glycoprotein (55,000 Da) from the glycolax in the molecule.51,55,56 This
particular transport carrier has a very high affinity for calcium and will bind approx
imately 1 mole of calcium per mole of protein.57,58 These carrier proteins do not move
across the cell membrane themselves but instead transfer the mineral ion from point
to point on the structure of the transport protein, and in so doing, the ion is carried
across the membrane from the lumen into the cytoplasm of the absorptive cell.
The active transport of free metal ions into the mucosal tissue occurs primarily
in the duodenum of the small intestine, where the luminal environment is acidic. An
acid environment not only encourages ionization of certain inorganic metallic salts
through solubilizing those salts, but also it is necessary to maintain the ionic charac
ter of the metal once in solution.59,60
The rate of absorption of metal ions can be influenced by the pH of the intestinal lumen.
The pH may affect the mucosal surfaces or the availability of the luminal contents by
altering solubility characteristics, or both effects may operate simultaneously. In gen
eral, however, the more alkaline the lumen is, the lower will be the rate of absorption.61
The absorption of an amino acid chelate does not equate to the absorption of
free metal ions.7,16 In one of the absorption studies elaborated on in the previous
chapter, the luminal uptake of a 65Zn amino acid chelate continued to rise as the
chelate descended through the lumen from the acid pH of the duodenum into the
alkaline environment of the jejunum and ileum, thus demonstrating that the amino
acid chelate was absorbed by a process that did not require the hydrolysis of the che
late. If the amino acid chelate had been hydrolyzed and its metal ion released into the
alkaline pH environment of the jejunum or ileum, the absorption of the 65Zn would
not have continued to rise when the 65Zn reached that alkaline environment. Instead,
it would have declined similarly to ionic zinc.60,61
There is a body of in vivo evidence demonstrating that amino acid chelates can be
absorbed as intact molecules. For example, heme iron from hemoglobin is reported
to be absorbed intact.62–65 The iron in the heme molecule is chelated.66 As reported by
Bezkorovainy:
Heme iron released from foods in the stomach enters the intestinal epithelial cell
as the heme moiety with the porphyrin ring intact. Within the mucosal cell, heme
is catabolized by heme oxygenase and enters the same pathways to either storage or
transport as non-heme iron.66
While the chelated heme molecule may be hydrolyzed in the mucosal cell following
its uptake into the absorptive cell, it is first absorbed through the mucosal cell mem
brane from the lumen as an intact chelated iron molecule.66 If the heme molecule
can be absorbed intact, then by following the same or a similar pathway, amino acid
chelates, which are much smaller molecules than the heme molecule, can also be
absorbed intact.
Mellander prepared a calcium amino acid/small-peptide chelate that contained
10% calcium.67 When submitted to gastric enzymes, this chelate was found to be
enzyme resistant and was not hydrolyzed. This calcium amino acid chelate was sub
sequently administered to infants and resulted in very high calcium absorption com
pared to inorganic calcium salts.68–70 Mellander concluded that the amino acid ligand
that originally chelated the calcium remained attached to the calcium ion while in
the gastrointestinal tract and acted as a physiological carrier for the calcium into the
body. He also speculated that the same consequence would have resulted if other
metals were chelated to the same small peptide/amino acids.70
There is a negative charge associated with the mucin that coats the glycolax of the
mucosal cell.10,71 In reality, however, in the microenvironment near the mucosal cell
membrane, the pH environment between the microvilli of the mucosal cell remains
acidic.10 Thus, if a neutral amino acid chelate/complex were formed in the alkaline
pH of the lumen, this complex could potentially reconfigure again as it entered the
acidic microenvironment between the microvilli and gain a positive charge on each
of its amine moieties. This would allow the metal amino acid chelate/complex to be
attracted to the mucin in preparation for uptake or be bonded to a negatively charged
transport molecule and carried into the mucosal cell.
Several studies, described previously, demonstrated that amino acid chelate can
remain intact when subjected to the gastric environment. In particular, Lough et al.
showed that the magnesium amino acid chelate did not ionize when subjected to
gastric solution.50 Their study caused an earlier in vitro study to be reexamined.
It compared the jejunal tissue absorption of magnesium from magnesium oxide to
magnesium amino acid chelate.72 The small intestines of adult Sprague-Dawley
rats were removed and their jejunums excised. Once removed, each jejunum was
severed along the mesenteric line before being cut into 10-cm segments. All of
the segments were triple washed in Ringer’s lactate bicarbonate buffer solution
to remove external contamination. Following randomization, the segments were
divided into three groups: control, magnesium oxide, and magnesium amino acid
chelate. Each segment was incubated in a Ringer’s lactate bicarbonate buffer solu
tion through which a mixture of 95% oxygen and 5% carbon dioxide gas was bub
bled. Gastric solutions each containing 50 mg of magnesium, as either the oxide
or amino acid chelate sources, were added to the incubation solutions. In each
case, the magnesium sources were allowed to solubilize in the solutions prior to
the jejunal segments being placed in them. Based on the work of Lough et al.,50
it was assumed that the magnesium oxide ionized as this magnesium source went
into solution, whereas the magnesium amino acid chelate, while in solution, did not
result in ionization of its magnesium. Each of the jejunal segments was immersed
in one of the two magnesium-containing solutions or in the control solution for
TABLE 7.1
Magnesium Absorption from Magnesium
Oxide or Magnesium Amino Acid
Chelate into Rat Jejunal Tissue
Control MgO Mg AAC
7b 23b 94a
Source: From Ashmead, H, Tissue transportation of

organic trace minerals, J Appl Nutr 22: 42–51,
Spring/Summer, 1970.
Note: Data expressed as parts per million magnesium
per milligram of wet tissue.
a,b p < 0.05.
60 seconds before being removed, triple washed, dried, weighed, and assayed for
absorbed magnesium by atomic absorption spectroscopy. Table 7.1 summarizes
the results.
The data in Table 7.1 demonstrated that there were clearly two different absorption
pathways operating in this study. If the magnesium from the amino acid chelate had
hydrolyzed in the gastric solution and created magnesium ions, the amount of mag
nesium taken up by the jejunal tissue would have been similar to the amount of mag
nesium absorbed from the ionized magnesium oxide source. Because there was a
significant difference in the uptake of the two magnesium sources, it was apparent
that the ionized source of magnesium (magnesium oxide) was absorbed via a differ
ent pathway than was the amino acid-chelated source. Not only was the uptake of
the magnesium amino acid chelate more rapid, but also the quantity absorbed was
significantly greater. These data also confirmed Tansy’s data; he reported that more
than one pathway for intestinal absorption of magnesium exists.73
One of the pathways discussed in the previous study is actually an amino acid
absorption pathway and does not respond to the presence of the magnesium in the
chelate. Given the significant increase of magnesium in the jejunal tissue following
exposure to the magnesium amino acid chelate, it is obvious that at least some, and
perhaps all, of the absorption of this magnesium source were by either active trans
port or facilitated diffusion. Since the magnesium in the molecule was unreactive,7,11
the active transport and/or facilitated diffusion had to be amino acid based.
Using similar logic, Bovell-Benjamin et al. designed a study in human subjects to
ascertain whether iron from chelated ferrous bisglycinate and iron from ferrous sul
fate shared the same absorption pathways.74 Ten adult males were selected for their
study. Following overnight fasting, a blood sample was obtained from each indi
vidual and analyzed. The mean serum ferritin concentration of these volunteers was
92.4 mg/L, and their hemoglobin concentrations ranged between 128 and 160 g/L,
indicating that none of the subjects was iron deficient. After providing this initial
blood sample, each study participant consumed 55 kBq 59Fe sulfate and 111 kBq
chelated 55Fe bisglycinate mixed together in a porridge made from maize fortified
with 0.011 mg of iron (iron sulfate) per gram of porridge on a dry weight basis and
which was equilibrated overnight. Prior to adding the chelated iron bisglycinate to
the porridge, the chelate was analyzed by FT-IR, and that iron was proven to be che
lated to glycine molecules.75 A second blood sample was drawn from each subject
14 days later, and the erythrocytes of each individual were analyzed for the two iron
sources based on their isotopes. The authors reported the following:
It can be stated with confidence that the process of absorption of iron from ferrous
bisglycinate in whole-maize meal porridge differs from that of the nonheme iron pool
and that its absorption is most probably higher because it is protected from the inhibi
tory factors present in the porridge.74
In elaborating on this study at a nutrition conference, Lindsey Allen, one of the

investigators, added:
When the FeSO4 and bisglycinate were mixed in the same meal, the ferrous bisgly
cinate did not affect the absorption of iron from the sulfate; absorption from FeSO4
was 1.59% in Experiment 1 and 1.34% in Experiment 2. Moreover, the absorption of
the Fe-55 (from the bisglycinate) was still much higher (6.95% in Experiment 1 and
8.69% in Experiment 2). This is a very important observation because it means that
there was no exchange of label between the FeSO4 and the bisglycinate in the intestinal
pool or before entering the mucosal cell. If the FeSO4 and the Ferrochel [iron bisgly
cinate] iron exchanged iron between themselves, the same proportion of label would
be absorbed from both compounds. However, the [bisglycinate] iron is consistently
absorbed about 5.3 times more than the FeSO4 and this is not modified by the simul
taneous mixing with FeSO4. The [bisglycinate] iron is probably entering the mucosal
cell as the chelate.76
These investigators concluded that the chelate did not dissociate in the gastro
intestinal tract because (1) as earlier reported by Jeppsen,11 the chelated iron was
protected, and (2) there was no exchange between the labeled irons because the
amino acid chelate did not hydrolyze in the gastrointestinal tract. The ferrous sulfate
required ionization before its iron could be absorbed. If iron amino acid chelate had
hydrolyzed prior to being absorbed, that hydrolyzation would have released the ionic
iron from the chelate. This iron would have been identical to the ionic iron from the
ferrous sulfate, and uptake would have been similar. The lack of exchange between
the labeled irons meant that the amino acid chelate did not hydrolyze into free amino
acids and ionic iron in the gastrointestinal tract prior to absorption. Because iron
absorption was different for the two iron sources that were provided in the same
meal, it was concluded that the amino acid chelate was absorbed into the mucosal
tissue via a pathway that was different from the pathway employed for uptake of the
ionic iron from the ferrous sulfate.
Olivares et al. attempted to repeat this study by preparing an iron bisglycinate
product in their laboratory.77 Unfortunately, the iron employed in their study was
not analyzed and proven to be chelated to the glycine prior to its administration.
It was simply assumed to be a chelate. The iron product they had created did not
behave similarly to the product tested by Bovell-Benjamin et al.75,76 Instead of the
iron being “protected in the gastrointestinal tract,” as Olivares et al. had expected,
they found that various foods, such as milk, suppressed the absorption of the iron
from the glycinate product, while other substances, such as ascorbic acid, enhanced
the absorption of iron. In their study, their iron glycinate behaved as if it were an
amino acid complex that hydrolyzed following ingestion and acted more like an ionic
iron salt than an amino acid chelate.
This study is mentioned simply to underscore the importance of relying on absorp
tion data from a product that has been analyzed and proven to be an amino acid che
late rather than simply assuming the product is a chelate. Without proof, an assumed
chelate could actually be a metal proteinate, metal amino acid complex, or some
thing else. As noted, a metal proteinate, an amino acid complex, and an amino acid
chelate each have different chemical structures with varying characteristics even
though the atoms composing the three molecules may be identical. Stability con
stants change, molecular size of the molecules changes, and so on. If the product
being examined exhibits significant ionization in the lumen prior to absorption of
the metal, it is unlikely that the product is an amino acid chelate. It may have been an
amino acid complex or a metal proteinate, both of which may be absorbed differently
from an amino acid chelate.
An amino acid chelate is not absorbed via a metal ion absorption pathway. An
amino acid chelate/complex remains intact throughout the gastrointestinal environ
ment, resulting in the metal ion being unreactive. The only moiety in the molecule
that is reactive is the amino acid ligand, which is capable of binding to the transport
molecule in the mucosal cell membrane and initiating facilitated diffusion or active
transport of the entire amino acid chelate molecule into the mucosal cell.
There is no disputing that several pathways have been proposed to explain the
absorption of complexed or certain types of chelated minerals. Some of the models
even have sufficient data to support that such a pathway may actually exist in the
lumen to facilitate their absorption, but none of these models focuses on the absorp
tion of a proven amino acid chelate. Each of the earlier proposed absorption models
requires that the metal ion be released from the original amino acid ligand and be
attached to some other type of transport molecule found in the lumen so that the
metal in the chelate can be absorbed like any other metal ion. A model that accu
rately describes the absorption of an amino acid chelate cannot rely on a metal ion
absorption model to describe the absorption of an amino acid chelate.
Chelating a metal ion with amino acids results in greater uptake of the metal com
pared to the absorption of an ion. If, as proposed by some investigators, a metal from
an amino acid chelate and a salt both share the same absorption pathway, the amount
of uptake of the two sources should be the same. Since the uptake is significantly
different when the absorption of the two sources is compared under controlled condi
tions, this strongly suggests that the amino acid chelate and metal ion do not share
the same absorption pathway.
From these studies, it is obvious that the argument that intact chelate absorption is
impossible is without merit. Several investigators have shown that not only is it possible
for intestinal absorption of intact amino acid chelate/complexes to occur, but also it can
be an everyday occurrence depending on the type of food (source of minerals) ingested.
REFERENCES
1. Miller, JJ, Chelation, Complexing, and Key Minerals, Miller Pharmacal Monograph,
Miller Pharmacal Company, Chicago, IL, 1967.
2. Miller, JJ, Absorption from the Intestinal Tract, Miller Pharmacal Monograph, Miller
Pharmacal Company, Chicago, IL, 1966.
3. Giese, AC, Cell Physiology (Philadelphia: Saunders College) 191, 1973.
4. Ericson, C, and Ashmead, SD, “Characteristics of amino acid chelates in different pH
environments,” unpublished research study, Albion Laboratories, Clearfield, UT, 1999.
5. Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary amino acid che
lates by the utilization of ninhydrin,” poster presentation at American Chemical Society
Annual Meeting, St. Louis, MO, September 2004.
research study, Albion Laboratories, Clearfield, UT, 2009.
7. Payne, JW, “Transport and hydrolysis of peptides by microorganisms,” in Elliott, K, and
O’Connor, M, eds., Peptide Transport and Hydrolysis (Amsterdam: Elsevier) 320, 1977.
8. Fisher, RB, and Gardner, MLG, “A kinetic approach to the study of absorption of solutes
by isolated perfused small intestine,” J Physiol, 241:211–234, 1974.
9. Hendrix, TR, “The absorptive function of the alimentary canal,” in Mountcastle, VB,
ed., Medical Physiology (St. Louis, MO: Mosby) v2, 181, 1976.
10. de Robertis, EDP, Saez, FA, and de Robertis, EMF, Cell Biology (Philadelphia: Saunders)
158–165, 1975.
11. Jeppsen, RB, “Biochemistry and physiology of Albion® metal amino acid chelates as
proofs of chelation,” Proceedings of International Conference on Human Nutrition, Salt
Lake City, UT, January 21, 1995.
12. Fordtran, JS, and Dietschy, JM, “Water and electrolyte movement in the intestine,”
Gastroenterology 50:263, 1966.
13. Visscher, MB, Fetcher, ES, Carr, CW, Gregor, HP, Bushey, MS, and Barker, DE,
“Isotopic tracer studies on the movement of water and ions between intestinal lumen
and blood,” Am J Physiol, 142:550–575, 1944.
14. Ashmead, H, Jensen, N, and Graff, D, Radioactive Study of Amino Acid Chelated Zinc,
Albion Monograph, Albion Laboratories, Clearfield, UT, 1975.
15. Ashmead, H, Ashmead, D, and Jensen, N, “Chelation does not guarantee mineral metab
amino acid chelates and inorganic metal salts,” in Subramanian, KD, Iyengar, GK,
and Okamoto, K, eds., Biological Trace Element Research (Washington, DC: America
17. Dow Chemical Company, Keys to Chelation (Midland, MI: Dow Chemical Company)
1980.
18. Princiotto, JV, Zapolski, EJ, Bagley, H, Laskey, A, Morgan, R, and Rubin, M, “Absorption
of oral chelated iron,” Biochem Med 3:289–297, 1970.
19. Rubin, M, Houlihan, J, and Princiotto, JV, “Chelation and iron metabolism I: Relative iron
binding of chelating agents and siderophilin in serum,” Proc Soc Exp Biol Med 103:663,
1960.
20. Rubin, M, and Princiotto, JV, “Synthetic amino acid chelating agents and iron metabo
lism,” Ann NY Acad Sci 88:450, 1960.
Chelates (Springfield, IL: Thomas) 1985.
22. Evans, GW, and Johnson, PE, “Defective prostaglandin synthesis in acrodermatis
enteropathica,” Lancet 1:52, 1977.
23. Song, MK, and Adham, NF, “Role of prostaglandin E2 in zinc absorption in the rat,” Am
J Physiol 234:E99, 1978.
24. Hahn, C, and Evans, GW, “Identification of a low molecular weight 65Zn complex in the
rat intestine,” Proc Soc Exp Biol Med 144:793, 1973.
25. Schneeman, BD, Lonnerdal, B, Keen, C, and Hurley, LS, “Zinc and copper in rat bile
and pancreatic fluid: Effect of surgery,” J Nutr 113:1165–1168, 1983.
26. Lonnerdal, B, Stanislowski, AG, and Hurley, LS, “Isolation of a low molecular weight
zinc binding from human milk,” Inorganic Biochem 12:71–78, 1980.
27. Hutchens, TW, Henry, JF, and Yip, TT, “Origin of intact lactoferrin and its DNA binding
fragments found in the urine of human milk-fed infants: Evaluation by stable isotopic
enrichments,” Pediatr Res 29: 243–250, 1991.
28. Powell, JJ, Jugdaohsingh, R, and Thompson, RPH, “The regulation of mineral absorp
tion in the gastrointestinal tract,” Proc Nutr Soc 58:147–153, 1999.
29. Weir, E, and Hastings, A, “The ionization constants of calcium proteinate determined by
the solubility of calcium carbonate,” J Biol Chem 114:397, 1936.
30. Wasserman, R, Corradino, RA, and Taylor, AN, “Vitamin D-dependent calcium binding
protein: Purification and some properties,” J Biol Chem 243:3978–3986, 1968.
31. Taylor, A, and Wasserman, R, “Vitamin D3 induced calcium binding protein: Partial
purification, electrophonic visualization and tissue distribution,” Arch Biochem Biophys
119:536, 1967.
32. Evans, GW, and Hahn, CJ, “Copper-and zinc-binding components in rat intestine,” Adv
Exp Med Biol 48:285, 1974.
33. Evans, GW, Grace, CI, and Votava, HJ, “A proposed mechanism for zinc absorption in
the rat,” Am J Physiol 228:501, 1975.
34. Evans, GW, “Zinc absorption and transport,” in Prasad, AS, and Oberleas, D, eds., Trace
Elements in Human Health and Disease (New York: Academic Press) v1, 181–187, 1976.
35. Lichten, LA, and Cousins, RJ, “Mammalian zinc transporters, nutritional and physi
ologic regulation,” Annu Rev Nutr 29: 153–176, 2009.
36. Huebers, H, and Rummel, W, “Iron binding proteins: Mediators in iron absorption,”
in Kramer, M, and Lautenbach, F, eds., Intestinal Permeation (Amsterdam: Excerpta
Medica) 377–380, 1977.
37. Guthrie, HA, Introductory Nutrition (St. Louis, MO: Times Mirror/Mosby College) 293,
1989.
38. Manis, J, and Schachter, D, “Fe59-amino acid complexes: Are they intermediates in Fe59
absorption across intestinal mucosa?” Proc Exp Biol Med 119:1185–1187, 1965.
39. Brown, TF, and Zeringue, LK, “Laboratory evaluations of solubility and structural integ
rity of complexed and chelated trace mineral supplements,” J Dairy Sci 77:181–189,
1994.
40. Cao, J, Henry PR, Guo, R, Holwerda, RA, Toth, JP, Littell, RC, Miles, RD, and
Aminerman, CB, “Chemical characteristics and relative bioavailability of supplemental
organic zinc sources for poultry and ruminants,” J Animal Sci 78:2039–2054, 2000.
41. Guo, R, Henry, PR, Holwerda, RA, Littell, RC, Miles, RD, and Aminerman, CB,
“Chemical characteristics and relative bioavailability of supplemental organic copper
sources for poultry,” J Animal Sci 79:1132–1141, 2001.
42. Krebs, S, ed., Official Publication (Oxford, IN: Association of American Feed Control
Officials) 310–311, 2008.
43. Ibid., 361.
44. Kratzer, FH, and Voha, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 25, 1986.
45. Mellor, DP, “Background and fundamental concepts,” in Dwyer, FP, and Mellor, DP,
eds., Chelating Agents and Metal Chelates (New York: Academic Press) 45, 1964.
46. Bell, CF, Metal Chelation Principles and Applications (Oxford, UK: Clarendon Press)
1977.
47. Martell, AF, “The relationship of chemical structure to metal-binding action,” in Seven,
MJ, ed., Metal Binding in Medicine (Philadelphia: Lippincott) 1–18, 1960.
48. Sillen, LG, and Martell, AE, Stability Constants of Metal-Ion Complexes (London:
49. Richards, JD, Mechanisms of Mineral Absorption, Novus Monograph, NOVUS, St. Louis,
MO, 2006.
50. Lough, DS, Beede, DK, and Wilcox, CJ, “Lactational response to and in vitro ruminal
solubility of magnesium oxide or magnesium chelate,” J Dairy Sci 73:413–424, 1990.
51. Hobbs, A, and Albers, W, “The structure of proteins involved in active membrane trans
port,” Annu Rev Biophys Bioeng 9:259, 1980.
52. Singer, S, and Nicolson, G, “The fluid mosaic model of the structure of cell membranes,”
Science 175:720, 1972.
53. Kramer, P, Plant and Soil Water Relationships: A Modern Synthesis (New York: McGraw
Hill) 220, 1969.
54. Giese, AC, Cell Physiology (Philadelphia: Saunders) 401, 1973.
55. Knowles, A, and Racker, E, “Properties of a reconstituted calcium pump,” J Biol Chem
250:3538, 1975.
56. Miyanmoto, H, and Kasai, M, “Asymmetric distribution of calcium binding sites of
sarcoplasmic reticulum,” J Biochem 85:765, 1979.
57. Ostwald, T, and Machennan, D, “Isolation of a high affinity calcium binding protein
from sarcoplasmic reticulum,” J Biol Chem 249:974, 1974.
58. Memmler, R, and Wood, D, Structure and Function of the Human Body (Philadelphia:
Lippincott) 142, 1977.
59. Code, CF, Handbook of Physiology (Baltimore: William & Wilkins) v3, 1125, 1968.
60. Waldron-Edward, D, “Effects of pH and counter-ion on absorption of metal ions,” in
Skoryna, S, and Waldron-Edward, D, eds., Intestinal Absorption of Metal Ions, Trace
Elements and Radionuclides (Oxford, UK: Pergamon Press) 381, 1971.
61. Wheby, N., Suttle, GE, and Ford, KT, “Intestinal absorption of hemoglobin iron,”
Gastroenterology 58:647–654, 1970.
62. Conrad, M., Benjamin, BI, Williams, HL, and Foy, AL, “Human absorption of hemoglo
bin iron,” Gastroenterology 53:5–10, 1967.
63. Brown, E., Hwang, YF, Nicol, S, and Ternberg, J, “Absorption of radioiron-labeled
hemoglobin by dogs,” J Lab Clin Med 72:58–64, 1968.
64. Weintraub, L, Weinstein, MB, Huser, HJ, and Rafal, S, “Absorption of hemoglobin iron:
The role of a heme-splitting substance in the intestinal mucosa,” J Clin Invest 47:531–539,
1968.
65. Cammack, R., Wigglesworth, JM, and Baum, H, “Iron-dependent enzymes in mam
malian systems,” in Ponka, P, Schulman, HM, and Woodworth, RC, eds., Iron Transport
and Storage (Boca Raton, FL: CRC Press) 20, 1990.
66. Bezkorovainy, A, Biochemistry of Nonheme Iron (New York: Plenum Press) 50, 1980.
67. Mellander, O, “The nutritional significance of some peptides,” Voeding 16:336, 1955.
68. Mellander, O, “Phosphorylated peptides Ca absorption in infants,” Acta Soc Med Upsal
55:247, 1950.
69. Mellander, O, “Protein quality,” Nutr Rev 13:161, 1955.
70. Mellander, O, and Folsch, G, “Enzyme resistant and metal binding of phosphory
lated peptides,” in Bigwood, E, ed., International Encyclopedia of Food and Nutrition
(Oxford, UK: Pergamon Press), VII, 569, 1972.
71. Pollock, EG, “Fine structural analysis of animal cell surfaces: Membrane and cell sur
face topography,” Am Zool 18:25–69, 1978.
72. Ashmead, H, “Tissue transportation of organic trace minerals,” J Appl Nutr 22: 42–51,
Spring/Summer, 1970.
73. Tasny, MF, “Intestinal absorption of magnesium,” in Skoryna, SC, and Waldron-Edward, D,
eds., Intestinal Absorption of Metal Ions, Trace Elements and Radionuclides (Oxford,
UK: Pergamon Press) 193–210, 1971.
74. Bovell-Benjamin, AC, Viteri, FE, and Allen, LH, “Iron absorption from ferrous bisgly
cineate and ferric trisglycinate in whole maize is regulated by iron status,” Am J Clin
Nutr 71:1563–1569, 2000.
75. Hartle, JW, and Ashmead, SD, “Single laboratory validation of the quantification of che
at 118th AOAC International Meeting, St. Louis, MO, September, 2004.
76. Allen, LH, “Properties of iron amino acid chelates as iron fortificants for maize,” Proc
Albion Int Conf Human Nutr, Salt Lake City, UT, 96–108, 1998.
77. Olivares, M, Pizarrom, F, Pineda, O, Name, JJ, Hertrampf, E, and Walter, T, “Milk inhib
its and ascorbic acid favors ferrous bis-glycine chelate bioavailability in humans,” J Nutr
127:1407–1411, 1997.
8 The Absorption of
Amino Acid Chelates
by Active Transport
To understand how an amino acid chelate is absorbed into the mucosal cell from
the lumen, comprehension of the structure of the small intestine is important. If
one were to examine its morphology, one would see that the luminal surface tissue,
known as the mucosal tissue, is folded into the villi, which extend into the lumen.
Depending on the locus in the small intestine, these villi can be continuous folds or,
in other areas of the intestine, they become more pronounced and resemble individ-
ual finger-like or leaf-shaped projections.1 Each villus greatly increases the effective
absorption area of the small intestine as well as the secretory surface of the mucosa.
A single layer of mucosal cells, also known as absorptive cells, lines each villus,
including the remainder of the mucosal tissue2 (Figure 8.1).
Mucosal cells have limited life spans that range from 2 to 4 days depending on
the species studied. These cells are formed in the crypts of Lieberkühn located in the
mucosal tissue at the base of the intestinal villus, where they commence life as undif-
ferentiated cells and then over the next few days migrate outward onto the mucosal
floor as well as upward to the tip of the villus.3 As they move out of the crypts of
Lieberkühn onto the villus, they develop brush border enzymes that are character-
istic of mature mucosal cells. In the process of migrating to the tip of each vil-
lus, these absorptive cells are continuously exfoliated and replaced by new mucosal
cells, which are being pushed up from the base of the villus.4 When an absorptive
cell is shed from the villus, the nutrients, which have been taken up by the cell but
not yet transferred to the plasma, are lost along with the exfoliated cell.4
The layer of mucosal cells is fused together onto the villus at approximately 0.1 to
0.2 mm below the lateral surface of the cells. Their attachment to the villus forms a
tight junction at the luminal ends of the mucosal cells that obliterates the intracel-
lular spaces between them.3 Below the tight junction, the mucosal cells are separated
by a distance of approximately 10 to 30 nm. The space contains water as well as
desmosomes, which consist of thickened plasma membrane and fibrils that lead from
the thickened membrane into the cytoplasm of each neighboring absorptive cell. The
desmosomes weld the mucosal cells together into what appears to be a single layer of
absorptive cells that are interconnected.3 This layer of mucosal cells is therefore able
to function as a unit instead of being millions of independent cells.3
At the basement lamina, where the mucosal cells are attached to the villus, half
desmosomes extend from the basement membrane of the mucosal membrane of
the villus into the lamina of the absorptive cell. These half desmosomes serve two
117
Microvilli Glycocalyx
Tight Junction
Desmosome
Cell Membrane
Rough Endoplasmic
Reticulum
Ribosome
Golgi Actin Absorptive Goblet

Sacculae Filaments Epithelial Cell
Nucleus Cell
Cell
Membrane
Mitochondrion
Central
Lacteal
Terminal Web Artery
Myosin Vein
Filaments
FIGURE 8.1 The structure of a mucosal cell. (From Ashmead, HD, Graff, DJ, and
Ashmead, HH, Intestinal Absorption of Metal Ions and Chelates (Springfield, IL: Thomas)
30–54, 1985.)
functions. First, they attach the absorptive cell to the mucosal membrane of the villus.
Second, they facilitate the rapid transport of nutrients from the absorptive cell into
the blood via the capillary system of the villus.5 Figure 8.2 illustrates the structure.5
On the luminal end of the mucosal cell, there is a striated apical border that proj-
ects into the lumen to form microvilli. These microvilli are about 0.6 to 0.8 mm long
and 0.01 mm in diameter.5 It has been estimated that in the human intestine there are
between 1,700 and 3,000 microvilli per absorptive cell.3,5 This effectively increases
the luminal surface area available for absorption of nutrients by 15- to 40-fold.3 In
total, the absorptive surface created by the villi and microvilli is increased about
600 times more than the surface space of a smooth tube of similar dimension to the
diameter and length of the small intestine.2
As also seen in Figure 8.1, the terminal web of the mucosal cell is located just
below the microvilli in the cytoplasm. The microvilli are formed from cytoplasm
with a core of microfilaments that are composed of protein similar to the contractile
protein actin (the filaments in muscle tissue).1 These microfilaments are responsible
for maintaining the shape of the microvilli and descend down the microvilli to a
layer of perpendicular fibers called the terminal web. The fibers from the microvilli
and the terminal web interlink, or mesh, similar to the fibers in felt.6,7 Thus, the ter-
minal web anchors the actin filaments of the microvilli.
Below the terminal web, the structure of the mucosal cell is relatively similar to
the structure of other cells. The nucleus is located deep in the cytoplasm toward the
basal third of the cell. Above it are the mitochondria, which are necessary to supply
large quantities of energy for the specialized operations of the cell, including the
active transport of nutrients from the lumen into the cell. There are also endoplasmic
The Absorption of Amino Acid Chelates by Active Transport 119
Fuzzy coat/Glycocalyx
(glycoprotein)
Microvilli
Terminal web Zonula occludens
Zonula adherens
Desmosome
(macula adherens)
Gap junction
(nexus)
Desmosome
Intracellular canaliculus
Zonula occludens
Basement lamina
FIGURE 8.2 Diagram of a mucosal cell showing its attachment to the mucosal tissue and
another mucosal cell. (Redrawn from de Robertis, EDP, Saez, FA, and de Robertis, EMF, Cell
Biology (Philadelphia: Saunders) 159, 1975.)
reticulum that assemble small proteins from absorbed amino acids and Golgi sac-
cules that attach carbohydrate side chains to proteins to create digestive and trans-
port enzymes and the like.8
Many of these digestive enzymes recovered in the chyme or embedded within the
mucosal cell membranes are synthesized in the cytoplasm before being processed
and packaged within the Golgi apparatus of the mucosal cell prior to sending them
to their final destination. Some of the proteinaceous macromolecules coming from
the Golgi apparatus are inactive digestive enzymes that, when secreted from the
mucosal cells into the lumen as a result of the chyme being propelled through
the small intestine by the muscular contractions in the intestine, become activated
by the luminal environment.4 Hendrix has suggested that the spatial arrangement
of the microvilli is essential to enable the cell to secrete these digestive enzymes
into the lumen and to provide transport sites for the active uptake of nutrients.3 The
TABLE 8.1
Brush Border Enzymes Located in the Cell Membrane of Microvilli
in the Small Intestine
Category Enzyme Comments
Disaccharidase Sucrase Digests sucrose to glucose and fructose; deficiency produces
gastrointestinal disturbances
Maltase Digests maltose to glucose
Lactase Digests lactose to glucose and galactose; deficiency produces
gastrointestinal disturbances (lactose intolerance)
Peptidase Aminopeptidase Produces free amino acids, dipeptides, and tripeptides
Enterokinase Activates trypsin (and indirectly other pancreatic juice
enzymes); deficiency results in protein malnutrition
Phosphatase Ca++, Needed for absorption of dietary calcium; enzyme activity
Mg++-ATPase regulated by vitamin D
Alkaline Removes phosphate groups from organic molecules; enzyme
phosphatase activity may be regulated by vitamin D
Source: Data from Fox, SI, Human Physiology (Dubuque, IA: Brown) 534, 1990.
narrow spaces between the microvilli form a kind of sieve through which nutrients
must pass prior to their luminal absorption.5 It is here, in the unstirred water layer,
that the acid-Schiff reaction occurs for uptake of polysaccharides.5 It is also here that
the intestinal environment remains acidic, thus potentially allowing amino acid che-
lates to reconfigure into the positively charged amino acid chelate/complexes.9 This
acid pH of the unstirred water layer between the microvilli also increases mucosal
cell membrane permeability.
As noted, not all of the enzymes found in the small intestine are secreted into the
lumen. Many of these digestive enzymes remain attached to the mucosal cell mem-
brane, with only their active sites exposed to the chyme. These enzymes are referred
to as brush border enzymes. Table 8.1 summarizes a few representative examples.4
Several of the enzymes listed in Table 8.1, such as aminopeptidase or ATPase
(adenosine triphosphatase), are mineral dependent for their activations.10 While these
minerals function as enzymatic cofactors, the minerals are probably not sourced
directly from the chyme. Generally, they first have to be absorbed into the mucosal
cell from ingested minerals in the food and so on, and while in that intracellular
state, they are incorporated into the enzymes that are being built in the mucosal cell
prior to being exported back to the brush border.11
The microvilli of the absorptive cell are covered with glycocalyx or “a fuzzy coat”
as seen in Figure 8.1. The glycocalyx consists of a network of fibers that are partially
composed of mucin and attached to the membrane of the microvilli.1,2 Mucin cov-
ers the surface of the gastrointestinal tract and in so doing protects its surface from
enzymatic and pH damage associated with digestion. Mucin further plays a direct
role in the absorption of charged particles, such as metal ions and certain amino acid
chelates/complexes as described in Chapter 7. Mucin contains oligosaccharide side
chains composed of glycolipids and glycoproteins. These side chains have negatively
charged sialic acid termini that are on both the glycoproteins and gangliosides.5
Due to their negative charges, the mucins are able to capture and hold available
cations or other positively charged molecules found in the intestine. In essence, this
“traps” these positively charged ions or molecules and subsequently allows them
to be carried into the mucosal cell by either facilitated diffusion or active trans-
port. Once the positively charged ion, such as a cation, is attached to the negatively
charged sialic acid terminus on the mucin of the glycocalyx, the ion is no longer
reactive and thus is prevented from entering into other chemical reactions that might
limit its absorption.12,13 When the mucin on the cell membrane is exposed to neur-
aminidase, an enzyme that removes the negatively charged sialic acid termini from
the glycocalyx, there is a reduction of the negative charge on the membrane.5 This
results in a “release” of the positively charged molecule and allows it to be bonded
to a negatively charged molecule in a transport carrier and absorbed by facilitated
diffusion or active transport into the mucosal cell. Since this absorption is highly
regulated,14 it is speculated that, in the case of metal ions, the tissue in the body sends
signals to the absorptive cells, which then control the rate of removal of sialic acid
and subsequently the absorption of the cations.15
Glycocalyx is also involved in the digestion of ingested protein as well as “con-
centrating” amino acids from the digested protein.5 The ability to collect the amino
acids lies in them being able to reconfigure into positively charged molecules in
that acid pH similar to the reconfiguring of amino acid chelates into amino acid
complexes in that same acid pH environment. Thus, glycocalyx is also involved
in concentrating amino acid chelate/complexes in preparation for absorption. This
concentrating process is essential for their absorption by diffusion, as discussed fur-
ther in this chapter.
In the acidic pH of the stomach, duodenum, or the microenvironment between
the microvilli, mucin is capable of capturing an amino acid chelate as the amino
bond is broken on the chelate molecule and positively charged amino acid chelate/
complex is formed.9 The bonding would not be from the negative charge on the
mucin/glycocalyx to the metal ion, but instead it would be to the positive charge on
the amino moiety of the reconfigured amino acid ligand.
Once attached to the glycocalyx, the metal amino acid chelate/complex is held
similarly to an amino acid/peptide until being released by exposure to neuraminidase
or some other releasing molecule. When that occurs, the chelate/complex is absorbed
into the mucosal cell by active transport similarly to an amino acid or by certain types
of facilitated diffusion, which is discussed in the next chapter. Another possibility
that may affect the release of the amino acid chelate/complex from the mucin may be
that the strength of the potential bond between the positive charged amino moiety of
the amino acid ligand and the negative charge on the transport molecule embedded
within the cell membrane may be greater than the stability constant created from
holding the amino acid chelate/complex to the mucin. The attraction of a potentially
higher formation constant of the transport molecule would result in a greater quantity
of the amino acid chelate being removed from the glycocalyx compared to the seques-
tering of metal ions that are released only on receiving a signal from the body that
the metal ion is required. A third possibility is that there may be a signal requiring
the release of amino acids from the mucin in preparation for their absorption. If such
a signal exists, it would also affect the absorption of an amino acid chelate/complex.
In considering any of the three possibilities for the release of an amino acid
chelate/complex from mucin, it becomes obvious that another role played by mucin
is that of coating the glycocalyx not only to protect it but also to attract the charged
amino acid chelate or chelate/complex and hold it until a transport molecule in the
mucosal cell membrane can pick up the chelate or chelate/complex molecule and
later transfer it into the cytoplasm of the absorptive cell. It should be remembered
that in the case of the amino acid chelate/complex, the body is not absorbing a metal
ion. It is absorbing an amino acid-like molecule. Thus, the role of the glycocalyx as
it relates to amino acid chelates is to capture the amino acid ligand, not the metal ion
attached to the ligand. As the ligand is absorbed, the metal ion is “smuggled” into the
mucosal cell as part of the amino acid-chelated/complex molecule.16
The glycocalyx has other roles that are also essential to nutrition, such as chang-
ing the concentration of different substances, including amino acids, amino acid che-
lates or chelate/complexes, and metal ions, at the surface of the cell.5 This is crucial
for diffusion of cations, amino acid chelates or chelates/complexes, and so on into
the mucosal cell. The membrane of the cell behaves as though it were lipid. It will
impede the flow of water and charged solutes (i.e., cations, amino acids, amino acid
chelates or amino acid chelate/complexes) in and out of cells.
Besides forming a covalent bond between the amino acid chelate/complex and
the glycocalyx, there can be another type of attraction. Known as the van der Waals
force, it results in a momentary nonsymmetrical electron distribution in the molecule
that results in a dipolar attraction. A dipolar attraction causes the positive end of one
polar molecule to be weakly attracted to the negative end of another polar molecule.
The van der Waals attraction can apply to both the amino acid chelate via its waters
of hydration or the amino acid chelate/complex.
However it occurs, when the concentration of an amino acid or amino acid chelate
or chelate/complex outside the cell is equal to the concentration of the amino acid
within the cytoplasm of the cell, there is no movement across the membrane by dif-
fusion. If, however, the glycocalyx increases the concentration of that amino acid
chelate or chelate/complex on the luminal side of the membrane to a greater concen
tration than the amino acid concentration within the cytoplasm of the mucosal cell,
certain types of diffusion can move more of that amino acid chelate or chelate/
complex into the cell until an equal concentration of the amino acid chelate (amino
acid) exists on both sides of the cell membrane. The rate of diffusion is dictated by
the relative solubility of the amino acid chelate or chelate/complex in the aqueous
extracellular fluid, the concentration of integral proteins in the lipid membrane, and
the proportions of amino acids (amino acid chelate or chelate/complex) on both sides
of the mucosal cell membrane.3
This difference is dramatically illustrated in Figure 8.3. Here, the in vivo absorp-
tion of iron from ferrous carbonate is contrasted to that of iron amino acid chelate
in microvilli.17,18 The masses of ferrous carbonate must be broken down into micel-
lular iron before being transported across the mucosal cell membrane (a). In the
Iron carbonate
Microvilli
Microvilli
Iron amino
acid chelate
Terminal web
(A) (B)
FIGURE 8.3 Uptake of iron from iron carbonate (a) and iron amino acid chelate (b). The
dark streaks in the microvilli of b are iron amino acid chelate/complex. (From Freeman,
JA, and Greer, JC, “Intestinal fat and iron transport, goblet cell mucus secretion and cel-
lular changes in protein deficiency observed with an electron microscope,” Am J Digest Dis
10:1005–1025, 1965. With permisson from Springer; and from Ashmead, HD, Graff, DJ, and
Ashmead, HH, Intestinal Absorption of Metal Ions and Chelates (Springfield, IL: Thomas)
157, 1985.)
case of the iron amino acid chelate (b), absorption of the molecule into the micro-
villi and cytoplasm is very rapid and in significantly greater amounts,19 as indicated
by the dark streaks at the terminal web and basal portion of the microvilli. Not all
of the iron amino acid chelate in Figure 8.3 was absorbed into the mucosal cell by
facilitated diffusion. Some of it was probably also taken up by active transport.
If the amino acid chelate is absorbed by facilitated diffusion, a carrier molecule
is still required to translocate it, but no energy is expended. The amino acid chelate
must be capable of forming an attraction with a moiety on the transport molecule to
attach the amino acid chelate to that carrier molecule. This carrier molecule can
be water, an integral protein, or some other molecule. The transport mechanism
for facilitated diffusion of an amino acid chelate/complex requires (1) substrate and
stereospecificity, (2) competition between related substances, and (3) saturability
of the mechanism with increasing concentration.14 Thus, most amino acid chelates
are probably absorbed by facilitated diffusion rather than reconfiguring as chelate/
complexes and being absorbed by active transport.
Like facilitated diffusion, to be absorbed by active transport the amino acid
chelate must be able to bond to a transport molecule. The bonding of the chelate
for active transport is via the positive charge on the amine moiety of the following
reconfiguring as a chelate/complex ligand. As previously noted, a metal ion, such
as zinc, can be chelated with one, two, or possibly up to five amino acids, such as
glycine. In the case of the zinc bisglycinate, the molecule remains in chelated form
at a pH range of approximately 5 to 9. When ingested, the initial environment in the
mouth and esophagus is slightly alkaline. At that pH, the zinc ion remains chelated,
and the zinc bisglycinate molecule is chemically unreactive (Figure 8.4).
Zn
FIGURE 8.4 Zinc bisglycinate at a pH of approximately 5 to 9. (Redrawn from Ericson, C,

and Ashmead SD, “A novel approach in confirming dietary amino acid chelates by the uti-
lization of ninhydrin,” poster presentation at American Chemical Society Annual Meeting,
September 2004.)
O
NH3+
O Zn O
+H N
3 O
FIGURE 8.5 Zinc bisglycinate at a pH of approximately 2 to 5.
As the zinc amino acid chelate enters the stomach and mixes with the stomach
acid, the pH drops, and the bonds between the metal and the nitrogen on the amino
moieties of the two glycine ligands begin to break (Figure 8.5).9 When the bonds
break, the stomach acid donates a hydrogen ion to the amine terminus of each ligand,
resulting in it gaining a positive charge on the two ligands of the newly formed zinc
bisglycine chelate/complex. The zinc ion continues to remain attached to the amino
acid ligands via carboxyl bonds.9
As noted, this zinc molecule, while a type of amino acid complex resulting from
the reconfiguring of the amino acid chelate, is not the same as a zinc amino acid com-
plex as defined by the Association of American Feed Control Officials (AAFCO).20,21
Under the AAFCO definition, the metal ion is attached to the amino acid by only an
amino bond. When an AAFCO-defined amino acid complex is subjected to an acid
or alkaline pH similar to the environments found in the gastrointestinal tract, the
amino bond is broken similarly to an amino acid chelate but with a different result.
The breaking of this bond causes the metal ion to be released. In the case of the zinc
amino acid chelate/complex that is created when the amino bonds of the chelate
are broken, the metal ion remains attached to the amino acid ligands via the car-
boxyl moieties. In an acid pH, this results in a positive charge on each terminus of
the amine moieties, which allows the entire molecule to be attached to a negative
charge on the glycocalyx or directly to a negative charge on a transport molecule.
The amino acid chelate/complex can then be carried across the stomach lining mem-
brane into those absorptive cells lining the stomach.
If the amino acid chelate molecule is soluble, and depending on the pathway taken,
it may also potentially be absorbed from the stomach or duodenum by facilitated dif-
fusion in the chelated form. In either case, while in the stomach, the metal remains
O
NH2
O Zn O
H2N O
FIGURE 8.6 Zinc bisglycinate at a pH of approximately 9 to 12.
attached to its amino acid ligands. As the chelate/complex molecule moves into the
duodenum, it starts to reconfigure as an amino acid chelate at the distal end of
the duodenum. When that occurs, passive diffusion can pull more of the chelate
into the mucosal cell provided the amino acid chelate remains soluble.
Further on in the small intestine, beginning in the proximal jejunum, the pH
environment of the lumen starts to become alkaline. If the alkalinity becomes high
enough, the zinc bisglycinate chelate will again reconfigure into a metal chelate/
complex as the bonds between the nitrogen on the amine moieties and the metal ion
are again broken (Figure 8.6).9 But now, because the environment is alkaline, there
are no hydrogen ions available to create a positive charge on the terminus of either
amine moiety of the reconfigured amino acid chelate/complex. When that situation
occurs, the amino acid chelate/complex will not immediately be attracted either to
a negative charge on the glycocalyx or to a negative moiety on a carrier molecule.
There are, however, positive charges on both the peripheral and the integral proteins
that are embedded in the cell membrane. These positive charges through ion-dipole
forces tend to attract the amino acid chelate/complex and pull the molecule toward
the cell membrane.22
As the amino acid chelate/complex approaches the cell membrane, it can be sub-
jected to another pH environment change between the microvilli in the unstirred
water layer. Even though the luminal environment, as a whole, may be alkaline, due
to the acid-Schiff reaction occurring between the microvilli,5 the microenvironment
in the unstirred water layer remains acidic, causing the amino acid chelate/complex
molecule to potentially change again. If that occurs, each NH2 moiety on the zinc
bisglycinate chelate/complex is able to gain a hydrogen ion. When the ligand of the
amino acid chelate/complex acquires a positive charge on its amine moieties, the
molecule can then be bonded to a negative charge on a transport molecule located
in the absorptive cell membrane, thus initiating its active transport across the mem-
brane into the cell.
As shown in Figure 8.7, the amino acid, like an amino acid chelate, gains a posi-
tive charge on its amine moiety when subjected to that same acid pH environment.
When the pH in the intestine reaches a certain level of basicity, this positive charge
is lost; the amino acid reconfigures and becomes neutral. But as noted, the pH envi-
ronment near the cell membrane between the microvilli remains acidic due to the
acid-Schiff reaction between the microvilli.5 Thus, when presented to the mucosal
cell membrane, the amino acid can potentially reconfigure into a positively charged
molecule by gaining a hydrogen ion on its amine moiety, which it acquires from the
O O O
–O –O
HO NH3+ NH3+ NH2
pH 1 pH 7 pH 12
FIGURE 8.7 The effect of pH on amino acid configuration.
acidic environment. At that point, like the amino acid chelate/complex, an amino
acid can be similarly attached to the glutathione and subsequently transferred to the
cytoplasm of the mucosal cell by using the same active transport system.16
It is important to understand that the changes that an amino acid/peptide or an
amino acid chelate undergo in the different pH environments are fundamental to
either type of molecule being able to be absorbed by active transport or possibly by
certain modes of facilitated diffusion. The membrane of the cell must act as a perme-
ability barrier that has sufficient mechanical strength to remain intact throughout the
lifetime of the cell. Concurrently, it has to be fluid enough to allow molecules to pass
through the membrane against a concentration gradient, but it cannot be so fluid that
it permits the wholesale exit of its cellular contents through leakage.8 This includes
metal ions, amino acid chelates, amino acids, peptides, and sugars. The fluid charac-
teristic of the membrane is critical.
Elucidation of the cell membrane has evolved over time. Most investigators now
believe that the Singer fluid mosaic model most accurately depicts the cell mem-
brane.23,24 Shown in three-dimensional form, Figure 8.8 depicts this model.23 The
Glycocalyx
Lipids
Peripheral
protein
Integral Terminal web

protein
Myosin filaments
FIGURE 8.8 A representation of the Singer fluid model of cell membranes. (Redrawn from
Singer, SJ, and Nicolson, GL, “The fluid mosaic model of the structure of the cell membrane,”
Science 175:720731, 1972.)
membrane consists of a lipid bilayer with the hydrophilic ends directed outward
and its hydrophobic ends directed inward. The solid bodies in Figure 8.8 represent
integral and peripheral proteins that are embedded within the lipid bilayer. They are
randomly distributed along the plane of the membrane, some spanning the entire
width of the membrane (integral proteins), while others are only partially embedded
(peripheral proteins). The degree to which the integral protein traverses the lipid
bilayer depends on the size and structure of the molecules forming the protein.23,24
Some integral proteins may be enzymes and, as such, function as transporters for
cellular uptake of certain nutrients, which includes the absorption of amino acid
chelates. Other integral proteins may form pores or carrier molecules that allow
facilitated diffusion to occur.
As suggested in Figure 8.9, some integral proteins in mucosal cells can be folded
polypeptide chains that partially protrude from the membrane into the cytoplasm of
the mucosal cell as well as into the lumen of the intestine. On both the luminal side
and the cytoplasm side of the membrane surface, these proteins contain ionic resi-
dues that can attract charged molecules. The nonionic residues are found mostly in
the embedded parts of the protein.25,26 In addition to the ionic charges on the integral
proteins, glycocalyx, which, as noted, carries a negative charge, extends from both
the integral and peripheral proteins into the lumen.27
Besides the integral proteins embedded within the lipid bilayer, as mentioned,
there is another group of proteins associated with the mucosal cell membrane called
peripheral proteins. The peripheral proteins are generally found on the surface
of the lipid bilayer. They are loosely held by an electrostatic interaction with the
hydrophilic lipid head.28 Due to their charges, these peripheral proteins are able to
attract charged molecules from the lumen. If the peripheral protein is an enzyme,
the bonding of a charged molecule to the peripheral protein may initiate enzymatic
activity. These peripheral proteins are easily removed from the cellular membrane
Lumen
Integral Peripheral
+ ––
protein – protein
– –+ –
+
Lipids
– +
–
+ –
Cytoplasm
FIGURE 8.9 Cross section of a mucosal cell membrane showing both integral and periph-
eral proteins embedded in the membrane. (Redrawn from Giese, AJ, Cell Physiology
(Philadelphia: Saunders) 192, 1973.)
and may, in some instances, constitute the digestive enzymes that are secreted into
the lumen.
As illustrated in Figure 8.9, not every integral protein is a single continuous poly-
peptide chain. Frequently, more than one independent peptide may comprise a par-
ticular integral protein. Furthermore, as noted, enzymes may compose all or part
of the integral protein and in some cases form a pore. These enzymes can reside on
the surface of the membrane, where they participate in digestion, or they may be
part of a transport molecule that facilitates the movement of nutrients across the cell
membrane.3 One such enzyme that forms part of a peptide system that transverses
the cellular membrane and would probably be classified as an integral protein is the
γ-glutamyl transport system.29–31 The enzyme γ-glutamyl transpeptidase is located
on the surface of the cell membrane. It has been identified as the enzyme that initi-
ates the active transport of amino acids or peptides into the absorptive cell from the
lumen.32,33 Because an amino acid chelate behaves similarly to an amino acid, this
same enzyme is also implicated in the active transport of that molecule as well.16,34
The active transport of an amino acid, small peptide, or amino acid chelate/
complex begins with the intracellular formation of glutathione, which is the most
abundant intracellular thiol ester except for glutamine. Glutathione is composed of
three amino acids: L-glutamine, L-cysteine, and glycine.33 The initial reactions are
as follows (ADP, adenosine diphosphate; ATP, adenosine triphosphate):
1. L-Glutamate + L-Cysteine + ATP → L-γ-Glutamylcysteine + ADP + Pi

2. L-γ-Glutamylcysteine + Glycine + ATP → Glutathione + ADP + Pi
Once the glutathione has been created in the cytoplasm of the mucosal cell, it is
moved to the surface of the mucosal cell membrane. There, the glutathione is bro-
ken down by the membrane-bound enzyme γ-glutamyl transpeptidase. This enzyme
attaches the γ-glutamyl moiety of the glutathione to the positively charged amino
moiety of an amino acid, peptide, or amino acid chelate/complex that has been
captured at the brush border of the absorptive cell. The binding of the amino moi-
ety to the γ-glutamyl moiety is illustrated by the following formula: Glutathione +
Amino acid (or peptide or amino acid chelate/complex) = γ-Glutamyl-Amino acid (or
γ-glutamyl peptide or γ-glutamyl amino acid chelate/complex) + Cysteinyl-glycine.
The γ-glutamyl amino acid (or peptide or amino acid chelate/complex) bound by
transpeptidation is subsequently converted to 5-oxoproline and the corresponding
amino acid (or peptide or amino acid chelate/complex) as a result of the action of a
second enzyme in the basal portion of the same integral protein, γ-glutamyl cyclo-
transferase: L-γ-Glutamyl-amino acid (peptide or amino acid chelate/complex) →
5-Oxo-L-proline + L-Amino acid (peptide or amino acid chelate/complex). The
amino acid (or peptide or amino acid chelate/complex) loses its charge and is sub-
sequently released into the cytoplasm of the mucosal cell, and 5-oxo-L-proline
commences its enzymatic steps to re-forming glutathione. This entire process is
illustrated in Figure 8.10.33
For the γ-glutamyl transport system to attach to the amino acid/peptide or amino
acid chelate/complex and initiate movements of either across the mucosal cell mem-
brane, there has to be a positive charge on the amine moiety of the molecule to bind
Amino Acid Chelate/Complex (glutathione)

or ADP + Pi
γ-glu-cysH-gly
Amino Acid
ATP
γ-Glutamyl Transpeptidase GSH Synthetase
glycine
AACC cysH-gly
γ-glu-AA or AACC
Peptidase γ-glu-cysH
γ-Glutamyl Cyclotransferase ADP + Pi

cysteine
γ-Glu-cysh Synthetase
ATP
Amino Acid
5-oxoproline glutamate
or
Amino Acid Chelate 5-Oxoprolinase
or
Amino Acid Chelate/Complex
ATP ADP + Pi
FIGURE 8.10 The γ-glutamyl transport system for the uptake of amino acids, peptides, or
amino acid chelate/complex into the mucosal cell. AA = amino acid; AACC = amno acid
chelate/complex. (Redrawn and adapted from Meister, A, Tate, SS, and Thompson, GA,
“On the function of the γ-glutamyl cycle in the transport of amino acids and peptides,” in
Elliott, K, and O’Connor, M, eds., Peptide Transport and Hydrolysis (Amsterdam: Elsevier)
123–143, 1977.)
it to the enzyme. When the amino acid/peptide or amino acid chelate/complex is

attached to γ-glutamyl transpeptidase via the positively charged amine moiety, it
initiates enzymatic activity. The enzyme can commence the hydrolyzation of the
glutathione, which results in the movement of these proteinaceous molecules into
the mucosal cell.
The hydrolytic activity of the glutathione does not affect the basic structure of the
amino acid/peptide any more than it affects the structure of the amino acid chelate/
complex. The γ-glutamyl moiety is attached to the amino acid at the amino terminus.
The binding of the amine moiety of the amino acid (peptide) or amino acid chelate
to that enzyme is the key that turns the enzyme on. When the transporter molecule is
finally attached to the γ-glutamyl moiety in place of the γ-glutamyl transpeptidase, the
enzyme is turned off, and the γ-glutamyl amino acid/peptide or amino acid chelate/
complex is moved toward the cytoplasm to be acted on by γ-glutamyl cyclotrans-
ferase. This last enzyme releases the γ-glutamyl amino acid/peptide or amino acid
chelate/complex by neutralizing the charge on the amino moiety as the amino acid/
peptide or amino acid chelate/complex enters the cytoplasm and thus returns the
amino acid/peptide or amino acid chelate/complex to its original state. The presence
of the metal ion in the amino acid chelate does not interfere with the absorption of
the proteinaceous molecule. The metal does not enter into any of the enzymatic reac-
tions required to transfer the amino acid chelate across the membrane.16 The metal
remains attached to the carboxyl moieties of the amino acids and just “goes along
for the ride.”
Payne has written, “When an impermeant substance is chemically linked to a
peptide and the resultant complex can be transported via a peptide permease the
impermeant moiety can then be ‘smuggled’ across the cell membrane, and the com-
plex is accordingly called a ‘smuggler’.”16 Ames et al. added that the γ-glutamyl
transport system is capable of transporting other molecules or atoms through the cell
membrane if those molecules or atoms have been chemically attached to the trans-
ported peptides prior to the absorption of the peptide.35
More recently, Cannon et al. conducted a study that tested the hypothesis that
amino acid transport systems are involved in the transport of mercury bonded to
cysteine in a 1:2 (metal-to-amino-acid) ratio.36 They reported that there are several
transport systems, including facilitated diffusion, available for the uptake of amino
acids, which tends to confirm an earlier study in which investigators found that
amino acids were absorbed via a hierarchy depending on the transport mechanism
available for the uptake of a specific amino acid.37 All of these systems seem to
operate similarly, in that they are all amino acid dependent, although each has some
degree of specificity for certain amino acids. Cannon et al. concluded that, “The
transport data obtained when L-cystine was added to the luminal fluid indicated
strongly that dicysteinalmercury is likely transported as a molecular homolog of
L-cystine.”36 By implication, their finding would apply to other metals that have been
chelated to other amino acids. The metal ion in an amino acid chelate is absorbed
into the mucosal cell as part of the amino acid to which it is attached. It is not
absorbed as a metallic molecule.
The research of Mellander also demonstrated this implication.38,39 He reported
that when calcium was chelated to amino acids, those same amino acids remained
attached to the calcium and facilitated the transport of the calcium across the cell
membrane. He concluded that small peptides could also function as carriers to trans-
port the calcium into cells.38,39
Hemmings and Williams came to a similar conclusion based on their studies
with iron ferritin.40 They administered labeled ferritin that was complexed to immu-
noglobulin to suckling rats. When blood samples from the pups were analyzed, 75%
of the serum ferritin did not precipitate when treated with a precipitating chemical.
This demonstrated that the ferritin that had been complexed to the peptide molecule
prior to administration was absorbed through the mucosal tissue and into the blood
still bound to the original peptide molecule. The remaining 25% may have also
been absorbed intact into the mucosal tissue, but it subsequently underwent intra-
cellular hydrolysis within the mucosal cells prior to passing the released ferritin on
to the blood.40
These researchers all demonstrated that metal ions or other molecules can be
transported across the cell membrane as part of an amino acid or small peptide
molecule without entering into any chemical reactions themselves when those metal
ions or other molecules are chemically bound to amino acids or small peptides. It
is only the amine moiety of the amino acid or peptide molecule that is involved in
the enzymatic reactions necessary for the facilitated diffusion or active transport of
the entire complex across the cell membrane. In essence, the metal is “smuggled” in
as part of an amino acid/peptide-like molecule, as was suggested by Payne.16
Ugolev and Kooshuck published data suggesting that most dipeptides are cleaved
into amino acids by brush border enzymes prior to absorptive cell uptake.41 In
their study, they exposed dipeptides, consisting of a large and a small amino acid,
to the intestinal membrane. They then measured the ratio of the two amino acids on
the luminal and cytoplasmic sides of the membrane. If no hydrolysis occurred, they
reasoned that the ratio between the amino acids should be the same on both sides. In
other words, the dipeptides were absorbed intact. To the contrary, however, the ratio
changed, suggesting that hydrolysis of some, or all, of the dipeptides occurred on the
brush border surface adjacent to the transport sites.
Even if dipeptides are hydrolyzed extracellularly, it still does not negate the amino
acid/peptide or amino acid chelate/complex absorption model described because in
either case, the initiation of the enzymatic process for active transport requires the
presence of a positive charge on the terminus of the amine moiety of the amino acid
molecule whether it be in the form of an amino acid, a peptide, or an amino acid
chelate/complex.42 Hydrolysis of a dipeptide will create two amino acids in place
of the peptide.38 When this occurs, the amine groups of each of the resulting amino
acids that were subsequently subjected to the same acid reaction in the duodenum
or at the brush border of the mucosal cells gain a hydrogen ion that allows it to be
bonded to the transport molecule.
Since it is not a peptide, if an amino acid chelate/complex is presented to the
membrane of the mucosal cell, further hydrolysis of the molecule is unnecessary for
its uptake into the cell; furthermore, such a molecule would not respond to peptidase
activity. However, a metal proteinate, which is a metal chelated to amino acids or
partially hydrolyzed protein,20 may contain some peptide bonds that require further
hydrolyzation before absorption of any part of that molecule could occur. Depending
on how the metal ion is attached in the proteinate molecule, the metal ion could
be released during luminal hydrolysis of the proteinate molecule. If the metal ion is
released, it will be absorbed into the mucosal cell similarly to any other free metal
ion and at about the same rate.
REFERENCES
1. Kessel, RG, and Kardon, RH, Tissues and Organs: A Text-Atlas of Scanning Electron
Microscopy (San Francisco: Freemen) 171–177, 1979.
ed., Medical Physiology (St. Louis, MO: Mosby) 1145–1147, 1974.
4. Fox, SI, Human Physiology (Dubuque, IA. Wm. C. Brown) 534, 1990.
158–166, 1975.
6. Moog, F, “The lining of the small intestine,” Sci Am 245:154–176, November 1981.
7. Brestcher, A, and Weber, K, “Localization of actin and microfilament-associated pro-
teins in the microvilli and terminal web of the intestinal brush border by microscopy,”
J Cell Biol 79:839–845, 1978.
8. Giese, AC, Cell Physiology (Philadelphia: Saunders College) 184–202, 1973.

9. Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary amino acid che-
10. Schutte, KH, The Biology of the Trace Elements (Philadelphia: Lippincott) 17–23, 1964.
11. Ashmead, HD, and Samford, RA, “Increasing protein/energy digestion by feeding metal
amino acid chelates,” Int J Res Vet Med 6:38–45, 2008.
12. Conrad, ME, Umbreit, JN, and Moore, EG, “A role for mucin in the absorption of inor-
ganic iron and other metal cations,” Gastroenterology 100:129–136, 1991.
13. Simovich, M, Hainsworth, LN, Fields, PA, Umbreit, JN, and Conrad, ME, “Localization
of the iron transport proteins mobilferrin and DMT-1 in the duodenum: The surprising
role of mucin,” Am J Hematol 74:32–45, 2003.
14. Roy, CN, and Enus, CA, “Iron homeostasis: New tales from the crypt,” Blood
96:4020–4027, 2000.
15. Conrad, ME, and Umbreit, JN, “Pathways of iron absorption,” Blood Cell Mol Dis
29(3):336–338, 2002.
16. Payne, JW, “Transport and hydrolysis of peptides by microorganisms,” in Elliott, K, and
O’Connor, M, eds., Peptide Transport and Hydrolysis (Amsterdam: Elsevier) 307, 310,
320, 329, 1977.
17. Freeman, JA, and Greer, JC, “Intestinal fat and iron transport, goblet cell mucus secre-
tion and cellular changes in protein deficiency observed with an electron microscope,”
Am J Digest Dis 10:1005–1025, 1965.
made from various protein sources into rat jejuna slices, in vitro,” paper presented at
Utah Academy Arts, Letters and Science, Salt Lake City, UT, April 1970.
officials) 361, 2008.
Officials) 310–311, 2008.
22. Hughes, M, The Inorganic Chemistry of Biological Processes (London: Wiley) 71–72,
1972.
23. Singer, SJ, and Nicolson, GL, “The fluid mosaic model of the structure of the cell mem-
brane,” Science 175:720–731, 1972.
24. Singer, SJ, “The molecular organization of membranes,” Annu Rev Biochem 43:805–833,
1974.
25. Lenard, J, and Singer, SJ, “Protein confirmation in cell membrane,” Proc Natl Acad Sci
USA 65:720, 1970.
26. Rothfield, R, ed., Structure and Function of Biological Membranes (New York:
Academic Press) 145, 1971.
27. Pollock, EG, “Fine structural analysis of animal cell surfaces: Membrane and cell sur-
face topography,” Am Zool 18:25–69, 1978.
28. Berliner, LJ, “The spin-label approach to labeling membrane protein sulfhydryl groups,”
Ann NY Acad Sci 44:153–161, 1983.
29. Tate, SS, and Meister, A, “Interaction of γ-glutamyl transpeptidase with amino acids,
dipeptides and derivatives and analogs of glutathione,” J Biol Chem 249:7593–7602,
1974.
30. Meister, A, and Tate, SS, “Glutathione and related γ-glutamyl compounds: biosynthesis
and utilization,” Annu Rev Biochem 45:559–604, 1976.
31. Meister, A, Tate, SS, and Ross, LL, “Membrane bound γ-glutamyl transpeptidase,”
in Martinosi, A, ed., The Enzymes of Biological Membranes (New York: Plenum) v3,
315–347, 1976.
Hill) 630–635, 1973.
33. Meister, A, Tate, SS, and Thompson, GA, “On the function of the γ-glutamyl cycle in
the transport of amino acids and peptides,” in Elliott, K, and O’Connor, M, eds., Peptide
Transport and Hydrolysis (Amsterdam: Elsevier) 123–143, 1977.
Chelates (Springfield, IL: Thomas) 126–152, 171–181, 1985.
35. Ames, BN, Ames, FL, Young, JD, Tsuchiya, D, and Lecorcq, J, “Illicit transport, the
oligopeptide permease,” Proc Natl Acad Sci U S A 70: 456–458, 1973.
36. Cannon, VT, Kalups, RK, and Barfuss, DW, “Amino acid transporters involved in lumi-
nal transport of mercuric conjugates of cysteine in rabbit proximal tubule,” J Pharmacol
Exp Ther 298:780–789, 2001.
37. MacInnis, A, and Graff, DJ, “Specificity of amino acid transport in the tapeworm,
Hymenolepsis iminuta, and its rat host,” Rice Univ Studies 62:183, 1977.
38. Mellander, O, “The nutritional significance of some peptides,” Voeding 16:363, 1955.
39. Mellander, O, “The physiologic importance of the casein phosphopeptide calcium salt,
II: peroral calcium dosage of infants,” Acta Soc Med Ups 55:247–255, 1950.
40. Hemmings, WA, and Williams, E, “Transport of ferritin across the ileum of suckling
rats,” Int Res Commun Syst Med Sci 3:215, 1975.
41. Ugolev, AM, and Kooshuck, RI, “Hydrolysis of dipeptides in cells of the small intes-
tine,” Nature 212:859–860, 1966.
9 The Absorption of
Amino Acid Chelates
by Facilitated Diffusion
Chapter 8 focused primarily on the intestinal absorption of amino acid chelates by
active transport. Amino acid chelates can also be absorbed into the mucosal cell
by diffusion.1 While more than one type of diffusion into the mucosa can occur,
the absorption of amino acid chelates primarily takes place by facilitated diffu-
sion, using one or more pathways that are available for this form of absorption.
It is unlikely an amino acid chelate molecule can cross the mucosal cell mem-
brane independent of a carrier molecule regardless of whether that transport process
is by facilitated diffusion or by active transport. The pathway assigned for intestinal
uptake of an amino acid chelate, at a given moment, depends on the luminal environ-
ment that the chelate is subjected to, including the pH. As discussed in the previous
chapter, if the pH is low enough to cause the amino bond between the metal ion and
the ligand to break,2 the amine moiety can pick up a hydrogen ion and the resulting
amino acid chelate/complex can be absorbed by active transport once it is bonded
to the enzyme system that functions as a transporter.3 If the pH is high enough to
break the amino bond, the chelate/complex molecule will probably be absorbed by
facilitated diffusion as described in this chapter. If the luminal pH is not low enough
to be conducive to reconfiguring the amino acid chelate into a chelate/complex, the
amino acid chelate can still be absorbed by facilitated diffusion. In those cases, it
will require a carrier molecule that will either affect the permeability of the mucosal
cell membrane in such a way that it admits the chelate and its carrier molecule into
the cytoplasm, shuttles the amino acid chelate from one side of the mucosal cell
membrane to the other (Figure 9.1), or both.1,4 Unlike active transport, in the case
of facilitated diffusion, no energy is imbued to the transport molecule that carries
the chelate across the mucosal cell membrane. If energy were expended, the amino
acid chelate would be absorbed against a concentration gradient by active transport.
While diffusion seems relatively easy to accomplish, the actual process can be
rather complex. “Diffusion is movement due solely to the kinetic energy and electri-
cal charge of the molecules and to the electrical field in which they exist.”5 Thus, for
an amino acid chelate molecule to be absorbed by diffusion, there must be some type
of attraction between it and its transport molecule that will allow the intact chelate to
be attached to the transport molecule. As a result, this will facilitate a modification
to the characteristics of the carrier molecule, which in turn will dictate a behavioral
change as it relates to the chelate in the luminal environment. That transporter mol-
ecule can be water molecules or other types of carrier molecules as long as there is
an attraction between the carrier molecule and the chelate.
135
Facilitated diffusion Cell membrane
FIGURE 9.1 Shuttling of an amino acid chelate across the mucosal cell membrane by facili-
tated diffusion. (Redrawn from Hunt, SM, and Groff, JL, Advanced Nutrition and Human
Metabolism (St. Paul, MN: West) 41, 1990.)
Unlike the process that occurs during the active transport of an amino acid chelate,
when the chelate is absorbed by facilitated diffusion, it does not always require recon-
figuring into a positively charged chelate/complex as a prerequisite for its absorption.
If it were to undergo such a change prior to uptake, the chelate would more likely be
absorbed by active transport because it could then be bonded to the glutathione in the
γ-glutamyl enzymatic cycle.3 Instead, the amino acid chelate is absorbed as either an
intact chelate or, in certain cases, a chelate/complex. When it is absorbed as an intact
amino acid chelate, its attachment to a carrier molecule for facilitated diffusion does
not affect the molecular structure of the chelate in the sense of changing the position
of any of the atoms that compose the amino acid chelate. The attachment of the car-
rier molecule to the chelate does increase the molecular weight of the total combined
chelate/carrier molecule, but this does not cause any change in the structure of the
chelate itself.
When an amino acid chelate diffuses into the mucosal cell, it can cross the mem-
brane through polar channels within the membrane or between the cells composing
the mucosal tissue or even directly through the bilayer lipid sheet of the mucosal cell
membrane.5,6 The movement of the amino acid chelate into or subsequently out of
the cell takes place only in the direction of the prevailing electrochemical gradient.5
Because the amino acid chelate molecule is taken up by the mucosal cell as if it were
an amino acid or a small peptide, there is competition for carrier molecules between
it and similar amino acids or small peptides that are also absorbed by facilitated dif-
fusion.5,7 There is a hierarchy that dictates which amino acid will be absorbed first,
second, and so on.8 Thus, it seems logical that an amino acid chelate with a ligand
that is higher in that hierarchy will be absorbed before a chelate or amino acid with
a lower position.
While water can diffuse directly through the membrane via its pores or even
through the lipid bilayer of the membrane, certain larger molecules, such as amino
The Absorption of Amino Acid Chelates by Facilitated Diffusion 137
acid chelates, cannot without the aid of a carrier molecule. As noted, they must first
be attached to a transport molecule and thus form a type of complex with the mol-
ecule. This carrier is then able to shuttle the intact chelate across the mucosal cell
membrane before liberating it in the cytoplasm. There are several carrier molecules
available to aid in the uptake of an amino acid chelate. When any of these carriers,
including water, facilitates the diffusion of the amino acid chelate without expend-
ing energy in the transfer process, it is known as facilitated diffusion. Frequently,
when water is involved in the diffusion, the process is termed solvent flow, but since
the amino acid chelate is carried into the mucosal cell by the water molecules, it is
included as part of the discussion on facilitated diffusion.
When absorption by facilitated diffusion occurs, the rate of movement of the
amino acid chelates levels off as its concentration increases within the mucosal
cell, whereas in the case of simple diffusion, its concentration continues to increase
linearly.9 Absorption data relating to the uptake of amino acid chelates have dem-
onstrated that a regulatory mechanism generally controls the rate and amount of
cell uptake of an amino acid chelate.10 As the concentration of the chelate increases
within the mucosal cell, its absorption rate from the lumen decreases. This regula-
tory control therefore suggests that a significant percentage of an ingested dose of
an amino acid chelate is taken up by facilitated diffusion. If absorption were accom-
plished by simple diffusion, no regulatory mechanism would be in evidence.
As noted previously, in the case of facilitated diffusion, a substance, such as water,
in combination with an amino acid chelate that is dissolved in that water, can freely
cross the cell membrane together as a single molecule. The concentration of the
water and dissolved amino acid chelate tends to equalize on both sides of the muco-
sal cell membrane so that the dissolved chelate moves from a higher concentration on
the luminal side of the membrane to a lower concentration within the mucosal cell.
In other words, it moves down a concentration gradient (Figure 9.2).1 Its uptake tends
Diffusion Cell membrane
Water & amino

acid chelate
FIGURE 9.2 The concentration gradient for facilitated diffusion of an amino acid che-
late. (Redrawn from Hunt, SM, and Groff, JL, Advanced Nutrition and Human Metabolism
(St. Paul, MN: West) 41, 1990.)
Facilitated diffusion pathway Facilitated diffusion

of an amino acid chelate pathway of an
amino acid chelate
Bulk
phase
Unstirred
water
layer
Cell
membrane
Transport
sites
Villus
Cytoplasm
FIGURE 9.3 The bulk water phase and unstirred water phase that surround a villus. Note
that the thickness of the unstirred water phase changes depending on its location in relation
to the villus. (Redrawn from Thomson, ABR, and Dietschy, JM, “Intestinal lipid absorp-
tion: Major extracellular and intracellular events,” in Johnson, LR, ed., Physiology of the
Gastrointestinal Tract (New York: Raven Press) v2, 1151–1158, 1981; and Thomson, ABR,
and Dietschy, JM, “Derivation of the equations that describe the effects of unstirred water
layers on the kinetic parameters of active transport process in the intestine,” J Theor Biol
64:277–294, 1977.)
to mimic the pathway of a solubilized amino acid that is absorbed as a solute in the
water by the same facilitated diffusion process.11
Even if the amino acid chelate is soluble in water, the concentration of that soluble
amino acid chelate in the chyme within the lumen will not be the same as its concen-
tration when it is presented to the mucosal cell membrane. This is due to a diffusion
barrier between the mucosal cell surface and the bulk perfusion medium. Known
as the unstirred water layer (Figure 9.3), this barrier consists of a series of layers of
water that extend outward from the mucosal cell membrane.12 As one moves away
from the cell membrane, each layer becomes progressively more stirred until they
blend imperceptibly with the bulk water phase in the lumen.5
If the amino acid chelate is soluble, it will generally commence going into solu-
tion as it enters the bulk water phase within the lumen. This dissolution is usually
completed in the bulk water phase, although the process can continue in the unstirred
water layer. Another alternative, which was alluded to in Chapter 8, is the attraction
of the amino acid chelate or chelate/complex to the glycocalyx on the microvilli via
the van der Waals force. In this situation, the amino acid chelate molecules can be
collected and concentrated extracellularly while waiting for dissolution and subse-
quent absorption by diffusion.13
O H O
H
O
O NH2 O NH2
H H H H
O M O O M O
H H H
H2H H
O H2N O
O
H H
O O
(a) (b)
FIGURE 9.4 The bonding of water molecules to an amino acid chelate molecule. (a) They
are attached to the metal ion by coordinate covalent bonds. (b) The water molecules are also
attached to the amino acids by an electrostatic attraction.
An amino acid chelate does not break apart into the metal ion and an anion
similarly to a dissolved sodium chloride molecule. Instead, the polar water mol-
ecules tend to be attracted either to the unoccupied ionic sites on the metal ion in the
chelate molecule or to the oxygen atoms in the ligand molecule.14,15 This is seen in
Figures 9.4a and 9.4b. In Figure 9.4a, two waters of hydration have been attached to
the metal ion by coordinate covalent bonds. In Figure 9.4b, water molecules not only
are attached to the metal ion by coordinate covalent bonds, but also are bonded to
the oxygen atoms on each ligand through hydrogen bonds. It is also possible to attract
another water molecule via a van der Waals attraction. Thus, while the resulting
chelate molecule example in Figure 9.4b contains four water molecules, it is theoreti-
cally possible to attach up to eight water molecules to a single amino acid chelate
molecule. As the water molecules are bonded or attracted to the chelate molecule, it,
in essence, goes into solution.
As explained in Chapter 2, depending on the valence of the metal ion as well as
the size of the amino acid ligands involved, more than one amino acid ligand can
chelate a metal ion. Thus, in the case of an iron cation, it can potentially be chelated
with one, two, or three amino acids depending on its valence, the availability of the
necessary free amino acids, and the size of each ligand. When a ferrous iron ion is
chelated to two amino acids of glycine, for example, the solubility of the resultant
molecule is approximately 1 chelate molecule per 10 parts of water. In other words,
it is relatively soluble.16 When the iron ion is further oxidized and becomes ferric
iron, it can be chelated to three glycine molecules, but in so doing, its solubility
may be reduced. To illustrate, a sample of an iron trisglycinate was subjected to
solubility tests and found to require approximately 10,000 parts of water per chelate
molecule to go into solution.16 The reason for the lower solubility is, of course, that
when the iron is chelated to three amino acids, it creates a molecule in which there
are no available sites on the metal ion to bind to a water molecule through coordinate
covalent bonds. The water can only be attracted to the oxygen atoms in the ligands
by hydrogen bonding as well as through a van der Waals attraction, both of which
are significantly weaker than are the coordinate covalent bonds that a water molecule
could have formed with the ferrous ion in a bisglycinate molecule.
The lower solubility of the ferric trisglycinate appears to affect overall intestinal
absorption of that iron amino acid chelate.17 In a study involving 21 human subjects,
3 mg of iron as 59ferrous bisglycinate, 59ferric trisglycinate, or 59ferrous sulfate were
administered to each person in a whole maize breakfast meal following a 12-hour
fast. On day 16 postdosing, blood samples were obtained from each subject and his or
her iron status measured from blood radioactivity. Mean iron absorption in the group
receiving the chelated bisglycinate iron source was 13.8% ± 3.4%, whereas the ferric
trisglycinate absorption was 2.3%. This was only slightly higher than absorption of
ferrous sulfate (2.2%). The absorptions of both chelate sources as well as the ferrous
sulfate were similarly regulated by the iron status of the body. Those individuals
with the lowest serum ferritin concentrations absorbed the greatest amount of iron
and vice versa, regardless of the iron source.17
Besides comparing the bioavailability of the bis- and trisglycinate iron che-
lates, the data in the study also suggest that a significant amount of the amino acid
chelate ingested from either chelate source was absorbed by diffusion as opposed
to active transport. If the two sources of amino acid chelates had been predomi-
nantly absorbed by active transport, then the percentage of iron absorption from
the iron bisglycinate and trisglycinate would have both been about the same since
the amino bonds between the iron and glycine molecules would have been broken
at the same pH regardless of whether the chelate was in a bis- or trisglycinate form.
The stability constant of the trisglycinate would certainly have been higher than was
the bisglycinate, but the pH at which the amino bonds would have been affected
would have been the same for both chelates. If they were similarly broken, then
they would have been similarly attached to the same active transport molecules and
thus would have been absorbed at approximately the same rates. Since the absorp-
tion of the two chelated sources was significantly different (p < 0.05), it points to the
effect of chelate solubility on absorption and ultimately the rate of facilitated diffu-
sion into the mucosal tissue. Oxidizing the ferrous iron to the ferric state and adding
a third amino acid to the iron chelate molecule resulted in significantly reducing
the overall solubility of that chelate. Similarly, salts formed with ferric iron are less
soluble than are salts formed with ferrous iron for the same reasons that trisiron
amino acid chelates are less soluble than is the bis form.18
Besides the valence of the metal ion, the choice of the amino acid employed as
a ligand will also affect the solubility of the resulting chelate. For example, just as
glycine is more soluble than methionine,18 zinc glycinate is more soluble than zinc
methionate.19 Thus, one would predict that more zinc glycinate would diffuse through
the intestinal cell membrane in a given period of time compared to zinc methionate if
everything else were equivalent. While direct in vitro diffusion studies have not been
conducted, the research20 discussed next tends to support this supposition.
A group of 16 zinc-sufficient adult male Sprague-Dawley albino rats was divided
into two groups consisting of 8 animals each.21 One was labeled the glycinate group
and the other the methionate group. Two zinc amino acid chelates were produced,
one with a glycine ligand and the other with a methionine ligand. Each chelate had a
molar ratio of two moles of amino acid to each mole of zinc. The zinc in each chelate
was tagged with 65Zn.
Following a 12-hour fast, all of the rats in both groups were anesthetized with
ether. Each animal was then orally administered a single dose of 20 mCi of 65Zn as
either the glycine or methionine chelate mixed with 95 mL of distilled and deionized
water. Each animal received the same amount of zinc in its dose. The only difference
was that the animals in the glycine group were administered zinc glycinate, whereas
zinc methionate was given to the rats in the methionine group. The zinc amino acid
chelate/water mixture was prepared immediately prior to administration, and no
attempt was made to ensure the dissolution of either chelate in the water.
The amino acid chelate/water mixtures were injected directly into the duode-
num via curved balled needles attached to syringes. This method of administra-
tion avoided the acid pH of the stomach, which reduced the chances of much of the
chelates being converted to chelate/complexes and subsequently absorbed by active
transport. At 4 hours postdosing, and while animals were still anesthetized, 15 mL
of blood was obtained from each animal by suborbital bleeding. The blood from
each animal was transferred to a stainless steel planchet, dried for 12 hours, and then
analyzed for 65Zn. Following obtaining the blood samples, all of the animals were
sacrificed and their tissues and organs harvested and also analyzed for 65Zn.
Although there were different percentage increases of 65Zn from the two chelate
sources depending on the tissues or organs analyzed, the differences all tended to
reflect what was also observed in the blood. At 4 hours postdosing, there was a sig-
nificant difference in the mean amount (p < 0.05) of 65Zn in the blood of the animals
administered the zinc glycinate compared to those receiving the zinc methionate.
This suggested a more rapid uptake of the zinc glycinate, which was possibly due to
this amino acid chelate being more soluble than the zinc methionate.18 While a small
portion of the zinc from either chelate may have been absorbed by active transport in
the intestines during that 4-hour period, nevertheless the majority of the two chelates
was probably absorbed by facilitated diffusion. In the case of facilitated diffusion, the
degree of solubility of the amino acid chelate appeared to influence how much of that
chelate molecule was potentially taken up during the 4-hour time period. Thus, it was
concluded that generally, in a given finite period of time, a more soluble amino acid
chelate will be absorbed in greater quantities by facilitated diffusion than will a less
soluble amino acid chelate, particularly if the pH of the lumen is not generally low
enough to break the amino bonds in the chelate and initiate active transport.
This study is not meant to suggest that chelated zinc glycinate is a superior source
of nutritional zinc when compared to chelated zinc methionate. It simply points out
that, in the case of facilitated diffusion, a more soluble amino acid chelate appears
to be taken up into the blood more rapidly compared to a less soluble source. Both
sources were absorbed in significantly greater quantities than was a highly soluble
inorganic 65Zn salt source, used in other similar bioavailability studies.21 Further,
other zinc absorption studies elaborated elsewhere in this book have demonstrated
that, regardless of the ligand employed, the uptake of zinc amino acid chelate will
continue past the 4-hour cutoff period used in this study, which would have allowed
for continued absorption of zinc from both chelate sources.
Returning, then, to the point that the amino acid chelate is dissolved in the bulk
water phase while in the lumen, the subsequent rate of movement of the chelate
across the unstirred water layer barrier is dictated by the thickness of that unstirred
water layer, the aqueous diffusion constant of the amino acid chelate, and the amino
acid or small-peptide concentration gradient between the bulk water phase and the
cell membrane. The aqueous diffusion constant is dependent on the solubility of the
amino acid chelate, whereas the amino acid concentration gradient is a function of
the concentration of the amino acid in the mucosal cell that corresponds to the amino
acid in the ligand of the chelate. It is essential to understand that the unstirred water
layer in and of itself forms a significant barrier to the diffusion of the amino acid
chelate. The higher the concentration of the amino acid chelate (or amino acid/small
peptides) in the bulk water phase compared to the amino acid chelate (or amino acid/
small peptide) concentration in the mucosal cell cytoplasm, the lower the resistance
to diffusion and the greater the flow of the amino acid chelate (or amino acid/peptide)
into the mucosal cell will be. Further, as seen in Figure 9.3, the effective thickness of
the unstirred water layer varies from about 100 to 500 mm.5,22 Obviously, a thinner
unstirred water layer will present less resistance to the diffusion of the solubilized
amino acid chelate when compared to a thicker layer.
Once the amino acid chelate has crossed the unstirred water layer, there are sev-
eral pathways available for its diffusion into the mucosal cell. One pathway bypasses
the mucosal cell altogether and allows the amino acid chelate to be absorbed directly
through the mucosal floor into the plasma. In the intestinal mucosa, the individual
absorptive cells are rather loosely attached to each other via tight junctions located
near the lumen. They are also connected by desmosomes found between the tight
junction and the mucosal floor.23 This is seen in Figure 9.5. This type of structure
leaves a significant amount of space filled with water between the mucosal cells. As
a result of these extracellular spaces, much of the net transfer of water into the body
bypasses the mucosal cells and is diffused directly into the mucosal floor to which
the mucosal cell is attached.5
The extracellular space between the mucosal cells that is occupied by water cre-
ates local differences in hydrostatic and osmotic pressures that can pull water and
its solutes into the paracellular spaces between the absorptive cells. The activation
energies of large nonelectrolytes across these tight junctions are similar to those
observed in free solutions, which suggests that the permeation pathway across the
tight junction is highly hydrated.5,24 Since dissolved metal salts (ions) can also be
absorbed via this pathway, it seems reasonable to suppose that this direct pathway
through the mucosal floor could be similarly shared by dissolved amino acid chelates
that are carried in the water as solutes.25 Further, if the dissolved amino acid chelate
is able to diffuse through the tight junction into the fluid between the cells, in addi-
tion to being absorbed through the mucosal floor, it could be taken into the mucosal
cells through the paracellular membranes.
The paracellular membranes between the absorptive cells have low electrical
resistance, high osmotic permeability, isotonic fluid transport, and only shallow
gradients built up by active solute transport.5 Thus, the cell membranes of the two
adjacent absorptive cells below the tight junction appear to be sufficiently perme-
able to allow amino acid chelate molecules that are in solution to diffuse into the
Glycocalyx
Dense plaque
Apical plasma
membrane
Central actin
filaments
Terminal Tight junction

web Intermediate
junction
Lateral plasma
membrane
Desmosome
FIGURE 9.5 Schematic diagram of the structural features of the apical plasma membrane,
the apical cytoplasm, and the junctional complexes of intestinal absorptive cells. (Redrawn
from Trier, JS, and Madara, JL, “Functional morphology of the mucosa of the small intestine,”
in Johnson LR, ed., Physiology of the Gastrointestinal Tract (New York: Raven) 930, 1981.
mucosal cells along with their diffused water carriers via this paracellular pathway.
The amount of uptake into the mucosal cell by this pathway would be largely depen-
dent on the solubility of the amino acid chelate. The more soluble the amino acid
chelate is, the greater the potential for its paracellular diffusion will be.
Facilitated diffusion of amino acid chelates can also occur at the brush border
of the mucosal cell. Intestinal water absorption is linked to the transfer of solutes
across the absorptive cell membrane, including dissolved amino acid chelates. As
previously described, the transfer of the amino acid chelate solute is a result of a
change in the osmotic pressure on the two sides of the mucosal cell membrane. This
requires a hyperosmolar region in the tissue of the villus that in effect “pulls” the
water and the solute it is carrying into the mucosal cell. Hyperosmolality is created
by a countercurrent exchanger in the vascular system located within the villus. The
dense system of capillary vessels in the villus surrounds the main ascending arterial
vessel. As the blood flows up the arterial vessel and down the capillary vessels, it
produces a counterflow that creates osmotic pressure because the linear blood flow
rate in the subepithelial capillaries is about a tenth that of the central artery.26 Water,
along with its amino acid chelate solute, can be drawn into the mucosal cells by this
pressure difference. The amino acid chelate may be released from the water mole-
cule following uptake into the mucosal cell leaving the water free to be transferred to
the capillary system in an attempt to equalize or reduce the osmotic pressure created
by the countercurrent of the vascular system within the villus.26 Or, it may remain in
solution and be transferred directly to the plasma. The potential fates of the absorbed
amino acid chelate are discussed in subsequent chapters.
In the following example, facilitated diffusion through the membrane of the

mucosal cell is accomplished by traversing the lipid bilayer of the membrane, which
is composed of amphilic lipids.27 These are membrane lipids that contain both
hydrophilic (OH groups, NH2 groups, phosphate groups, and carboxyl groups) and
hydrophobic (hydrocarbon chains) regions. The lipids and proteins are bound by
electrostatic bonds, hydrogen bonds, and hydrophobic bonds.27
The membrane permeability to hydrophilic compounds, including amino acid
chelate solutes, depends to a great degree on the concentration of H+ and Ca+2 ions
in the unstirred water layer. Depending on their ratio to each other, these ions can
increase or decrease membrane permeability.28 This interaction between the Ca+2
and H+ is in opposition to each other. If the Ca+2 is removed by exchanging H+ ions
from the acid pH of the unstirred water layer, the membrane permeability is sig-
nificantly increased with a resulting increased diffusion of low molecular weight
organic compounds, such as solubilized amino acid chelates across the membrane.29
The permeability of the membrane to hydrophilic molecules, including amino
acid chelate solutes, also depends, to a degree, on the fatty acid component in the
lipid bilayer. Long saturated hydrocarbon chains reduce its permeability, whereas
unsaturated ones loosen the structure and increase permeability.30 There is a direct
relationship between an increased proportion of steroids and reduced membrane per-
meability. In this case, the phosphate group of the phospholipids may be bound to
the NH3+ group in the proteins by electrostatic forces. Furthermore, the phosphate
group can be bridged by Ca2+ to a carboxyl group in the proteins. As noted, in the
acid pH of the unstirred water layer between the microvilli, the Ca+2 can be replaced
by H+ ions, which will result in this bond being broken. When that happens, the per-
meability of the membrane will be increased.31
In addition to the membrane lipid bilayer, there are protein molecules that are
embedded within this bilayer that are known as integral proteins. They generally
protrude from both sides of the membrane. The protein found within the membrane
is hydrophobic, whereas the protein moieties that extend past the membrane lay-
ers are hydrophilic.32,33 Many of the integral proteins are enzyme systems3 that are
capable of transferring the amino acid chelate/complex into the mucosal cells by
active transport as described in Chapter 8. Other integral proteins may form pores in
the membrane that facilitate diffusion of amino acid chelates via the pore pathway
that they have created.27
These pores are composed of polypeptide chains that extend through the lipid
bilayer. The globular proteins forming the pore are each anchored to the membrane by
20 to 28 hydrophobic amino acids. The terminal ends of the protein that extend into the
cytoplasm contain either lysine or arginine. More than one protein chain is involved
in forming a pore, as seen in Figure 9.6, which is based on the structural analysis of
pores in a bacteriorhodopsin.34 Each of the protein rods is in the form of a helix with
the backbone of the chain coiled tightly into an α-helix. In this way, each rod forms
a cylinder that is stabilized by bonds between the NH and CO groups of its backbone
chain. This occurs between the successive turns of the α-helix. The exception is where
the helices are linked together. There, the stabilizing bonding may not occur.34–36
As illustrated in Figure 9.6, several of the amino acids on each end of the seven
rods that form the pore carry either a positive or a negative charge on them. These
The Absorption of Amino Acid Chelates by Facilitated Diffusion
A B C D E F G
ALA GLU GLY
GLU ALA PHE
ALA ALA
ASP ALA ILE
STR THR GLY
PRO SER MET PRO ALA
ASP ASP
GLU SER
7 VAL ARG GLU GLY
ARG
Inside GLY
ALA ALA
PRO PRO ALA SER
6 LYS ASP GLU ??
5 MET ALA
GLY
LYS ??
GLU
??
GLY LYS ASP SER LEU
LYS THR VAL
PHE VAL
3 VAL
LEU ILE THR LEU
1 PHE
TYR
ALA
LEU
LEU
PHE SER
ALA ILE
??
GLY
2 4 LEU
TYR
ILE ALA
LEU
ILE
ALA
PHE
THR PHE
PHE
GLY
THR THR LEU VAL LYS
ASP PHE GLY
GLY THR LEU GLY VAL LEU VAL
LEU ALA LEU
LEU VAL VAL ARG LYS
LEU ASP
GLY PRO TYR ASN ALA
LEU GLY SER
MET ALA LEU LEU VAL THR
LEU VAL
LEU ILE MET ILE
PRO VAL ASP
ALA ALA THR TYR
THR GLY SER VAL ?? LEU
THR PHE THR MET
PHE TRP VAL
GLY THR GLY ALA
LEU SER MET
MET LEU ALA
LEU TRP ALA PHE
ALA TYR ASP VAL THR TYR LEU
LEU GLY PRO ALA
LEU ALA SER
SER TYR ALA VAL THR
TRP ARG ILE ALA
MET LEU VAL GLU
ILE ALA THR TRP ILE
LEU ??
TRP LEU TRP LYS VAL ASN
PHE ??
GLU GLY TYR VAL LEU PRO
PRO ILE ARG ILE
TYR VAL
PRO TYR GLY
GLY ??
LEU ASN TYR SER ??
GLY SER GLY
Outside ARG THR THR
GLN
GLU GLU GLY ALA
ILE MET
VAL GLY
GLN ALA GLU PRO GLY
PHE
FIGURE 9.6 The structure of a membrane pore based on the analysis of a bacteriorhodopsin. (Redrawn from Warren, G, “Membrane proteins:
Structure and assembly,” in Finean, J, and Michell, R, eds., Membrane Structure (Amsterdam: Elsevier) 213–257, 1981.)
145
FIGURE 9.7 The opening and closing of a pore in the membrane of a mucosal cell.
(Redrawn from Warren, G, “Membrane proteins: Structure and assembly,” in Finean, J,
and Michell, R, eds., Membrane Structure (Amsterdam: Elsevier) 213–257, 1981; Unwin,
PNT, and Zampighi, G., “Structure of the junction between communicating cells,” Nature
283:545–549, 1980; and Flag-Newton, J, and Loewenstein, W, “Asymmetrically permeable
membrane channels in cell junction,” Science 207:771–773, 1980.)
charges allow each of the rods to change its angle in relation to the other rods as they
come in contact with other charged molecules that break the bonds in the integral
protein rods or in other situations, either neutralizing or enhancing the force between
the charged amino acids. When any of these events occur, the angle of the protein
rods shifts position, which then allows the passage of soluble or suspended molecules
through the pore from the lumen into the mucosal cell. As the molecules arrive in the
cytoplasm, they affect the charged moieties at the opposite end of the rods in reverse,
causing the rods to shift back to their original positions. This results in the pore
being closed again. This process is illustrated in Figure 9.7.34,37,38
In the previous chapter, it was noted that the lumen in the small intestine com-
mences to change from an acid pH in the duodenum to an alkaline pH beginning
in the proximal jejunum. If the alkalinity becomes high enough, the amino acid
chelate will reconfigure into an amino acid chelate/complex. The amino acid chelate/
complex has electronegative atoms (oxygens and nitrogen) in each amino acid ligand
as seen in Figure 9.8.2
O
NH2
O Zn O
H2N O
FIGURE 9.8 Zinc bisglycinate that is subjected to a highly alkaline pH.

Because of these electronegative atoms, the amino acid chelate/complex can be

attracted to the positively charged moieties that are part of both the peripheral and
integral proteins. If attracted to a peripheral protein that is not enzymatic in nature,
it is likely the chelate/complex molecule will remain there until it is either attracted
to a molecule with a stronger potential bonding constant or the peripheral protein
and amino acid chelate are jointly released. As noted, most peripheral proteins are
loosely held by an electrostatic attraction with the hydrophilic head of the mucosal
cell membrane.40 They are easily removed from the membrane. When this occurs,
these peripheral proteins may possibly serve as transport molecules for facilitated
diffusion of the amino acid chelate/complex molecule through a membrane pore.
There is a third possibility arising out of the release of the peripheral protein/
amino acid chelate/complex molecule. These peripheral proteins are bathed by the
unstirred water layer, which has an acid pH.40 In this acid pH environment of the
unstirred water layer, hydrogen ions could potentially be donated to the amino acid
chelate/complex molecule and allow it to reconfigure as a chelate. If that were the
case, its attraction to the peripheral protein would be lost. It could then be dissolved
and absorbed by a different pathway through facilitated diffusion. If the pH became
acidic enough in the unstirred water, the amino acid chelate could reconfigure again
as a chelate/complex, but this time it would carry a positive charge on its ligand. The
molecule could then be absorbed by active transport as described in Chapter 8.
The amino acid chelate/complex could be attracted through ion-dipole forces in its
ligand to a positively charged moiety on an integral protein. If that were to occur, the
amino acid chelate/complex may potentially be transferred through the mucosal cell
membrane by facilitated diffusion using the integral protein as the transport molecule.
In addition to water or protein molecules functioning as a carrier molecule for
facilitated diffusion, this form of diffusion can occur when certain other transport
molecules are employed. The attachment of the amino acid chelate to these other
transport molecules may be the result of a van der Waals attraction between the carrier
molecule and any of the amino acid ligands, or the bond may be a result of displacing
one of the waters of hydration on the metal ion with a different carrier molecule.41
If an amino acid ligand containing a charged polar side chain is employed in
forming the chelate, a bond between the amino acid chelate and the carrier molecule
may also possibly occur at the site of the side chain on that amino acid ligand. Amino
acids can be grouped into those with nonpolar side chains, those with uncharged
polar side chains, and finally those with charged polar side chains.42 To produce a
charge on the side chains of a particular amino acid, it requires a change in the pH
environment to which the amino acid is subjected. That requisite pH change is usu-
ally so low (acidic) that if an amino acid chelate ligand were to pick up a charge on
its side chain, the amino bond between the metal ion and the amine moiety of each
of its ligands would probably also be broken.2,42 If the amino bond were broken in
an acid environment, the resultant amino acid chelate/complex would probably be
absorbed into the mucosal cell by active transport rather than by facilitated diffu-
sion. The amino acid chelate/complex formed in a high-pH environment would prob-
ably be preferentially absorbed by some other facilitated diffusion pathway rather
than bonding the chelate/complex molecule to an amino acid side chain. Thus, facili-
tated diffusion of an amino acid chelate into the intestinal cell must generally focus
on a source of attraction between the amino acid chelate molecule and the transport
molecule that excludes the consideration of amino acid side chains as moieties to
bond to transport molecules.43
To illustrate one such alternative, one potential way amino acids can be absorbed
is by bonding them to sodium ions rather than to water molecules. There are four
different sodium-dependent facilitated diffusion transport systems employed for the
uptake of amino acids.44 Each is specific for uptake of a certain group of amino
acids (i.e., acidic, basic, neutral, and specific amino acids). In spite of their specific-
ity, each of these transport systems requires the cotransport of sodium and utilizes a
sodium-potassium pump to move the sodium transporter and the molecule it carries
(i.e., an amino acid or amino acid chelate or chelate/complex) across the mucosal
cell membrane.
Sodium and potassium ions are distributed along both sides of the mucosal cell
membrane in such a way that it causes the fluid inside the cell to have a slightly lower
concentration of positively charged ions compared to the extracellular fluid.45 This
results in a voltage difference across the membrane. Thus, if one were able to poke a
hole in the membrane, it would result in the flow of negatively charged anions out of
the cell in an attempt to neutralize the charge difference. In the living cell, however,
that ion flow is controlled by channels in the membrane that mediate the passage of
specific ions, such as sodium.46 An amino acid or amino acid chelate can be bonded
to the Na+-dependent transport system via a hydroxyl group.40 As the sodium enters
the cell, its transport system concurrently pulls the amino acid or amino acid chelate
that is attached to the sodium through the membrane.
It should be recognized that the sodium pump was originally hypothesized as a
mechanism for the active transport of certain organic molecules.47 That hypothesis
was based on the fact that “uphill” transport of affected substrates depended on the
difference in sodium concentration in the cytoplasm of the cell and its environment.
Thus, the driving force for the uphill transport was believed to be the sodium concen-
tration gradient. Since the presentation of Crane’s original work on the sodium pump
in 1962, additional research has indentified at least two types of sodium pumps.48
Goldner et al. reported that it has been clearly demonstrated that mucosal-to-serosal
water movement increases the permeability of the intestinal epithelium. Therefore,
they stated that the sodium in the mucosal solution may either directly couple to the
movement of solutes or it may indirectly increase the permeability of the intestine to
the substrate because the movement of the water is linked to the sodium movement,
and the water movement changes the structure of the epithelial cell membranes.48
From the discussion, one could conclude that the sodium potassium pump could be
involved in both active transport, if energy were involved, and facilitated diffusion of
amino acids and possibly amino acid chelates if no energy is expended.49
At the conclusion of an international conference on intestinal permeation, the
chair, D. S. Parsons, summarized the meeting by stating that there was more than
one mode of entry for amino acids and, by implication, amino acid chelates into the
intestinal cell. Both active transport systems and diffusion pathways are concur-
rently in play for the mucosal cell uptake of these chelate molecules. The accumula-
tion of the majority of these nutrients in the villus of the intestine depends on blood
flow. This is not true in the case of active transport, but when substrate saturation in
the mucosal cell is reached, facilitated diffusion, which appears to be the main path-
way for uptake of amino acid chelates, levels off until the blood is able to pick up and
carry the excess nutrient (in the mucosal cell) into the various tissues and organs of
the body.50 Depending on the life span of the mucosal cell in question, removal of the
saturated nutrient may or may not occur before the migrating mucosal cell arrives at
the tip of the villus and is sloughed off. Further, as discussed in a subsequent chapter,
if the body had achieved sufficiency of a particular amino acid chelate, there will
be less of that molecule or its individual components picked up by the blood from
the mucosal cell and translocated to body tissues compared to when the body has
an acute need for the nutrient. Thus, the efficient function of a facilitated diffusion
process is somewhat dependent on the need of the body for that chelate.
In that same concluding lecture, Parsons also noted that the unstirred water layer
adjacent to the mucosal cell membrane plays a significant role in the uptake of the
nutrients. In particular, besides functioning as a barrier, it will provide a transport
system which facilitates the diffusion of nutrients into and out of the mucosal cell.
Further, it is essential for the facilitated diffusion of nutrients between the mucosal
cells via the paracellular route.50
He also emphasized that there is compartmentalization within the mucosal cells
in respect to uptake of nutrients as well as their subsequent metabolism. This is dis-
cussed in greater detail in the chapters that follow. As far as transport is concerned,
there is at least one active transport system and several facilitated diffusion transport
systems that move amino acid chelates from the lumen into the mucosal cells. These
pathways include different types of pores, different enzyme systems as part of an
integral protein, or even facilitating absorption directly through the lipid bilayer of
the mucosal cell membrane.50
Finally, Professor Parsons reported that not all nutrients that are absorbed from
the intestinal lumen “are accompanied by an ion. Some are packaged in the intestinal
lumen before being absorbed by the mucosa, [and] others during the absorption pro-
cess.”50 Under certain circumstances, this packaging process could include amino
acid chelates that have been bonded to both a nonionic active transport system or to
certain facilitated diffusion carrier molecules. The amino acid chelate can be modi-
fied by changes in pH, become a chelate/complex, and be absorbed enzymatically by
active transport,2 or the amino acid chelate can retain its original chelate configura-
tion and be taken into the mucosal cell by any one of the several facilitated diffusion
pathways described.
REFERENCES
1. Hunt, SM, and Groff, JL, Advanced Nutrition and Human Metabolism (St. Paul, MN:
West) 41, 1990.
chelates by the utilization of ninhydrin,” AOAC poster, St. Louis, MO, September, 2004.
4. Stein, WD, “Concepts of mediated transport,” in Bonting, SL, and dePont, J, eds.,
Membrane Transport (Amsterdam: Elsevier) 143, 1981.
5. Thomson, ABR, and Dietschy, JM, “Intestinal lipid absorption: Major extracellular and
intracellular events,” in Johnson, LR, ed., Physiology of the Gastrointestinal Tract (New
York: Raven Press) v2, 1151–1158, 1981.
7. Roy, CN, and Enus, CA, “Iron homeostasis: New tales from the crypt,” Blood
96:4020–4027, 2000.
8. MacInnis, A, and Graff, DJ, “Specificity of amino acid transport in the tape worm,
Hymemolepsis diminuta, and its rat host,” Rice Univ Studies 62:183, 1977.
10. Mazariegos, DI, Pizarro F, Olivares, M, Nunez, MT, and Arredondo, M, “The mecha-
nism for regulating absorption of Fe bis-glycine chelate and Fe-ascorbate in Caco-2
cells are similar,” J Nutr 134:395–398, 2004.
11. Burston, D, Marrs, TC, Sleisenger, MH, Sopanent, T, and Matthews, DM, “Mechanisms
of peptide transport,” in Elliott, K, and O’Connor, M, eds., Peptide Transport and
Hydrolysis (Amsterdam: Elsevier) 79–108, 1977.
12. Thomson, ABR, and Dietschy, JM, “Derivation of the equations that describe the effects
of unstirred water layers on the kinetic parameters of active transport process in the
intestine,” J Theor Biol 64:277–294, 1977.
ed., Medical Physiology (St. Louis, MO: Mosby) 1145–1147, 1974.
14. Seager, SL, and Slabaugh, MR, Introductory Chemistry (Glenview, IL: Scott Foresman)
134, 1979.
15. Sanjit, K, Gagnon, K, Clearfield, A, Thompson, C, Hartle, J, Ericson, C, and Nelson, C,
16. Ashmead, HD, “The solubility of ferrous bisglycinate,” unpublished research report,
Albion Laboratories, Clearfield, UT, 2002.
17. Bovell-Benjamin, AC, Viteri, F, and Allen, LN, “Iron absorption from ferrous bisglyci-
nate and ferric trisglycinate in whole maize is regulated by iron status,” Am J Clin Nutr
71:1563–1569, 2000.
18. O’Neil, MJ, Smith, A, and Heckelman, eds., The Merck Index (Whitehouse Station, NJ)
2001.
19. Ashmead, HD, “Solubility of zinc methionate and zinc glycinate,” unpublished research
report, Albion Laboratories, Clearfield, UT, 2010.
20. Peck, A, and Graff, DJ, “Absorption of 65Zn in chelated form into the blood of rats,”
unpublished study, Weber State University, Ogden, UT, 1973.
amino acid chelates and inorganic metal salts,” in Subramanian, KD, Iyengar, GV, and
22. Westergaard, H, and Dietschy, JM, “Delineation of dimensions and permeability char-
acteristics of the two major diffusion barriers to passive mucosal uptake in rabbit intes-
tine,” J Clin Invest 54:718–732, 1974.
23. Trier, JS, and Madara, JL, “Functional morphology of the mucosa of the small intes-
tine,” in Johnson, LR, ed., Physiology of the Gastrointestinal Tract (New York: Raven)
930, 1981.
24. Smulders, AP, and Wright EM, “The magnitude of nonelectrolyte selectivity in the gall-
bladder epithelium,” J Membr Biol 5:297–318, 1971.
25. Frizzell, RA, and Schultz, SG, “Ionic conductances of extracellular shunt pathway in
rabbit ileum: Influence of shunt on transmural Na transport and electrical potential dif-
ferences,” J Gen Physiol 59:318–346, 1972.
26. Boyd, CAR, “Vascular flow and the compartmental distribution of transported sol-
utes within the small intestinal wall,” in Kramer, M, and Lauterbach, F, eds., Intestinal
Permeation (Amsterdam: Excerpta Medica) 41–47, 1977.
27. Ussing, HH, “Ion transport across biological membranes,” in Clarke, HT, ed., Ion
Transport across Membranes (New York: Academic Press) 5, 1954.
28. Kavanau, JL, Structure and Function in Biological Membranes (San Francisco: Holden-
Day) v1, 132, 1965.
29. Mengle, RG, and Heald, WR, “Distribution of potassium, rubidium, cesium, calcium,
and strontium within plants grow in nutrient solutions,” Soil Sci 80:287–293, 1955.
30. Van Deenen, L, “Phospholipids. Relationship between their chemical structure and bio-
logical membranes,” Naturwissenschaften 59:485–491, 1972.
31. Mengel, K, and Kirkby, EA, Principles of Plant Nutrition (Bern, Switzerland:
International Potash Institute) 114, 1982.
32. Singer, SJ, “A fluid lipid-globular protein mosaic model of membrane structure,” Ann
NY Acad Sci 195:16–23, 1972.
33. Lodish, HF, and Rothman, JD, “The assembly of cell membranes,” Sci Am 240:38–53,
1979.
34. Warren, G, “Membrane proteins: Structure and assembly,” in Finean, J, and Michell, R,
eds., Membrane Structure (Amsterdam: Elsevier) 213–257, 1981.
35. Unwin, N, and Henderson, R, “The structure of proteins in biological membrane,” Sci
Am 250:78–94, February 1984.
36. Engelman, DM, Henderson, R, McLachlan, AD, and Walace, BA, “Path of the polypep-
tide in bacteriorhodopsin,” Proc Natl Acad Sci USA 77:2023–2027, 1980.
37. Unwin, PNT, and Zampighi, G., “Structure of the junction between communicating
cells,” Nature 283:545–549, 1980.
38. Flag-Newton, J, and Loewenstein, W, “Asymmetrically permeable membrane channels
in cell junction,” Science 207:771–773, 1980.
39. Berliner, LJ, “The spin-label approach to labeling membrane protein sulfhydryl groups,”
Ann NY Acad Sci 44:153–161, 1983.
40. deRobertis, EDP, Saez, FA, and deRobertis, EMF, Cell Biology (Philadelphia: Saunders)
158–166, 1975.
41. Seager, SL, and Slabaugh, MR, Introductory Chemistry (Glenview, IL: Scott Foresman)
75, 1979.
42. Voet, D, and Boet, JG, Biochemistry (Hoboken, NJ: Wiley) 67–68, 2004.
43. Munc, BG, “Intestinal transport of amino acids,” in Kramer, M, and Lauterbach, F, eds.,
Intestinal Permeation (Amsterdam: Excepta Medica)126, 1977.
45. Caspary, WF, “Mechanism and specificity of intestinal sugar transport,” in Kramer, M,
and Lauterbach, F, eds., Intestinal Permeation (Amsterdam: Excepta Medica) 75–83,
1977.
47. Crane, RK, “Hypothesis for mechanism of intestinal active transport at sugars,” Fed
Proc 21:891–895, 1962.
48. Goldner, AM, Sarracino, S, and Estep, JA, “Coupling of sodium and substrate fluxes in
the intestine: Galactose transport in the jejunum,” in Kramer, M, and Lauterbach, F, eds.,
Intestinal Permeation, 282, 1977.
49. Eddy, AA, “Current status of the ion gradient hypothesis in some selected systems,”
in Kramer, M, and Lauterbach, F, eds., Intestinal Permeation (Amsterdam: Excepta
Medica) 332–349, 1977.
50. Parsons, DS, “Closing summary,” in Kramer, M, and Lauterbach, F, eds., Intestinal
Permeation (Amsterdam: Excepta Medica) 432, 1976.
10 The Fate of Amino
Acid Chelates in
the Mucosal Cell
After an amino acid chelate or an amino acid chelate/complex has been taken up by the
absorptive cell, is it transferred directly to the plasma as the intact molecule that was
absorbed or is it hydrolyzed into a metal ion and free amino acids in the mucosal cell
prior to sending the mineral on to the plasma? To help answer this question, a study
was designed to simulate the fate of an iron amino acid chelate following its absorption
into the mucosal tissue and subsequent transfer to the serosa.1 Using 14C-tagged amino
acids and 59iron, this in vitro study allowed investigators to observe potential mucosal
cell hydrolysis and the rates of deliveries of the ligand and metal to the serosa.
Six nonfasted, iron-sufficient adult male Sprague-Dawley rats were used for each
portion of this two-part study. The animals were individually housed in identical
stainless steel cages. All were fed a commercial diet that met National Research
Council (NRC) dietary requirements, including iron.2 At the time of sacrifice, all of
the animals were approximately the same age and weight plus or minus 15 g.
After anesthetization with ether, an incision was made at the midline of each rat
and its intestines exposed. The ileum was located and excised beginning at 5 cm
proximal to the ileocecal valve and continuing for 15 cm distal. Following removal
of that portion of the small intestine, each rat was sacrificed by cervical dislocation
while still anesthetized.
Each ileum segment was everted and subsequently washed three times in Krebs-
Ringer bicarbonate buffer (pH 7.4) solution to remove the luminal contents. Both ends
of the everted intestinal segments were tied with 000 silk thread to form intestinal loops.
During the preparation of the intestinal loops, the segments were maintained in
beakers containing a Krebs-Ringer bicarbonate solution into which an oxygen and
carbon dioxide gas mixture (95%:5% concentration) was bubbled. Once a loop was
prepared, it was transferred into a vial containing mucosal solution (pH 7.1), and that
vial immediately was immersed in a preheated water bath maintained at a constant
temperature of 37°C. The mucosal solution in the vial bathed the luminal sides of
the everted intestinal loop. Isotonic serosal solution was then injected into the loop
by using a syringe pump. The same oxygen and carbon dioxide (95%:5%) gas mix-
ture used during the intestinal loop preparation continued to be bubbled through the
mucosal solution into the vials containing the intestinal loops (Figure 10.1).
This investigation was designed to study the fates of the absorbed amino acid
chelate as it moved from the mucosal solution through the mucosal tissue into the
serosal solution. Radiolabeling of both the iron and the chelating ligands allowed
153
CO2:O2 Gas flow Serosal Solution in 50cc Syringe
Mucosal
solution
Syringe Pump
Rat
intestine
Lead
shield
Temp.
probe
Water bath
Fraction Collectors
Magnetic stirrer
temp. controlled
FIGURE 10.1 Laboratory setup to measure absorption and hydrolyzation of 59iron

14C-amino acid chelate in rat ileum. (From Ashmead, HD, Graff, DJ, and Ashmead, HH,
Intestinal Absorption of Metal Ions and Chelates (Springfield, IL: Thomas) 205–212, 1985.
With permission.)
segregated observation of the fate of the chelated iron source as well as its several
components in the mucosal and serosal solutions and in the mucosal tissue.
Two amino acid chelates were prepared. The first consisted of 59Fe chelated to
14C-lysine. Quantifications of the iron and its chelating ligands were determined by
liquid scintillation. The ratio of iron to the lysine in the chelate was 1:2.7 (metal
to amino acid) based on corrected counts per minute of 59Fe and 14C. The second
amino acid chelate also used 59Fe as the metal, but in this case, it was chelated to
14C-methionine. The ratio of iron to methionine was 1:1.45 (metal to amino acid),
also based on corrected counts per minute of 59Fe and 14C. On a molar basis, both
chelates had a ratio of iron to either amino acid of 1:2 (metal to amino acid).
Then, 10 mL of either chelate were added to each liter of mucosal solution and the
appropriate amount of mucosal solution transferred to each vial containing an intes-
tinal loop. A liquid scintillation analysis for 59Fe and 14C was conducted on the muco-
sal solutions prior to their introduction into the vials containing the intestinal loops.
As described, isotonic serosal solution was injected into each everted intestinal
loop using a 50-cc syringe connected to a syringe pump. Hourly, that serosal solution
was removed from each intestinal loop via the fraction collectors beginning at 1 hour
“postdosing” and concluding 5 hours later. As the serosal perfusate was collected, it
was replaced by fresh isotonic serosal solution. Each serosal solution that was col-
lected was separately assayed for 59Fe and 14C by liquid scintillation. A sample of the
isotonic serosal solution was also obtained at the study commencement and analyzed
by liquid scintillation to verify that it contained no 59Fe and 14C. The data obtained
from analyzing the initial isotonic serosal solution was not included in any statistical
analysis of the analytical results since it represented zero.
The Fate of Amino Acid Chelates in the Mucosal Cell 155
TABLE 10.1
Ratios of 59Iron Chelated to 14Carbon Lysine in Mucosal
Solution, Mucosal Tissue, and Serosal Solution
Mucosal Solution Mucosal Tissue Serosal Solution
1:2.7 Proximal 1:2.7 Hour 1 1:1.5
Middle N/A Hour 2 1:2.7
Distal 1:3.4 Hour 3 1:1.3
Hour 4 1:1.2
Hour 5 1:1.1
Mean 1:2.70a 1:3.05b 1:1.56b
Standard deviation ± ±0.49 ±0.65
Source: Based on data from Ashmead, HD, Graff, DJ, and Ashmead, HH,
Intestinal Absorption of Metal Ions and Chelates (Springfield, IL:
Thomas) 209, 1985.
a/b p < 0.05.
At the conclusion of taking the last serosal solution sample, each intestinal loop
was divided into two equal parts for the first study and three equal parts (the proxi-
mal, the middle, and the distal sections) for the second study. Each of the segments
from each intestinal loop was weighed and then digested with Soluene-350 before
being analyzed for 59Fe and 14C by liquid scintillation.
All of the data generated from the analysis of the six intestinal loop segments and
the hourly collections of the six serosal solutions were statistically analyzed using
t tests to determine significance. Table 10.1 summarizes the results of the first por-
tion of that experiment. The data are expressed in ratios of 59Fe to 14C-lysine.
In the first study, when the iron was chelated to the amino acid lysine, the
ratio of 59Fe to 14C in the ligand, based on corrected counts per minute, was 1:2.7
(59Fe to 14C-lysine). As the chelate was taken up into the mucosal tissue, presumably
by facilitated diffusion based on the pH of the mucosal solution, the mean ratio of
59Fe to 14C changed to 1:3.05, indicating that, on average, for every 59Fe atom the
mucosal tissue contained 13% more 14C-lysine than did the mucosal solution. The
serosal solution, on the other hand, contained approximately 42% less 14C-lysine
than was in the mucosal solution and about 49% less than in the mucosal tissue based
on the ratio of 59iron to 14C-lysine.
The difference between the concentration of 14C-lysine in the mucosal solution
compared to its mean concentration in the serosal solution was significant (p < 0.05).
An analysis of the data in Table 10.1 suggested that once the iron lysinate was taken
up by the mucosal tissue, much of the amino acid chelate appeared to be dissociated
into its individual components, iron and lysine, while in the mucosal cells. The two
components iron and lysine were then transferred to the serosal solution, but at dif-
ferent rates. During the study period, more iron appeared to be sent to the serosal
solution compared to lysine, based on changes in their 59Fe to 14C ratios. This is
indicated by a significantly (p < 0.05) larger ratio of 59Fe to 14C in the serosal solution
(1:1.56) compared to the mucosal solution (1:2.70). The transfer of lysine out of the
mucosal tissue to the serosal solution was slower than the transfer of iron.
At initial glance, the difference in the 14C-lysine between the proximal and distal
portions of the intestinal loops suggests that the iron lysinate did not remain evenly
dispersed throughout the mucosal solution and during the 6-hour study period.
Instead, it tended to “settle” toward the bottom of the vial containing the mucosal
solution. Iron lysinate is highly soluble, so this does not seem to be a reasonable
explanation. The 14C-lysine difference between the proximal and distal section was
not statistically significant, indicating that in statistical terms, there was no differ-
ence between the groups of segments.
Based on the ratio of 59Fe to 14C-lysine, the rates of transfer of the iron and the
lysine into the serosal solution from the mucosal tissue were relatively consistent
during the 5-hour analytical period except for the second hour, when over twice as
much lysine appeared to be transferred compared to the mean transfers during the
other measured periods. With the exception of the second hour, there were no other
significant differences in the hourly rates of lysine transfer from the mucosal tissue
to the serosal solution during the study period. The reason for the increased transfer
of lysine during the second hour is unknown.
It is improbable that all of the amino acid chelate that was absorbed into the
mucosal tissue was hydrolyzed into its individual components following its uptake.
Some of the amino acid chelate was probably transferred to the serosal solution as an
intact molecule. Pineda calculated that the maximum amount of original chelate that
potentially remained intact was approximately 46.8% of the absorbed amount.3 That
amount reflected the entire 5-hour period except for hour 2, which was considered
to be an outlier and not used in his calculations. Based on his calculations, approxi-
mately 55% of the iron lysinate in the serosal solution was in the form of an iron
lysinate in hour 1, 48% in hour 3, 44% in hour 4, and 40% in hour 5 with a mean
amount of 46.75% ± 6.40% for the entire period.
These calculations suggested that more of the absorbed iron lysinate was initially
transferred intact to the serosal solution following its uptake into the mucosal tissue,
but as time progressed, more of the absorbed amino acid chelate in the mucosal tis-
sue was either hydrolyzed due to longer exposure to intracellular enzymes, or due to
a “reduction in serosal need” for the chelate, it was simply retained in the mucosal
tissue for a longer period of time. After the initial transfer of what was presumably
mostly intact iron lysinate from the mucosal solution into the serosal solution, the
rate of cellular hydrolysis of the amino acid chelate increased in the mucosal tissue
as a way of controlling further movement of iron out of the mucosal tissue and into
the serosal solution. There may have been some inducement to continue to transfer
intact iron lysinate into the serosal solution due to hourly replacement of the serosal
solution, but the change in 59iron to 14C-lysine ratio indicated that mucosal hydrolysis
of the chelate also occurred. As discussed further in this chapter, to exert maximum
control on the rate of flow of iron from the mucosal tissue to the serosa, the iron ion
must have been removed from the amino acid ligand and bound to other intracellu-
lar molecules. These data demonstrated that mucosal cell hydrolysis of at least part
of the absorbed iron lysinate took place in the absorptive cells because the rates of
movement of the iron and lysine out of the mucosal tissue into the serosal solution
did not occur equally. The iron was transferred to the serosal solution more rapidly
than was the lysine. Had the study continued for a longer period of time, one may
have seen the ratio of 59iron to 14C-lysine in the serosal solution continue to move
closer to a 1:1 ratio as more lysine was transferred to the serosal solution and as the
transfer of iron became less as a result of increased regulation of the movement of
iron ions into the serosal solution.
While this theoretical description of the fate of the iron lysinate seems reason-
able, it should be remembered that no measurements were made to determine what
percentages of the iron and lysine existed in the form of intact iron amino acid che-
late molecules in the serosal solution. The percentage of iron amino acid chelate
that potentially remained intact in the mucosal tissue, as suggested, is only a calcu-
lated theoretical percentage. There may have been less actual transfer of intact iron
lysinate chelate molecules to the serosal solution than these calculations suggest.
One thing is certain: If a chelate of iron and lysine did enter the serosal solution as
intact molecules, the almost neutral pH of the serosal solution would have encour-
aged those iron lysinate molecules to configure in the form of a chelate just as the
mucosal solution did previously. But, the same could possibly be said of the free iron
and lysine if they entered the serosal solution as reactive moieties. They also may
have reacted together and re-formed in the serosal solution as newly created amino
acid chelate molecules.4
Not every amino acid chelate is hydrolyzed in the mucosal tissue at the same rate.
Table 10.2 summarizes the results of absorption, the probable mucosal tissue hydro-
lyzation of the 59Fe that was chelated with 14C-methionine, and the rates of transfer
of both isotope moieties to the serosal solution.1
In this second study, while the molar ratio of metal to ligand was 1:2, the ratio of
59Fe to 14C-methionine, based on corrected counts per minute, was 1:1.4 in the muco-
sal solution. In the mucosal tissue, following absorption of the chelate by diffusion,
TABLE 10.2
Mean Mucosal Tissue Uptake and Transfer to the
Serosal Solution of 59Iron Chelated to 14C-Methionine
Mucosal Solution Mucosal Tissue Serosal Solution
1:1.4 Proximal 1:4.7 Hour 1 N/A
Middle 1:4.9 Hour 2 1:7.9
Distal 1:4.5 Hour 3 1:10.5
Hour 4 1:11.5
Hour 5 1:10.2
Mean 1:1.4a 1:4.7b 1:10.05b
Standard deviation ± ±0.2 ±1.53
Source: Based on data from Ashmead, HD, Graff, DJ, and Ashmead, HH,
Intestinal Absorption of Metal Ions and Chelates (Springfield, IL:
Thomas) 208, 1985.
a/b The means are significantly different at p < 0.05.
that ratio changed to a mean ratio of 1:4.7, resulting in approximately 3.36 times
more 14C-methionine being recovered in the mucosal tissue per atom of 59Fe com-
pared to their ratios in the mucosal solution. This suggested that significant intracel-
lular hydrolysis of the absorbed amino acid chelate occurred following its absorption
and prior to the transfer of the iron and methionine to the serosal solution. Further,
the ratio of 59Fe to 14C-methionine in the mucosal tissue remained relatively constant
in the proximal, middle, and distal segments of the intestinal loops. There was, how-
ever, a significant difference (p < 0.05) in the amount of methionine in the mucosal
tissue when the ratio of 59Fe to 14C-methionine in the mucosal solution (1:1.14) was
compared to the ratio of 59Fe to 14C-methionine in the mucosal tissue (1:4.70).
Contrary to what was observed in the first portion of this study when the iron
lysinate was administered, the transfer of the 14C-methionine to the serosal solu-
tion seemed to be much more rapid than was the transfer of 14C-lysine in the pre-
vious study. Approximately 7.2 times more 14C-methionine was recovered in the
serosal solution compared to the amount of 14C-methionine in the mucosal solution
(p < 0.05), indicating not only rapid hydrolysis of the chelate within the mucosal tis-
sue but also possibly more complete hydrolysis of that particular absorbed chelate.
On a theoretical basis, all of the iron recovered in the serosal solution could have
potentially been chelated to methionine since there were adequate amounts of methi-
onine in the serosal solution to accomplish this. Unfortunately, there was no analysis
of the serosal solution to determine whether the recovered iron was chelated, so
either position, chelate or free iron in the serosal solution, could be advocated.
What is interesting to note from these two studies, however, is the difference
in the rates of probable hydrolysis of the two chelated iron sources following their
absorptions into the mucosal tissue. The only variable in this two-part study was
the source of the ligand. One iron chelate used lysine as the ligand, while the other
employed methionine. On a molar basis (not a corrected counts per minute basis),
every mole of iron was chelated to 2 moles of amino acid regardless of whether that
amino acid was lysine or methionine. The intestinal loops were sourced from healthy
and iron-adequate animals and were randomized prior to study initiation. Thus, an
explanation for the difference in the two studies appears to lie in the difference in the
inherent stability constants of the two amino acid chelates. Iron lysinate has a stabil-
ity constant of 9.010 compared to 6.4810 for iron methionine.5 An amino acid chelate
with a lower stability constant would potentially exhibit greater hydrolyzation in the
mucosal tissue following intact absorption than would an amino acid chelate with a
higher stability constant. Thus, based entirely on the stability constants of these, the
two amino acid chelates and the observations from these studies, cellular hydroly-
sis of chelates made with individual amino acids would theoretically progress from
weakest to the strongest as follows: arginine < methionine < phenylalanine < threo-
nine < valine < leucine < glutamic acid or serine < alanine < proline < glycine <
lysine < histidine < cysteine < asparagine < tyrosine.5
In these studies, most, if not all, of the iron lysinate and iron methionate were
probably absorbed by facilitated diffusion. The pH of the mucosal solution and of
the serosal solution, as well as of the mucosal tissue, all encouraged the formation or
continued stability of chelates rather than causing them to reconfigure as amino acid
chelates/complexes. Had the mucosal solution been more acidic, a chelate/complex
may have resulted. If that were the case, then more of the iron lysinate and methi-
onate may have entered the mucosal cells as chelate/complexes possibly following
active transport, and there may have been additional intracellular hydrolysis of the
absorbed molecules as a consequence.
In spite of the differences in the rates of mucosal cell hydrolysis of the amino
acid chelates made with different amino acids, other studies indicated that amino
acid chelates still appeared to be transferred from the lumen to the serosal tissue
more rapidly and in greater amounts when compared to the absorption and transfer
of metal ions under the same conditions. These different transfer rates suggest that
some of the amino acid chelate or chelate/complex that is absorbed into the mucosal
tissue is not hydrolyzed in that tissue. If complete hydrolysis of the chelate or chelate/
complex occurred, then the rate of transfer of the metal out of the mucosal tissue into
the serosa would be similar to the rate of transfer of a metal ion, as demonstrated by
the study discussed next.
In a double-isotope study of similar design to the two studies described previ-
ously, the affinity of zinc for ligands sourced from mucosal tissue was measured in
both the presence and the absence of histidine.6 The investigators determined that
when histidine was added to the mucosal solution bathing the intestinal loops, the
presence of the amino acid reduced zinc binding by mucosal-sourced ligands by a
factor of 3.7 times. Instead, the zinc ions in the mucosal solution were preferentially
bound to the histidine in the mucosal solution, resulting in the formation of zinc
histidine chelates.
This study demonstrated that following uptake of the zinc from either the ionic or
the chelated source, there was a different metabolic response to the two sources of
zinc in the mucosal tissue. The absorbed zinc histidine chelate was a much smaller
molecule compared to the absorbed zinc ions, which had to be complexed to ligands
sourced from the mucosal tissue prior to their absorption. Smaller molecules gen-
erally require less intracellular hydrolyzation compared to larger molecules. The
transport ligands, sourced from mucosal tissue, resulted in the formation of much
larger zinc-containing molecules, which required intracellular hydrolysis before the
zinc could be transferred to the serosa. The need for additional hydrolysis of the zinc
complex formed with ligands derived from mucosal tissue resulted in less zinc being
transferred to the serosa in a given amount of time compared to zinc histidine. The
investigators concluded that, due to the lower molecular weight of the amino acid
chelate, less intracellular hydrolysis of zinc histidine was required before transfer-
ring that form of zinc to the serosal solution, which resulted in a more rapid transfer.
This zinc study assumed that the zinc ions were complexed to transport molecules
acquired from the mucosal cell membrane and remained complexed to these same
transport molecules until the zinc was prepared to enter the serosal solution. While
this may not be entirely accurate as far as the site for intracellular hydrolysis is con-
cerned, the postulated conclusion is probably still valid.
The fate of the metal ion that remains attached to the amino acid ligand is the
subject of another chapter. The remainder of this chapter focuses on that portion of
the absorbed amino acid chelate/complex that is hydrolyzed within the mucosal cell.
It will be recalled from the previous discussion on stability constants that iron
ethylenediaminetetraacetic acid (EDTA) chelate, which has a stability constant of
13.7210, is readily taken up from the gastrointestinal tract, but most of that original
iron EDTA chelate can be recovered in the urine as an intact molecule.5,7 The stabil-
ity constant of that chelate is higher than the potential formation constant of most
ligands found within the cells of the body. Following its absorption, the EDTA will
generally release the iron to the cells only if it can exchange that metal for a different
metal on the electromotive scale that would potentially have a greater affinity for the
EDTA than the iron had.8 EDTA is not readily metabolized by the body, so following
absorption, it continues to function as an active ligand, particularly if it is induced to
release the iron to which it was originally chelated.
The potential formation constants of ligands produced within the cell often have
higher affinities for a given metal ion than the stability constants of certain absorbed
metal chelates/complexes. Thus, the amino acid chelate or chelate/complex ligand
that carries the metal into the cell can frequently be induced to release its metal to
another ligand that was produced intracellularly. When the body has an acute need
for the metal that is absorbed as part of the amino acid chelate or chelate/complex,
there appears to be less intracellular hydrolysis and greater direct transfer of the
intact molecule into the plasma, provided the mucosal cell is able to recognize the
absorbed amino acid chelate or chelate/complex as a metallic-based molecule and
not an amino acid/peptide molecule. In times of minimal need for the metal, more
of the amino acid chelate or chelate/complex will be induced to give up its metal to
an intracellular ligand while still in the mucosal cell. Unlike the EDTA ligand, when
intracellular hydrolysis takes place, the amino acid becomes free to enter into other
metabolic processes that require that particular amino acid. The metal ion, which the
amino acid ligand initially carried into the absorptive cell, is now free to participate
in the metabolic and biochemical roles assigned to any nonchelated metal ion that
has been absorbed into the mucosal cell.
It is not clear how the mucosal cell/body is able to recognize the metal ion
within the amino acid chelate or chelate/complex molecule following its absorption.
Mucosal cell hydrolysis of an amino acid chelate or chelate/complex may simply be
a natural response to the uptake of such a molecule. Or, it may be more complex than
that. The most logical complicated answer seems to lie in the molecular structure of
the amino acid chelate or chelate/complex as it enters the mucosal cell. In the amino
acid chelated form, the metal ion is bonded to the ligand on the amino and carboxyl
moieties of the amino acid ligands. That bonding protects the metal ion by rendering
the metal ion unreactive.9
If the amino acid chelate/complex enters the cytoplasm of the cell via an active
transport system, it retains the metal complex configuration because the chelate/
complex molecule enters the cytoplasm still attached to the transport molecule from
the active transport system.10 Based on the pH of the cytoplasm of the mucosal cell,
the amino acid chelate/complex may reconfigure into an amino acid chelate or it may
not. It depends on the pH that the molecule is exposed to as it enters the cytoplasm.
The overall pH of the cell is mildly acidic, although the pH can change depending on
the location in the cell or its cellular contents at a given moment in time. Some vacu-
oles may be alkaline, while others are very acidic (pH 5). In other regions, particu-
larly around the nucleus of the cell, the pH may be 7 to 7.8. In general, however,11 the
pH of the cell is about 6.8. This pH environment encourages an amino acid chelate/
complex that is entering the cell to reconfigure as a chelate.
In spite of the tendency to reconfigure into an amino acid chelate following its
immediate entry into the cell, the stability constant of the chelate/complex molecule
is half what it was when the molecule existed in the form of an amino acid che-
late.5 There is a strong possibility that while still in this chelate/complex configu-
ration, the less-protected metal can be attracted to another ligand in the cytoplasm
with a higher potential formation constant. Concurrently, while the metal remains in
the amino acid chelate/complex form, the mucosal cell may more easily recognize the
amino acid chelate/complex as a molecule that contains a metal rather than simply
an amino acid/peptide possibly due to the fact that not all of the bonding sites of the
metal ions are satisfied while it is in the form of an amino acid chelate/complex. If the
mucosal cell recognizes that the molecule contains a metal bonded to an amino acid,
in times of greater mineral need the molecule could be sent directly to the plasma
without further hydrolysis even as it probably reconfigures into a chelate molecule
while transiting through the cell and into the plasma. This is discussed in greater
detail in the following chapter. Alternatively, some of the absorbed chelate or chelate/
complex may simply traverse the mucosal unhydrolyzed and be taken into the plasma
as such.
Zinc can be used as an example to illustrate the concept of intracellular hydrolysis
as well as a possible mode of intracellular recognition of an absorbed amino acid
chelate/complex. Like most transitional metals, zinc has six binding sites. When in
the form of a bisamino acid chelate, four of those binding sites are satisfied by two
amino acid ligands. Two sites are occupied by nitrogen atoms and two by oxygen
atoms. This still leaves two unoccupied bonding sites on the metal. As noted in a pre-
vious chapter, these sites are frequently satisfied by attaching two water molecules
to the metal while it continues to exist in the amino acid chelate state. When the
pH changes to an acidic environment, the amino acid chelate rings open, resulting
in a bisamino acid chelate/complex. The bonds in this molecule emanate from the
carboxyl moieties of the ligands to the metal ion. This leaves four available binding
sites on the metal ion. It is in this molecular configuration that the zinc amino acid
chelate/complex molecule enters the cytoplasm of the absorptive cell (Figure 10.2).
While in the form of a bisamino acid chelate/complex, the metal is more likely
to be recognized, and due to the chelate/complex stability constant being lower in
this molecular configuration, the zinc is removed from the amino acid ligands by
another ligand located in the cytoplasm that has a potential formation constant that is
stronger than the carboxyl bonds holding the amino acid ligands to the metal ion. If
the metal is removed from the amino acid chelate/complex molecule, the amino acid
chelate/complex is destroyed. As an amino acid chelate, the stereochemistry of the
ligands tends to protect the metal from other reactions, particularly if an additional
two or the other two binding sites are also satisfied.9 When the chelate ring opens due
to environmental pH changes, the metal now has four available binding sites, which
are not protected by the stereochemistry of the chelate rings.
If the amino acid chelate were formed with three amino acid ligands, the strength
of its formation constant would increase,5 resulting in less mucosal cell hydrolysis.
O O N
N H2O N
Zn Zn
Zn N N OH2
O
O O
Ionic Zinc Chelated Zinc Chelated Zinc with Zn
without water water molecules
molecules
O
N
Zinc amino acid
chelate/complex
without water molecules
FIGURE 10.2 The bonding sites available on ionic zinc and bisglycine molecule zinc amino
acid chelate/complex with and without water molecules.
There may also be less recognition that the absorbed molecule contains a metal ion.
Ultimately, this may result in less potential metabolic utilization associated with that
metal, although other factors may also play roles in reducing overall metabolic activity.12
The cell and its organelles, as well as the environment surrounding them, are
maintained by buffers composed of a mixture of weak acids and their conjugate
bases. When an acid or a base is added to the buffer, the pH, which depends on the
ratio of associated acid to dissociated acid, changes slowly as the dissociation of the
acid or base by the common ion effect is suppressed. While carbonic acid and phos-
phoric acid are the main buffers within the cell, proteins, amino acids, fatty acids,
and other various organic acids, such as citric acid, can also serve as cellular buffers.13
It is estimated that within a single cell there can be as many as 1,000 different
enzymes.8 Each enzyme functions as a catalyst that can accelerate the multitude of
chemical reactions within the cell 108 to 1010 times faster than a noncatalyzed chemi-
cal reaction.14 Many of these enzymes have phosphate or hydroxide ions within their
structures. The phosphates have a higher potential formation constant for a given
metal than do the ligands that form the amino acid chelates or chelate/complexes.
For example, a zinc bisglycinate chelate has a stability constant of 109. A zinc bis-
glycinate chelate/complex, which is created by subjecting the chelate to a low acid
pH or high alkaline pH environment, has a stability constant of 104.5, or just half that
of the amino acid chelate.5 The potential formation constant of the phosphate ion is
1012.44 if it is able to bind to that same zinc ion.5 Thus, if an amino acid chelate or
chelate/complex with the lower stability constant were to enter the cytoplasm and
come in contact with a phosphate ion in the cytoplasm of the mucosal cell, it would
be a very simple process for the phosphate ion to remove the metal from the amino
acid ligand. In the case of a chelate/complex, when the bonds holding the metal ion to
the carboxyl moieties of the amino acid ligands were broken, the amino acids would
then be liberated from the metal and metabolized in the same manner as any other
absorbed amino acid. The metal that was picked up by the phosphate ion would be
stored or utilized similarly to any absorbed metal ion that was similarly picked up
by a phosphate ion.
Once the amino acid chelate/complex reconfigures as an amino acid chelate or if
an amino acid chelate is absorbed intact into the cytoplasm, it is more difficult for
the phosphate ion to remove the metal ion from the molecule, depending, of course,
on the stability constant of the chelate molecule as well as the concentration of the
phosphate ion. That does not mean it is impossible for the phosphate ion to remove a
metal ion from an amino acid chelate; it is simply less likely to occur.
To illustrate the mucosal cell hydrolyzation of an amino acid chelate or chelate/
complex concept more fully, one can consider the absorption of nonheme iron from
an iron salt. As the salt enters the stomach, gastric juice and other digestive secre-
tions ionize the iron salt and then help maintain its continued ionic state. While a
small amount of iron may be absorbed from the stomach,15 most of it is absorbed
from the duodenum and, to a lesser extent, the jejunum.16 The uptake of the iron ion
by the absorptive cells begins immediately on the iron ion reaching the brush border
of the mucosal cells and is generally completed within 30 to 60 minutes following
its initial exposure.17 The size of the dose of an iron salt presented to the small intes-
tine will dictate, to a degree, which ionic pathway will be employed for absorption
of the iron. The pathway chosen also determines the rate of metal ion uptake from
the lumen. If a low dose of iron is presented to the brush border, more of the iron
will be picked up by transporter molecules and carried into the cell by facilitated
diffusion or active transport. If the dose is higher, most of the absorption may be
accomplished by passive diffusion.18 This is reflected as more of the low-dose iron
being absorbed than a high dose of the same element. On entering the mucosal cell,
the iron is used as a metabolically active cell component, or if it is not immediately
required by the body, it is bound to intracellular ligands and stored within the muco-
sal cell.19 These intracellular ligands regulate the passage of the ionic iron across the
cell and include transferrin in the plasma, ferritin, hemosiderin, hepcidin, and newly
created amino acid chelates within the cell in which somehow the body is able to rec-
ognize the presence of the metals that they carry.18 As a result of controlling the pas-
sage of ionic iron across the mucosal cell to the plasma, these intracellular ligands
tend to reduce or prevent iron overload within the body. These same molecules also
regulate the transfer of iron from the hydrolyzed amino acid chelate/complex, as will
be discussed below.12
Ferritin is created when the absorbed iron ions are attached to an apoferritin mol-
ecule. Apoferritin is a protein that acts as a “shell” capable of trapping ferrous iron
atoms. Apoferritin has the shape of a hollow sphere, and each molecule of apoferri-
tin can hold as many as 4,500 iron atoms.20 Inside the hollow shell of the apoferritin
molecule are six channels that are situated symmetrically and connect the hollow
interior with the outside environment. These channels provide a pathway for the iron
and bioflavonoids to enter and exit from the apoferritin molecule.21,22 This is seen in
Figure 10.3, which is a schematic drawing of the iron-laden molecule.23
FIGURE 10.3 A cross section of the ferritin molecule showing its subunits and the chan-
nels that lead to the interior of the molecule. The core of crystalloid iron is represented by the
triangular lattices, while the phosphate molecules are shown as black dots. (Redrawn from
Treffy, A, and Harrison, PM, “Incorporation and release of inorganic phosphate in horse
spleen ferritin,” Biochem J 171:313–320, 1978.)
To sequester the excess iron taken up by the mucosal cell, the ferritin content of
the absorptive cells increases with greater iron administration. This prevents the
large influx of iron that is absorbed into the mucosal cells from being transferred
directly to the body, with subsequent toxic results. When iron is required by the
body, the stored iron in the ferritin molecule is reduced to the ferrous state and
released from the ferritin. There is a close relationship between iron transfer into the
body from the absorptive cells and ferritin. When the ferritin concentration is low,
iron transfer to the plasma is high and vice versa.
There are no specific iron-binding ligands in the ferritin molecule. The iron is
present in a crystalline aggregate form of ferric hydroxide and ferric phosphate.
Figure 10.3 shows how the phosphate molecules are located on the surface of the
iron core.23 The absorbed iron is reduced to the ferrous state as a result of the acid
pH of the cellular cytoplasm and then incorporated into the apoferritin molecule.
Once introduced into the apoferritin molecule, the ferrous iron is oxidized to the
ferric state in the ferritin cavity as it combines with amino acid side chains, known
as histidyl groups. These histidyl groups catalyze the oxidation of the ferrous iron
to the ferric state in the presence of an oxidizing agent, including O2.24 To leave the
apoferritin molecule, the iron must be reduced back to the ferrous state with the help
of bioflavonoids such as reduced riboflavin. Figure 10.4 illustrates this.25 This intra-
cellular chelating agent removes the ferrous iron on a last-in, first-out basis.26
When reduced iron leaves the ferritin molecule, it enters what is described as the
“chelatable Fe+2 pool.” The chelating agent in this pool can be an amino acid, citric
acid, and so on. Once chelated, the iron migrates to the basement membrane of the
Fe(II)
Oxidant
Loosely
bound Fe
c
Reductant
Oxidant
Fe(II)
Fe(II)
c
Chelator
Fe-chelate
FIGURE 10.4 The uptake and release of loosely bonded iron by ferritin. The crystalline
lattice indicates iron oxide initiation sites associated with amino acid side chains. (Redrawn
from Harrison, PM, Hoy, TG, Macara, IG, and Hoare, RJ, “Ferritin iron uptake and release
structure function relationships,” Biochem J 143:445–451, 1974.)
absorptive cell. There, it crosses the membrane of the cell and enters the plasma. As
soon as this ferrous iron enters the blood, it is oxidized once again to the ferric state
by oxygen in the presence of ceruloplasmin, after which it is immediately seques-
tered by serotransferrin.27
If the absorbed iron is needed by the mucosal cell for its internal metabolic activi-
ties or by the body for some other purpose, it will be reduced to the ferrous state and
released from the ferritin molecule as described. If, following absorption into the
mucosal cell, the iron is not required for metabolic purposes in the body or mucosal
cell, it will remain in the ferritin molecule and ultimately be excreted from the body
as part of the short-lived mucosal cells, which, when they reach the tip of the villus,
are sloughed off into the intestinal lumen.28
As the discussion suggests, the phosphate ions located in the cytoplasm as well as
in the apoferritin molecule both have stronger binding potentials for the metal than do
the amino acid ligands from any bisamino acid or monoamino acid metal chelate or
chelate/complex entering the mucosal cell. When metal phosphates are created with
the released metals from the hydrolyzed amino acid chelates or chelate/complexes, it
results in the formation of metallic compounds that are chemically identical to those
that would have been created following the absorption of those same metals in the
ionic state. The subsequent metabolism of the two sources of metal (ionic or che-
lated) throughout the body would also be identical. Thus, in the two studies described
at the beginning of this chapter, whatever portion of the absorbed iron amino acid
chelate/complex that was hydrolyzed in the mucosal tissue would have resulted in the
creation of iron ions that potentially could have been incorporated into apoferritin,
hemosiderin, and so on or otherwise treated similarly to absorbed ionic iron.
The transferrin molecule in the blood is a transport vehicle for iron whether it is
sourced from ionic iron absorption or from iron that was derived from mucosal cell
hydrolyzation of an absorbed iron amino acid chelate or chelate/complex. One mole-

cule of apotransferrin can bind two atoms of iron-forming transferrin. Apotransferrin
is a b-globulin that is formed in the liver and sent to the plasma. As the iron stored
in the ferritin molecule is reduced and transferred out of the absorptive cell, the apo-
transferrin in the plasma complexes that iron ion and carries it into the body’s cells by
endocytosis. The iron is released within the body’s cells by an energy-dependent pro-
cess, and the apotransferrin is returned to the plasma to pick up additional iron ions.28
Hemosiderin is a deposit of degraded ferritin and coalesced iron atoms. It forms
in the lysosomal membranes of the cell and can be as high as 50% iron. It can be
labilized to supply free iron, but the rate of release of that iron is much slower than
that of iron from ferritin.28
It is estimated that approximately 95% of the iron from absorbed ionic iron or
hydrolyzed iron amino acid chelate/complex that is transferred from the mucosal
cells into the plasma enters the portal blood. The remainder is sent to the lympho-
cytes. There are receptors on the serosal side of the absorptive cell that affect the
transfer of the ionic iron out of the cell to the plasma. The pathway for the exit of
ionic iron from the mucosal cell at the basement membrane is not the same as the
pathway for its entry into the cell. Uptake of ionic iron into the cell is more rapid than
is its release. With an increasing iron concentration at the brush border, the uptake
increases, while serosal transfer remains relatively constant. The pathway employed
by iron when it is absorbed into mucosal tissue from the duodenum, jejunum, or
ileum is similar, but serosal transfer of iron is maximum from the proximal duode-
num.18 All of these observations relating to the transfer of iron into the plasma apply
similarly whether the iron is sourced from ionic iron or iron from an amino acid
chelate or chelate/complex that has been hydrolyzed into the mucosal cell.
It was noted that when the ferritin concentration was low, iron uptake by the
absorptive cells increased and vice versa. This feature also applies to iron amino acid
chelate or chelate/complex. The only way this intracellular regulatory mechanism
can logically function is if the absorbed iron from either source is chemically avail-
able to be collected by the apoferritin. In other words, even though the iron amino
acid chelate or chelate/complex may have been absorbed intact into the absorptive
cell, to be complexed within the apoferritin molecule and its rate of transfer to the
plasma controlled, it must be hydrolyzed within the mucosal cell. It is only then that
the release of the iron from the amino acid chelate or chelate/complex into the cyto-
plasm as a free cation can occur as the following study demonstrated.12
Ten iron-sufficient adult males consumed a whole-maize breakfast meal contain-
ing 55.5 kBq 59Fe as ferrous sulfate and 111 kBq 55Fe as ferrous bisglycinate mixed
together following an overnight fast. Blood was drawn from each individual in the
morning prior to consuming the breakfast meal to establish their baseline serum fer-
ritin concentrations and hemoglobin levels. A final blood sample was obtained from
each person enrolled in the study 14 days later.
Absorption of the amino acid-chelated source of iron was approximately 4.7
times greater than was the absorption of iron from ferrous sulfate (p < 0.05). The
study demonstrated that the amino acid chelate did not dissociate in the gastroin-
testinal tract but instead was absorbed intact into the mucosal tissue. If the amino
acid chelate had dissociated in the stomach or intestines with the subsequent release
of the iron prior to absorption, then the iron absorption would have been about the
same from either source of iron because both would have been in the form of ionized
iron and absorbed by the same pathway as the iron from ferrous sulfate. Because the
absorption of the iron amino acid chelate was significantly higher than that of fer-
rous sulfate, the data supported the position that the amino acid chelate did not share
the ionic absorption pathway employed by ferrous sulfate. The amino acid-chelated
source entered the mucosal cell as an intact, nonionized molecule via the amino acid
absorption pathway.
In spite of entering the mucosal cell as an intact amino acid chelate or amino acid
chelate/complex, a significant portion of these absorbed iron molecules were subse-
quently hydrolyzed in the mucosal cells, as demonstrated by the scattergram illus-
trating the percentages of iron absorption from both iron sources separately and their
relationships to the natural logarithm of plasma ferritin (Figure 10.5).12 The linear
correlation coefficients between the natural logarithm of the baseline plasma ferri-
tin and the natural logarithm of iron absorption from both iron sources were similar
and significantly different from zero. This indicated that both sources of iron were
being sequestered in the mucosal tissue following their uptakes from the lumen, and
their rates of transfer to the plasma were controlled. This can only have occurred if
a significant portion of the absorbed iron amino acid chelate was dissociated in the
mucosal tissue prior to transferring that source of iron to the plasma.
The linear regression equations for the two iron sources, however, were very dif-
ferent. The percentage of iron absorption was 22.6% ± 3.4% for the natural logarithm
3
In Iron Absorption (%)
0
2 4 6 8
In Serum Ferritin (µg/L)
FIGURE 10.5 The relation between natural logarithm (In) serum ferritin and percentage
iron absorption from ferrous sulfate (■; r = –0.62, p < 0.04) and iron amino acid chelate
(ferrous bisglycinate) (○; r = –0.66, p < 0.05) consumed concurrently in whole-maize meal
porridge. (Redrawn from Bovell-Benjamin, AC, Viteri, FE, and Allen, LH, “Iron absorption
from ferrous bisglycinate and ferric trisglycinate in whole maize is regulated by iron status,”
Am J Clin Nutr 71:1563–1569, 2000.)
of ferritin for iron amino acid chelate compared to 4.2% ± 0.5% for ferrous sulfate.
This observation tends to support the conclusion of the in vitro study reviewed at the
beginning of this chapter that reported that the uptake and transfer of metals chelated
to amino acids were not only higher but also much more rapid than were the uptake and
transfer of that same metal when it had to source its transport ligands from mucosal tis-
sue.6 The negative slopes in relation to plasma ferritin reached the lowest iron absorp-
tion values of 2.6% for the iron amino acid chelate and 0.4% ferrous sulfate, which
indicated that iron absorption from either iron source was effectively and similarly
regulated by the iron reserves within the body. There was an almost-perfect correla-
tion between the percentage of iron absorbed from iron amino acid chelate and ferrous
sulfate when they were both administered concurrently in the whole-maize breakfast
meal porridge (r = 0.99, p < 0.001). That correlation demonstrated that the individu-
als who absorbed the lowest or the highest amounts of iron from ferrous sulfate also
absorbed the lowest or the highest amounts of iron from the amino acid chelate.12
In a subsequent study, these same investigators examined the absorptions of iron
amino acid chelate (ferrous bisglycinate) and ferrous ascorbate as they related to
serum ferritin when both were administered in water.25 As with the previous study, the
absorption of the iron amino acid chelate was greater than was the absorption of the
iron from the ferrous ascorbate. However, as shown in Figure 10.6, the absorption into
the plasma of both iron sources was inversely related to the serum ferritin concentra-
tion.12 As the serum ferritin concentration increased, the transfer of the iron into the
plasma decreased, again demonstrating mucosal cell hydrolyzation of the amino acid
chelate with subsequent regulation of the transfer of its iron to the plasma.
4
In Iron Absorption (%)
0
0 1 2 3 4 5
In Serum Ferritin (µg/L)
FIGURE 10.6 The relation between natural logarithm (In) serum ferritin and the percent-
age iron absorption from a reference dose of ferrous ascorbate (■; r = –0.61, p < 0.03) and
iron amino acid chelate (ferrous bisglycinate) (○; r = –0.78, p < 0.001) administered in water.
(Redrawn from Bovell-Benjamin, AC, Viteri, FE, and Allen, LH, “Iron absorption from fer-
rous bisglycinate and ferric trisglycinate in whole maize is regulated by iron status,” Am J
Clin Nutr 71:1563–1569, 2000.)
That portion of the study comparing absorption of ferrous ascorbate and the iron
amino acid chelate, plus the previous study comparing this same chelate with fer-
rous sulfate, both illustrated the influence of ferritin on the transfer of the iron amino
acid chelate into the tissues after its absorption into the mucosal tissue. The inverse
relationship shown in Figures 10.5 and 10.6 can only result if a significant portion
of the ingested chelates was hydrolyzed in the mucosal cells and the iron released
from the original ligands so that the iron was free to be sequestered by the apofer-
ritin molecule. The fact that the chelated source of iron was absorbed 4.7 times more
than was the iron from the sulfate source indicates that intact absorption of the iron
amino acid chelate into the absorptive cells occurred via a different pathway that did
not require gastric hydrolyzation of the chelate. Since the ferritin level regulated the
transfer of the absorbed iron from the mucosal tissue to the plasma similarly for both
the iron from the amino acid chelate and the ionic iron, these studies also indicated
that, prior to iron’s entry into the plasma, mucosal cell hydrolysis of much of the
amino acid chelated or chelated/complexed source of iron also took place.
In summary, then, the ingestion of a metal amino acid chelate delivers a mol-
ecule to the gastrointestinal tract that is capable of generally withstanding hydrolysis
within the gastrointestinal tract. While the molecular structure of the molecule may
change, based on the pH of the luminal environment to which it is subjected, that
environment does not generally destroy the molecule. The amino acid chelate is
absorbed into the mucosal cell by facilitated diffusion, and the amino acid chelate or
chelate/complex is absorbed by an active transport system. Once introduced into the
mucosal cell, cytoplasmic ligands with much higher potential formation constants
than the stability constants of the amino acid chelate or chelate/complex can initiate
the hydrolyzation of a significant portion of absorbed molecule into the metal ion
and its amino acid ligands. The metal ion that is released from the intracellularly
hydrolyzed amino acid chelate or chelate/complex is subsequently handled within
the mucosal cell in the same way any other similar metal ion would be handled fol-
lowing its absorption. It is utilized or stored within the mucosal cell according to the
body’s immediate need for that particular metal. If required by the body, the metal
ion from the hydrolyzed amino acid chelate or chelate/complexes will be transferred
to the plasma similarly to the same metal ions from other sources. If it is not required
by the body, the metal ion remains sequestered within the mucosal cell until that
mucosal cell completes its trip from the crypt of Lieberkühn to the tip of the villus.
Here, the cell is sloughed off and becomes part of the fecal material.29 Thus, the
transfer of the metal ion from the hydrolyzed portion of the amino acid chelate or
chelate/complex is effectively regulated as it enters the plasma.
REFERENCES
2. Overton, J, ed., Nutrient Requirements of Laboratory Animals (Washington, DC:
National Academy Press) 1955.
3. Pineda, O, personal communication, 2001.
chelates by utilization of ninhydrin,” AOAC poster, St. Louis, MO, September, 2004.
6. Fang, SM, Burton, SA, and Petersen, RV., “Bioavailability of zinc: Effect of amino acid
chelation,” in Ashmead, D., ed., Chelated Mineral Nutrition in Plants, Animals and Man
(Springfield, IL: Thomas) 137–151, 1987.
1986.
Spring/Summer 1970.
proof of chelation,” Proceedings International Conference on Human Nutrition, Salt
Lake City, UT, January 21, 1995.
10. Meister, A, Tate, SS, and Thompson, A, “On the function of the g-glutamyl cycle in the
transport of amino acids and peptides,” in Elliott, K, and O’Connor, M, eds., Peptide
174, 1975.
12. Bovell-Benjamin, AC, Viteri, FE, and Allen, LH, “Iron absorption from ferrous bisglyci-
71:1563–1569, 2000.
58–69, 1975.
15. Dagg, JH, Kuhn IN, Templeton, FE, and Finch, CA, “Gastric absorption of iron,”
Gastroenterology 53:918–922, 1967.
16. Wheby, MS, “Site of iron absorption in man,” Scand J Haematol 7: 56–62, 1970.
17. Hallberg, L, and Solvell, L, “Iron absorption studies,” Acta Med Scand Supp 358,
168:1–69, 1960.
18. Bezkorovainy, A, Biochemistry of Nonheme Iron (New York: Plenum Press) 70–126, 1980.
19. Richmon, VS, Worwood, M, and Jacob, A, “The iron-content of intestinal epithelial
cells and its subcellular distribution: Studies on normal, iron overloaded and iron defi-
cient rats,” Br J Haematol 23:605–614, 1972.
20. Finch, CA, and Huebers, H, “Perspectives in iron metabolism,” N Engl J Med
306(25):1520–1528, 1982.
21. Hoare, RJ, Harrison, PM, and Hoy, TG, “Structure of horse-spleen apoferritin at 6 Å
resolution,” Nature 255:653–654, 1975.
22. Harrison, PM, “Ferritin: An iron-storage molecule,” Semin Hematol 14:55–70, 1977.
23. Treffy, A, and Harrison, PM, “Incorporation and release of inorganic phosphate in horse
spleen ferritin,” Biochem J 171:313–320, 1978.
24. Treffy, A, Sowerby, JM, and Harrison, PM, “Oxidant specificity in ferritin formation,”
FEBS Lett 100:33–36, 1979.
25. Harrison, PM, Hoy, TG, Macara, IG, and Hoare, RJ, “Ferritin iron uptake and release
structure function relationships,” Biochem J 143:445–451, 1974.
26. Hoy, TG, Harrison, PM, and Shabbir, M, “Uptake and release of ferritin iron. Surface
effects and exchange within crystalline core,” Biochem J 139:603–607, 1974.
27. Osaki, S, Johnson, DA, and Frieden, E, “The mobilization of iron from the perfused
mammalian liver by a serum copper enzyme, ferroxidase,” J Biol Chem 246:3018–3023,
1971.
28. Hunt, S, and Groff, JL, Advanced Nutrition and Human Metabolism (St. Paul, MN:
West) 292, 1990.
29. Fox, SI, Human Physiology (Dubuque, IA: Brown) 534, 1990.
11 The Uptake of Amino
Acid Chelates into
and out of the Plasma
Data presented in the previous chapters demonstrated that following uptake into the
absorptive cell of either an amino acid chelate or chelate/complex, significant intra-
cellular hydrolysis can, and probably does, occur prior to transferring its metal ion
into the plasma. However, not all of the absorbed amino acid chelate or chelate/
complex is hydrolyzed within these cells. Some of these molecules remain intact and
are transferred to the plasma as such.
Using iron as an example, Bezkorovainy has reported that, in addition to transfer-
rin, intracellular iron can be bound to amino acids, which carry the metal through
the cytoplasm to the serosal side of the basement membrane.1 While he did not iden-
tify any specific amino acid functioning as this transport molecule, there is no reason
not to believe that his observation could also include amino acid chelates and chelate/
complexes that have been absorbed intact into the mucosal cell from the lumen.
In general, as noted in Chapter 10, the transfer of iron into the plasma from absorp-
tive cells is highly regulated and is tied to ferritin levels in the body.2,3 The transfer
increases when iron reserves are low and decreases as the ferritin level increases.
The rate of erythrocyte production also appears to influence iron absorption.4 Thus,
if the amino acid chelate or chelate/complex taken up by the mucosal cells were
completely hydrolyzed in the absorptive cells before transferring the iron into the
plasma, its rate of iron transfer to the plasma would be equivalent to that of absorbed
nonheme iron from inorganic salts regardless of how much iron amino acid chelate
was initially taken up into the mucosal cells. To the contrary, however, the iron con-
centration in the plasma is higher when the amino acid chelated source is ingested
compared to ferrous sulfate. As the previously described study2 in which 10 fast-
ing adult males consumed a whole-maize breakfast porridge equilibrated with 59Fe
sulfate or 55Fe bisglycinate demonstrated, more iron from the amino acid-chelated
source was recovered in the plasma compared to iron from ferrous sulfate. This indi-
cated that more than one form of iron was transferred to the plasma from the muco-
sal cells. Based on data from studies also previously discussed, the iron derived from
the ferrous sulfate source was ionized in the lumen. On entry into the mucosal cell, a
portion of this ionic iron was picked up by transfer molecules in the cytoplasm of the
absorptive cells and delivered to the plasma. The ionic character of that iron was pre-
served in the form of a salt or complex as it was carried from location to location by
various transport molecules. The only way the plasma iron level following ingestion
of the amino acid chelate or chelate/complex could be the same as the iron plasma
171
level after intake of ferrous sulfate is if the chelate were completely hydrolyzed in
the mucosal cells prior to iron transfer to the plasma. Not all the amino acid chelate
or chelate/complex that was taken up by the mucosal cells appeared to be hydrolyzed
prior to being sent to the plasma.
The uptake of the iron amino acid chelate into the blood as described in the study
was 4.7 times greater than uptake of ferrous sulfate. The only explanation for these
results is if some of the iron amino acid chelate or chelate/complex was taken up
and transferred to the plasma by one or more pathways that were different from the
pathway employed by the iron from ferrous sulfate. If the iron amino acid chelate or
chelate/complex were completely hydrolyzed in the mucosal cells before being taken
up into the plasma, then its rate and the quantity transferred to the plasma would be
similar to that of iron from ionized ferrous sulfate.
Since the amount of iron gained from the amino acid-chelated source was greater,
it suggests that at least part of the iron delivered to the plasma originated from a
source that had a different molecular configuration compared to the ionic iron from
the ferrous sulfate and was taken into the blood differently than the iron from the fer-
rous sulfate. The observation does not preclude some intracellular hydrolyzation of
a portion of the absorbed iron amino acid chelate or chelate/complex. As discussed
previously, a part of those molecules would have been hydrolyzed in the mucosal
cells prior to transferring the iron to the plasma.
The amount of hydrolyzation of an absorbed amino acid chelate or chelate/
complex within the mucosal cells is dependent on several factors, including the
ligand selected. The choice of the amino acid ligand used to chelate the metal origi-
nally prior to ingestion dictates the strength of the stability constant of the resulting
chelate or subsequent chelate/complex.5 If everything else is equal, the greater the
stability constant is, the less intracellular hydrolyzation will occur, as was demon-
strated in the in vitro double-isotope studies described in Chapter 10. The mucosal
tissue hydrolyzation of iron chelated to lysine was less than was the hydrolyzation of
iron chelated to methionine.6 The stability constant of the iron methionate is lower
than the stability constant of iron lysinate.5
The pH in the locus of the mucosal cell to which the amino acid chelate migrates
following absorption can also affect its rate of intracellular hydrolyzation, partic-
ularly if the amino acid chelate has configured as an amino acid chelate/complex
within the mucosal cell and another intracellular ligand with a higher potential for-
mation constant comes in contact with this chelate/complex. It will be remembered
that in certain acid pH environments, which may exist under certain conditions in
various areas of the cytoplasm, some amino acid chelates may reconfigure from
amino acid chelates to chelate/complexes.7 The exact pH at which this occurs is not
the same for every amino acid chelate. If an amino acid chelate is formed using an
amino acid ligand with a higher potential formation constant, then it will require a
lower acid pH environment to break the bond between the nitrogen on the amine
moiety and the metal ion, resulting in the molecule reconfiguring as an amino acid
chelate/complex.8 If the stability constant is not as strong, then the amino acid chelate
will more readily reconfigure into a chelate/complex in an acid pH that is not as low.8
This same concept is true in the alkaline pH environment, which also exists in
certain loci of the mucosal cell. The pH environment must be higher to induce an
The Uptake of Amino Acid Chelates into and out of the Plasma 173
amino acid chelate with a higher stability constant to break its bond between the
nitrogen on the amine moiety and the metal ion compared to an amino acid chelate
with a lower stability constant.8 If an alkaline pH environment causes this bond to
break, it is easier for a ligand with a higher potential formation constant to remove
the metal ion from the amino acid chelate/complex and sequester it.
The mean pH of the cell is about 6.8, which, generally speaking, encourages the
formation of an amino acid chelate.7–10 The pH can change within the cell depending
on the locus. For example, the pH of the content of the vacuoles can be as low as 5.0,
whereas a pH of 7.6 to 7.8 can occur in the nucleoplasmic matrix.10 The pH of the
cytoplasm of the cell can be altered by its absorption of acidic or alkaline substances,
but the cell will ultimately reestablish its original pH through its own internal buffer-
ing system.10 Thus, the precise location of the amino acid chelate or chelate/complex
within the mucosal cell at a given time will often dictate whether its molecular struc-
ture is that of an amino acid chelate or a chelate/complex. If the molecule assumes
the form of an amino acid chelate/complex, then the molecule is more likely to be
hydrolyzed within the absorptive cell.
Following absorption by active transport, or in certain cases facilitated diffusion,
if the amino acid chelate/complex reconfigures into its original chelate structure in
the cytoplasm of the mucosal cell, then it will provide greater stability and protec-
tion to the metal ion.5,11 The stability will be similar to the stability of an amino acid
chelate that is absorbed by other modes of facilitated diffusion. In this case, intracel-
lular hydrolysis of the chelate will be less likely. If it is not hydrolyzed in the mucosal
cell, the amino acid chelate migrates to the basement membrane of that mucosal cell
and exits the cell as the same chelate molecule that was ingested.12 Intracellular
hydrolysis of an absorbed amino acid chelate is not a prerequisite to transferring the
molecule from the lumen to the plasma, although if it is not hydrolyzed, it is likely
that the amino acid chelate exiting the mucosal cell and entering the plasma does so
as if it were an amino acid or peptide-like molecule rather than a metal ion bonded
to amino acids. At what point the body finally recognizes the metal ion within the
amino acid chelate molecule or what triggers that recognition is presently unknown.
That it occurs is without question as numerous metabolic studies described in this
book attested.
There is still insufficient information relating to the way iron, and by inference
other minerals, actually leave the intestinal absorptive cells.13 Bezkorovainy has
identified three transport carriers for iron in the absorptive cell.1 The first is iron
transferrin. As described previously, transferrin picks up iron from a pool that con-
tains iron amino acid complexes and carries the acquired iron to the plasma. The
second carrier is a protein molecule that is similar to transferrin. It also picks up
iron from the same pool as transferrin and delivers that iron to the basement mem-
brane. Unlike iron transferrin, however, this protein molecule does not carry the
iron into the plasma. Instead, it releases the iron at the basement membrane, where
other carriers transfer the metal ion to the plasma. This particular protein carrier is
not influenced by the iron status of the body. During cases of iron deficiency, the
amount of protein molecules capable of carrying iron to the basement membrane
in the absorptive cells remains constant.14 And finally, there are amino acids that
are able to bind iron ions, transport them through the mucosal cell cytoplasm to the
Fe - Ferritin Lysosome Fe - T Fe - T
Fe+++ Fe++ T
Fe - aa Fe - T like protein
aa
Fe - aa T like protein +UNK Fe - T “sloughed off
Fe Fe aa T
Mitochondria
aa Fe - enzymes
Intraluminal Factors Villus

Gut Secretions Capillary
Composition of Meal
Form of Iron
Corporal Factors
Hormones
Iron Status
Serum Ferritin Crypt
FIGURE 11.1 Summary of possible mechanisms that transfer iron absorbed into the muco-
sal cell to the plasma. aa = amino acid; T = transferrin. (Redrawn from Bezkorovainy, A,
Biochemistry of Nonheme Iron (New York: Plenum Press) 108–109, 1980.)
basement membrane, and then carry them into the plasma. Figure 11.1 summarizes
these three intracellular transport systems.1
While amino acid chelates may mimic the pathway employed by amino acids that
bind metal ions in the cytoplasm, there is a fourth pathway for the transfer of iron
that does not require ionization for the pathway to function. This pathway is used by
some amino acid chelates and follows the same pathway from the mucosal cell to the
plasma that is employed by proteoses, peptones, and small peptides.15
The pathway used by the amino acid chelate for efflux from the basement mem-
brane is probably, to a degree, a function of the configuration of the molecule when
it is presented to the basement membrane. If an amino acid chelate/complex with a
charge on each of its amine moieties arrives at the basement membrane and subse-
quently loses the positive charge on each amine moiety due to a shift in the pH at
the basement membrane, then an amino acid chelate would be re-formed, and this
chelate molecule could exit from the absorptive cell as if it were an amino acid via
diffusion. In the case of facilitated diffusion, the flow of the amino acid chelate
molecules will move from the higher concentration in the cytoplasm at the base-
ment membrane side of the absorptive cell to the lower concentration of amino
acids on the serosal side of the membrane. The carrier for the facilitated diffusion
of the amino acid chelate, which is a protein molecule, water, or certain ions such
as sodium, “shuttles back and forth between the two surfaces of the membrane.”16 It
does not require energy to function, but it does require that the amino acid chelate
be attached to a transport molecule through some type of electronic attraction as
Cytoplasm
Amino Acid
(glutathione)
Chelate/Complex (AA) ADP + Pi
γ-glu-cysH-gly
γ-Glutamyl Transpeptidase Gsh Synthetase ATP
glycine
γ-glu-AA cysH-gly
Basement Membrane Peptidase γ-glu-cysH
γ-Glutamyl Cyclotransferase ADP + Pi
cysteine
γ-Glu-cysh Synthetase
ATP
5-Oxoprolinase
ATP ADP + Pi
Amino Acid Chelate Serosa

or
Amino Acid Chelate/Complex (AA)
FIGURE 11.2 The proposed transfer of amino acid chelate/complex from the absorptive
cell to the serosa by active transport. ADP = adenosine diphosphate; ATP = adenosine tri-
phosphate. (Redrawn from Ashmead, HD, The Roles of Amino Acid Chelates in Animal
Nutrition (Park Ridge, NJ: Noyes) 464–465, 1993.)
previously described. When equilibrium is reached on both sides of the cell mem-
brane, the flow stops.17 This same pathway could be employed by amino acid chelates
that are absorbed into the mucosal cell by facilitated diffusion. As is the case when
describing luminal uptake of the amino acid chelate by facilitated diffusion through
the mucosal cell, the metal simply “goes along for the ride”.17
If the amine group or groups in the molecule are no longer bonded to the metal
ion (an amino acid chelate/complex), the isoelectric point of the chelate/complex
can shift from a neutral molecule to an acidic molecule depending on the pH at the
basement membrane. As described, if that occurs, the molecule now has a positive
charge on its amine moieties, which allows it to be attached to the negative charge on
a transport molecule located in or near the basement membrane.16,18 At that point, the
chelate/complex can potentially exit the absorptive cell by active transport.16 The dif-
ference is that in active transport, the movement across the cell membrane is facili-
tated by a transport molecule that requires energy (Figure 11.2).
There is a suggestion that a carrier molecule similar to that employed by facili-
tated diffusion may also be involved in the active transport of the amino acid
chelate/complex across the mucosal cell membrane.16 While bonding the amino
acid chelate/complex to a transport molecule may be a similar concept, it is unlikely
that a protein transport molecule used for facilitated diffusion and active transport
are the same. The mode of operation of each transporter is different. Enzymatic reac-
tions and the use of energy are required to move the chelate/complex across a cell
membrane by active transport. No energy is needed for facilitated diffusion of an
amino acid chelate to occur. Whatever the structure of either transport molecule, it
is necessary that the amino acid chelate or chelate/complex molecule have an attrac-
tion to either carrier system. If transported by facilitated diffusion, the release of the
amino acid chelate on the serosal side of the basement membrane may require that
the pH in the microenvironment change, whereas the release of the chelate/complex
molecule from an active transport carrier would probably be the result of an enzy-
matic reaction.
Ponka et al. stated that, in the case of iron, some of the iron leaving the mucosal
cells enters the plasma in the form of a chelate.13 The amino acid chelate they refer
to may have been created within the absorptive cell, or it is equally possible that it
could be an amino acid chelate molecule that entered the serosa as the same amino
acid chelate that was ingested and absorbed into the mucosal cell.
To illustrate the movement of amino acid chelates in the plasma, placental trans-
fer of ionic-sourced iron from the mother to the fetus occurs through the chorionic
villi of the areola. The iron is bound to specific protein molecules synthesized by
the placenta for the purpose of transporting the maternal plasma iron across the pla-
centa to the fetus.19–22 These transport protein molecules employed in sequestering
plasma iron appear to be similar in structure and function to the transport molecules
found in the intestinal lumen. One such membrane-bound molecule identified in
rats has a molecular weight of 52,000 Da.23 Another identified protein transport mol-
ecule has a molecular weight of 100,000 Da.24
Both of these protein transport molecules are capable of carrying ionic iron across
cellular membranes, but as in the case of intestinal absorption of iron, for which
more than one transport pathway exists, there are additional transport mechanisms
available for the movement of iron to the fetus. Some plasma iron is able to diffuse
across the placental barrier from the mother to the fetus. An iron amino acid chelate
delivered to the maternal plasma can also potentially move across this placental bar-
rier by diffusion or by a different pathway. This may result in greater fetal uptake of
iron, as demonstrated in the study discussed next.25
In this study,25 a single dose of 4.4 mc (microcuries) of 59Fe was orally adminis-
tered by gavage to fasted and anesthetized gestating rats as either an iron amino acid
chelate or ferric chloride approximately 4 days before expected parturition. At 72
hours postdosing, the animals in both groups were sacrificed, their fetuses removed,
and a whole-body scan for 59Fe conducted on each fetus. The fetuses from those
mothers receiving the amino acid chelate source of iron had a mean corrected counts
per minute per gram of tissue of 46 compared to 16 for the fetuses taken from moth-
ers that had been administered ferric chloride.25 The mean 188% increase in iron
fetal content was significantly different between the two groups of fetuses (p < 0.05).
The argument could be made that the mean increased uptake of 59Fe by fetuses
from mothers administered the amino acid chelate was simply due to a greater
amount of iron from the maternal plasma being delivered, as previous studies have
demonstrated. While it is true that in this current gestating rat study the mean mater-
nal tissue uptake of iron increased approximately 120% in those mothers receiving
the amino acid-chelated source of iron compared to mothers administered ferric
chloride, this does not completely explain the increased transfer of iron to the fetuses
of the chelate group. With the exception of the mean percentage differences of 59Fe
in the hearts of the mothers of the two groups, none of the other 59Fe percentage
differences in tissues or organs of the dams in the two groups were as high as the
percentage difference of 59Fe content of the two groups of fetuses. Obviously, the
demand for iron by the fetuses was high during the last stages of gestation, but if
the increase in 59Fe in the fetuses from mothers receiving the amino acid-chelated
source of iron were due entirely to higher maternal plasma iron, then the percentage
difference between the two groups of fetuses should have been closer to 120% rather
than the 188% that was reported.
During the last stages of gestation, there is a significant fetal demand for amino
acids, as seen in Figure 11.3. If some of the iron in the maternal blood remained
chelated to the original amino acids and was delivered to the maternal blood as such,
then it is very probable that the iron in that chelate would be “smuggled” into the tis-
sues, including the fetus, as if it were part of an amino acid/peptide molecule.17 Since
the demand for amino acids by the fetus tends to be higher in the last stages of gesta-
tion for fetal growth than is the demand for those same amino acids by the mother
for maintenance of body tissue, this would help explain why more of the amino acid
chelate that was deposited into the maternal plasma went to the fetus rather than to
the maternal tissue in the study discussed.
A subsequent study further substantiated what has been proposed. The drug
2,2-dichlorvinyl dimethyl phosphate (DDVP) has been reported to cause increased
transport of maternal iron to the fetus.26 When administered to the gestating dam,
DDVP tends to dilate the blood vessels within the placental wall, thereby enhancing
Total Weight
Total Protein
3
Weight Grams
Total Iron
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Days of Gestation
FIGURE 11.3 The relationship of fetal growth to protein and iron requirements.
the flow of blood across the placenta from the mother to the fetus. With the greater
flow of maternal blood, there is greater opportunity for fetal tissue to extract mater-
nal plasma iron.
Thirty-six Sprague-Dawley gestating rats were divided into four groups of
nine rats each.19,27 All the mothers had been synchronously impregnated and were
expected to give birth on the same day. At 3 days before expected parturition, 18
fasted rats were arbitrarily assigned to one of two groups and partially anesthetized
with ether. Using an automatic pipette, each animal was subsequently given an oral
dose of 59Fe as either iron amino acid chelate or ferric chloride in combination with
DDVP. The remaining 18 pregnant rats were similarly anesthetized, divided into two
groups, and dosed with 59Fe from either of the two iron sources but without the addi-
tion of DDVP. Thus, nine animals received iron amino acid chelate plus DDVP and
nine the chelate alone without DDVP. Another nine animals received ferric chloride
with DDVP and nine with ferric chloride but without the DDVP. The preparation of
the two iron sources is described next.
The animals in group 1 each received 4 mc of 59FeCl3 in 25 mL of water to which
was added 50 mL of a pH 10 bicarbonate-carbonate buffer solution. Group 2 animals
each received the same amount of 59Fe as an amino acid chelate (glycine), which was
also buffered in 50 mL of the buffer solution. The rats in group 3 each received the 4
mc of 59FeCl3, which was buffered as before. In addition, each rat was also adminis-
tered 2 mg of DDVP immediately following dosing with the 59FeCl3. Finally, the rats
in group 4 were each administered the 4 mc of the buffered 59Fe amino acid chelate
(glycine) immediately followed by the dosing of 2 mg of DDVP similarly to group 3.
At 48 hours postdosing and 1 day before expected parturition, all of the animals
were sacrificed by asphyxiation with ether. Their fetuses were removed and each
assayed for 59Fe by liquid scintillation. Table 11.1 summarizes the results in cor-
rected counts per gram per minute per fetus, which was adjusted to reflect an equiva-
lent number of pups per dam.
When comparing the fetal uptake of the iron from the amino acid chelate source
to the chloride source, approximately 141% more iron was recovered from each fetus
taken from mothers receiving the amino acid chelate. When DDVP was adminis-
tered concurrently with either iron source, the expected increase in the fetal uptake
of iron from either source of iron due to placental vasodilation occurred, but the per-
centage difference between the two groups was greater than between the two groups
that did not receive DDVP.
Had the two sources of ingested iron been deposited into the maternal blood in
the same molecular form and if the original percentage increase in the fetuses were
completely due to more iron being delivered to the plasma of dams administered the
amino acid-chelated source of iron, then the percentage increase in fetal uptake of
iron between the two groups of dams receiving DDVP should have been identical
to the percentage increase between the two groups that did not receive the DDVP
(141%). Admittedly, there should have been a greater amount of fetal uptake of iron
by the fetuses from both groups of mothers following administration of DDVP, but
still, the percentage difference between the two groups should have been the same.
To the contrary, however, when the two iron sources were administered concurrently
with DDVP, the percentage difference of iron between the two groups increased from
TABLE 11.1
Mean 59Fe Recovery from Rat Fetuses
Whose Mothers Received 59Fe as
FeCl3 or Iron Amino Acid Chelate
with or without 2,2-Dichlorvinyl
Dimethyl Phosphate (DDVP)
Group 59Fea
Group 1: FeCl3
59 90.3b
Group 2: 59Fe AAC 127.0c
Group 3: 59FeCl3 + DDVP 111.1d
Group 4: 59Fe AAC + DDVP 293.5e
Source: Based on data from Ashmead, D, and

Graff, D, “Placental transport of che-
lated iron,” Proceedings of the
International Pig Veterinary Society,
Mexico City, Mexico, 207, July 1982.
a Corrected counts per minute per gram wet tissue
per fetus.
b/c Means is significantly greater at p < 0.05.
d/e Means is significantly greater at p < 0.05.
141% to 264%, which was an 87.2% increase (p < 0.5) compared to the original per-
centage differences between the mothers that did not receive DDVP. This indicated
that at least some of the plasma iron from the dams receiving the amino acid chelate
was in a different molecular form than was the plasma iron from dams receiving
ferric chloride.
The iron amino acid chelate had a calculated molecular weight of less than
300 Da. Rat iron transferrin has a molecular weight of between 67,000 and
83,000 Da.28,29 Thus, when the DDVP enhanced vasodilation within the placenta,
that action would have automatically favored the fetal uptake of a smaller iron mol-
ecule for two reasons.
First, in a given volume of blood, more iron could be “packed” into that volume
if it existed in the form of smaller molecules. The administration of ferric chloride
would have resulted in most of that source of iron arriving in the plasma bound to
transferrin.30 Transferrin binds a maximum of two atoms of ferric iron to each trans-
ferrin molecule.31 Thus, in the same volume of blood that contained one transferrin
molecule, approximately 280 atoms of iron chelated to the amino acid glycine in a
molar ratio of two amino acids to one atom of iron could potentially reside. Under
theoretical conditions, this would provide approximately 140 times more iron in that
same volume of blood.
The second reason for greater fetal uptake of smaller iron molecules focuses
on the transfer of the molecule from maternal blood to the fetus. Ionic iron from
maternal blood is taken up by the placental tissue, which then transfers the iron to
fetal transferrin and ultimately to the fetal tissue itself.32 While this system is highly
effective, it is, by its very nature, somewhat regulated.33 On the other hand, amino
acids are absorbed more efficiently as well as rapidly transported across the placenta
to support the growth of the fetus.34 Since essential amino acids and peptides are
capable of “smuggling” minerals not only into mucosal cells but also into other cells,
including fetal tissue that is supported by nutrients from maternal plasma, then when
the amino acid portion of the chelate is absorbed into the fetal tissue by active trans-
port, it carries or “smuggles” the mineral with it.16,35,36
In spite of this discussion, it should be remembered that not all of the iron in
the maternal plasma of the dams that were administered the amino acid chelate
remained in the original chelate or chelate/complex molecular form. In all probabil-
ity, a significant percentage of the absorbed amino acid chelated iron was hydrolyzed
in the mucosal cells, as previously described, prior to its iron being transferred to the
plasma. Only a portion of the plasma iron from the dams administered the amino
acid chelate remained in the original chelate or chelate/complex form regardless of
whether it was administered with or without DDVP.
A similar conclusion was drawn following the completion of another study, which
measured the uptake of not only 59Fe by fetuses from gestating mothers adminis-
tered either an amino acid chelate or ferrous sulfate sources, but also the 59Fe in the
blood of the mothers following dosing with either iron source. This study was also
conducted in synchronously impregnated gestating albino rats similarly to the previ-
ously described isotope study in rats. Both iron sources were administered 3 days
before expected parturition and their fetuses removed 48 hours later. Fetal tissue and
maternal blood were analyzed for 59Fe. There was no significant difference in 59Fe in
the hemoglobin or hematocrit of the blood samples from the mothers in either group,
even though the difference in the transfer of 59Fe to the fetuses from the two iron
sources was significantly different (p < 0.001).37 This implies that the portion of the
amino acid-chelated iron that was not hydrolyzed in the mucosal tissue was probably
transferred to the serum and was rapidly taken up by the placenta and ultimately the
fetus. The fact that 59Fe levels in the dams’ hemoglobin or hematocrit were not signifi-
cantly different between the two groups also indicates that the amount of iron used for
hemoglobin or hematocrit production was similarly regulated in both groups of dams.
That portion of the amino acid chelate that was probably hydrolyzed in the mucosal
cells and exited these absorptive cells was about the same as the amount of iron pro-
vided by the ferrous sulfate. If the female animals had not been pregnant, then there
would have probably been a significant increase and greater movement of production
of hemoglobin and hematocrit following ingestion of the iron amino acid chelate.38
While this discussion has focused primarily on the metabolism of iron in the
mucosal cell due to the ease of using iron isotopes to elucidate its movement and
metabolism in the body, other metals are similarly handled. For example, it has been
reported that most of the non-protein-bound copper in the blood plasma is chelated
to amino acid ligands, primarily histidine and cysteine.39,40 Martin suggested that
a large percentage of these copper chelates are formed in the mucosal cell before
being transferred to the plasma.41 If this is the case, it also demonstrates the possibil-
ity of an intact chelate or chelate/complex being able to exit from the mucosal cell
into the plasma. If that occurs, and he and others have presented considerable data
demonstrating that it does, then there is no reason not to believe that a portion of the
amino acid chelate or chelates/complexes that are absorbed intact into the mucosal
cell from the lumen can also exit to the plasma as the same intact molecules.
The point of the discussion in this chapter is simple. A percentage of the amino
acid chelate that is ingested and subsequently absorbed into the intestinal mucosal
tissue as either an amino acid chelate or chelate/complex is ultimately transferred
to the plasma as the same basic amino acid chelate molecule that was ingested.
Depending on its absorption pathway, the chelate may change molecular configura-
tions during its trip from the lumen to the plasma, but the metal remains attached to
the original amino acid ligands throughout the trip. When that occurs, due to a large
demand for amino acids by the body, the uptake of this amino acid chelate molecule
into the tissues from the plasma, based on metal analysis, appears to be more rapid
and in greater quantities compared to absorbed ions or minerals from that portion
of the amino acid chelate that was hydrolyzed in the mucosal cells. Further, as dis-
cussed in the following chapter, the metabolism of those intact chelates is different.
REFERENCES
1. Bezkorovainy, A, Biochemistry of Nonheme Iron (New York: Plenum Press) 108–109,
1980.
2. Bovell-Benjamin, AC, Viteri, FE, and Allen, LH, “Iron absorption from ferrous bisglyci-
71:1563–1569, 2000.
3. Olivares, M, Pizarro, F, Pineda, O, Name, J, Hertrampf, E, and Walter, T, “Milk inhibits
and ascorbic acid favors ferrous bis-glycine chelate bioavailability in humans,” J Nutr
127:1407–1411, 2007.
4. Finch, A, and Cook, J, “Iron deficiency,” Am J Clin Nutr 39:471–477, 1984.
5. Sillen, LG, and Mortell, AE, Stability Constants of Metal-Ion Complexes (London:
Chelates, (Springfield, IL: Thomas) 205–212, 1985.
chelates by utilization of ninhydrin,” AOAC poster, St. Louis, MO, September, 2004.
174, 1975.
proofs of chelation,” paper presented at International Conference on Human Nutrition,
Salt Lake City, UT, January 21–22, 1995.
12. Ashmead, HD, The Roles of Amino Acid Chelates in Animal Nutrition (Park Ridge, NJ:
Noyes) 464–465, 1993.
13. Ponka, P, Schulman, HM, and Woodworth, RC, Iron Transport and Storage (Boca
Raton, FL: CRC Press) 254, 1990.
14. Conrad, ME, Parmley, RT, and Osterloh, K, “Small intestine regulation of iron absorp-
tion in the rat,” J Lab Clin Med 110:418–426, 1987.
15. Greyton, A, Textbook of Medical Physiology (Philadelphia: Saunders) 768, 1971.
16. Payne, JW, “Transport and hydrolysis of peptides by microorganisms,” in Elliott, L, and
O’Connor, M, eds., Peptide Transport and Hydrolysis (Amsterdam: Elsevier) 307, 310,
320, 329, 1977.
17. Giese, AC, Proceedings of the International Pig Veterinary Society, Cell Physiology
(Philadelphia: Saunders College) 14, 1973.
18. Schepartz, B, Regulation of Amino Acid Metabolism in Mammals (Philadelphia:
Saunders) 33–35, 1973.
19. Ashmead, D, and Graff, D, “Placental transport of chelated iron,” Proceedings of the
International Pig Veterinary Society, Mexico City, Mexico, July 1982.
20. Bagley, DH, Zapoliski, EJ, Rubin, U, and Princiotto, JV, “Placental transfer of chelated
Fe in the guinea-pig,” Am J Obstet Gyn 102:291–296, 1968.
21. Beaconsfield, P, Birdwood, G, and Beaconsfield, R, “The placenta,” Sci Am 243:91–102,
August 1980.
22. Hoskins, FH, and Hansard, SL, “Placental transfer of iron in swine as a function of
gestation age,” Proc Soc Exp Biol Med 116:7, 1964.
23. Stremmel, W, Lotz, G, Niederau, C, Teschke, R, and Strohmeyer, G, “Iron uptake by
rat duodenal microvillous membrane vesicles: Evidence for a carrier mediated transport
system,” Eur J Clin Invest 17:136–145, 1987.
24. O’Donnell, MW, and Cox TM, “Microvilliar iron binding glycoproteins isolated from
the rabbit small intestine,” Biochem J 202:107–115, 1982.
amino acid chelates and inorganic salts,” in Subramian, KS, Iyengar, GV, and Okamoto,
K, eds., Biological Trace Element Research (Washington, DC: American Chemical
Society) 306–319, 1991.
26. Batte, EG, Fogg, TJ, and Schooley, MA, “Treatment of young animals with dialkyl beta-
chlorovinyl phosphate,” U.S. Patent 3,507,956, April 21, 1970.
27. Ashmead, HH, “Synergistic combination of metal proteinates with beta-chlorovinyl
dialkyl phosphates,” U.S. Patent 4,076,803, February 28, 1978.
28. Charlwood, PA, “Ultracentrifugal characteristics of human, monkey and rat transfer-
rins,” Biochem J 88:394–398, 1963.
29. Gordon, AH, and Louis, LN, “Preparation and properties of rat transferrin,” Biochem J
88:409–414, 1963.
30. Bates, GW, and Schlabach, MR, “The reaction of ferric salts with transferrin,” J Biol
Chem 248:3228–2323, 1973.
32. Fairbanks, VF, “Iron in medicine and nutrition,” in Shils, ME, Olson, JA, and Shike, M,
eds., Modern Nutrition in Health and Disease (Philadelphia: Lea & Febiger) 195, 1994.
33. Zimmerman, DR, “Iron in swine nutrition,” in Conrad, HR, Zimmerman, DR, and
Combs, GF, eds., Iron in Animal and Poultry Nutrition (West Des Moines, IA: National
Feed Ingredients Association) 31, 1980.
34. Crim, MC, and Munro, HN, “Proteins and amino acids,” in Shils, ME, Olson, JA, and
Shike M, eds., Modern Nutrition in Health and Disease (Philadelphia: Lea & Febiger)
26, 1994.
35. Fickel, T, and Gilvarg, C, “Transport of impermeant substances in E. coli by way of
oligopeptide permease,” Nat New Biol 241:161–163, 1973.
36. Ames, BN, Ames, GF, Young, JD, Tsuchiya, D, and Lecocq, J, “Illicit transport, the
oligopeptide permease,” Proc Natl Acad Sci USA 70:456–458, 1973.
37. Ashmead, H, Jensen, N, and Graff, D, Iron Metabolism as Related to Hemoglobin in
Pregnancy, Albion monograph, Albion Laboratories, Clearfield, UT, 1975.
38. Ashmead, HD, “Summary and conclusions,” in Ashmead, HD, ed., The Roles of Amino
Acid Chelates in Animal Nutrition (Park Ridge, NJ: Noyes) 463, 1993.
39. May, PM, Linder, PW, and Williams, DR, “Computer simulation of metal-ion equilibria
in bio-fluids: Models for the low-molecular-weight complex distribution of calcium (II),
magnesium (II), manganese (II), iron (III), copper (II), zinc (II) and lead (II) ions in
human blood plasma,” J Chem Soc Dalton Trans 588–595, 1977.
40. Perrin, DD, and Agarwal, RP, “Multimetal-multiligand equilibria: A model for biologi-
cal systems,” in Sigel, H, ed., Metal Ions in Biological Systems (New York: Dekker) v2,
167–206, 1973.
41. Martin, RB, “Complexes of a-amino acids with chelatable side chain donor atoms,” in
Sigel, H, ed., Metal Ions in Biological Systems (New York: Dekker) v9, 1–39, 1979.
12 Tissue Metabolism of
Amino Acid Chelates
As described in the previous chapters, a certain portion of an amino acid chelate or
chelate/complex that is absorbed into the mucosal tissue survives the natural hydro-
lytic activity occurring in those mucosal cells. This surviving portion is transferred
to the plasma intact, contributing, in part, to the greater tissue uptake of the mineral
from the plasma compared to inorganic metal salt sources.1,2 To a degree, this may
be due to the body responding to the amino acid chelate as if it were a proteinaceous
molecule and transferring it to the plasma as such. The percentage of metal coming
directly from an ingested amino acid chelate or chelate/complex that survives mucosal
cell hydrolysis is difficult to quantify because that percentage may vary from situation
to situation depending on other nutrients that are concurrently absorbed, the current
metabolic activity within the mucosal cell, the immediate nutritional need of the body
at a given time, and even the choice of ligands employed for chelating that mineral.
Thus, when investigating mucosal cell hydrolysis of amino acid chelates or chelate/
complexes, the question should not focus so much on how much of an ingested dose
is delivered intact into the plasma as it should on how the body responds to whatever
amount is delivered. That there is generally greater metal uptake from the amino
acid-chelated source into the various tissues and organs of the body has been amply
demonstrated,1,2 but is that uptake simply increased retention of those metals in the
tissues, or does that uptake provide intact active amino acid chelate metabolites that
can enhance or interfere with normal metabolic activities within the body?
While it is difficult to unequivocally prove that the ingestion of a specific metal
atom has resulted in a specific biological response, it is reasonable to state that if a
certain metabolic response occurred following ingestion of a specific mineral supple-
ment, that particular mineral supplement was probably responsible for the observed
response. Thus, animal reactions to amino acid chelate supplementation provide
great mediums for the study of the metabolic responses because changes in their per-
formance or production records that are easily measured can reflect metabolic activ-
ity. Furthermore, since diets are easier to control in animals than in humans, it allows
a single variable, such as effect of the inclusion or omission of a particular mineral or
amino acid chelate, to be examined. The studies discussed next illustrate this.
As swine breeding stock become older, the numbers of stillborn pigs normally
increases with each successive parity. Further, piglet mortality following live births
also tends to increase in litters born to older sows. With each additional parity, it
becomes more difficult to rebreed the sow and achieve a successful conception.
Generally, breeding sows become progressively thinner with age, and by the sixth
farrowing most reach the stage where it is not economically viable to retain them as
breeding stock.
185
TABLE 12.1
Effect of Iron Amino Acid Chelate on Piglet Mortality
Treatment Sows Parity
2 3 4 5 6 7 8
Average number born/litter 10.91 11.73 12.65 12.84 13.19 13.73 13.07
Average number dead/litter 0.73 0.93 1.40 1.11 1.84 1.18 2.47
Average number reared/litter 9.55 10.23 10.55 11.05 10.03 10.73 10.07
Mortality (% of number born) 12.47 12.78 16.60 13.94 23.96 21.85 22.95
Control Sows Parity

2 3 4 5 6 7 8
Average number born/litter 11.03 12.15 12.92 13.56 12.81 12.91 13.3
Average number dead/litter 0.82 1.85 1.28 1.61 .07 2.09 2.03
Average number reared/litter 9.41 9.81 10.68 10.94 10.94 9.27 10.25
Mortality (% of number born) 14.69 19.26 17.34 19.32 14.60 28.20 22.93
A trial involving 200 breeder sows of varying parities was used to test the long-
term effects of feeding those pigs an iron amino acid chelate supplement.3 All of the
animals enrolled in the study were the same genetic stock. Each sow was arbitrarily
assigned to one of two groups, treated or control. All of the pigs in the study received
the same types and quantities of feed and water throughout the study period except
for the sources of the supplemental iron fed to both groups. In the treatment group,
each animal was hand fed approximately 1.7 g of iron amino acid chelate (171 mg of
iron) daily as a top dressing on its feed beginning 28 days before expected farrowing.
The control sows received an equivalent amount of iron daily, but in the form of fer-
rous sulfate. The iron supplements were discontinued each time the sows farrowed.
The 28-day supplement periods described continued for 2.5 years, during which
time piglet mortality rates and sow breeding records were maintained. Table 12.1
summarizes the results related to piglet mortality in each parity.
Except for the sixth parity, the piglet mortality was lower in each parity of pigs
farrowed from the group of sows receiving the iron amino acid chelate compared to
the control group. In the sixth parity, the number of sows in the control group was
about half that of the treatment group, which affected the results for that parity and
resulted in the data for that parity being an outlier. In all of the other parities, the
number of sows in either group was about the same. The average mortality rate for
the eight parities of piglets farrowed from the sows receiving the iron amino acid
chelate was 17.8% compared to 20.4% in the piglets from sows fed ferrous sulfate.
This was a 14.5% reduction in the mean mortality rate.
The veterinarian who supervised the study concluded that the 171 mg of supple-
mental iron from the amino acid chelate that was provided daily to each sow for
28 consecutive days prior to farrowing resulted in lower piglet mortality rates, par-
ticularly as the sows’ parities increased. The improved livability of the piglets from
the sows in the treatment group indicated that supplementing iron amino acid chelate
Tissue Metabolism of Amino Acid Chelates 187
to the gestating sows during the last 28 days of pregnancy resulted in a metabolic
response to the chelate.
Defining why supplementing the iron amino acid-chelated source resulted in an
improvement in piglet livability was outside the purpose of this research. It could be
postulated from previously described rat experments1,4 as well as other swine stud-
ies1,4,5 that the piglets farrowed from the sows receiving the iron amino acid chelate
probably had higher iron levels in their tissues and organs than did the baby pigs far-
rowed from sows receiving ferrous sulfate. An optimum source of iron is essential for
maximum fetal growth.6 The larger a baby pig is at birth, the lower will be its mor-
tality rate, as reported by a study of over 17,000 baby pigs at Iowa State University.7
Thus, it may be that the baby pigs farrowed from sows fed the iron amino acid chelate
were slightly larger than the piglets coming from sows supplemented ferrous sulfate
and that increased birth weight contributed to the lower mortality rate.
Table 12.2 compares the reproductive performance of both groups of sows dur-
ing the study. The average total empty days in Table 12.2 was defined as the number
of days from weaning a litter until the occurrence of estrus, followed by rebreeding
and conception. If conception was not achieved following the initial breeding by
artificial insemination, the number of days that passed before rebreeding the sow a
second or possibly a third time to achieve conception was included in the total empty
days. In other words, empty days were the number of days from weaning a litter of
piglets to a new conception.
Table 12.2 indicates that the inclusion of the supplemental iron amino acid chelate
to the diet of the treatment group reduced the number of empty days as well as the
number of times the sows in this group had to be rebred by artificial insemination
to achieve conception. Furthermore, the differences between the sows in the treat-
ment group, which were fed the amino acid chelate, and the control group became
greater with each successive parity. Some of the older sows, particularly in the con-
trol group, were unable to conceive even after being rebred three or more times;
ultimately, they had to be removed from the herd. At the conclusion of the 2.5-year
study, 60% of the control sows had been culled from the original herd compared to
only 10% of the sows in the treatment group.
Because the differences between the two groups of sows became larger with each
successive parity, it was suggested that not only was more iron transferred to the
fetuses from sows receiving iron amino acid chelate but also greater tissue retention of
TABLE 12.2
Effect of Iron Amino Acid Chelate Intake on Reproductive
Performance of Sows
Treatment Control Improvement (%)
Conception rate on first service 91% 84% 8.3%
Conception rate on all services 96% 93% 3.2%
Weaning to service interval (days) 7.11 8.79 –19.1%
Average total empty days 10.42 17.27 –39.7%
the absorbed iron by the sows was also occurring. That increased iron retention could
only take place if at least part of the molecular forms of the plasma iron from the two
sources, the chelate and the sulfate, were different. If the iron from both sources were
in the same molecular form when it was presented to the pigs’ tissues, then the trend
lines showing any improvements from parity to parity between the two groups of sows
would have also been the same. Further, the metabolic response, as exhibited by the
metabolism or metabolic rates resulting from that iron, would have also been the same.
In this study, with each successive parity, the mean differences in reproductive
performance between the two groups became greater. Because the slopes of the two
trend lines demonstrated increasingly greater disparity between the groups, as a
function of time (increasing parities), that difference suggests that at least part of the
iron from the amino acid chelate was being transferred from the mucosal cells into
the plasma and ultimately the tissues in a molecular form that was different from the
iron that was being sourced from ferrous sulfate. It was this molecular difference
that facilitated the greater tissue retention of iron and ultimately the improvement in
reproductive performance.
Pigs are not the only animal species that has demonstrated different metabolic
responses through performance changes following supplementation of amino acid
chelates. It has been reported that increased intake of certain metal salts, including
manganese, copper, magnesium, and zinc, has resulted in increased milk produc-
tion in dairy cows.8–14 Thus, investigators decided to look at milk production as a
metabolic response in cows following ingestion of minerals sourced from amino
acid chelates or metal salts. Since amino acid chelates were known to be absorbed
and retained in greater quantities,15,16 it was believed that this source of minerals
would increase milk production even more.15
The study involved 135 registered Holstein cows that were matched for age
and lactation period, then divided into two groups.17 One group, the control group,
received a pelleted supplement as a top dressing that supplied 364 mg of manganese
(as oxide), 727 mg of zinc (as oxide), 182 mg of copper (as sulfate), and 355 mg of
magnesium (as oxide) per animal per day during the first trimester. During the sec-
ond trimester of lactation, the mineral supplementation was reduced to 128 mg of
copper, 256 mg of manganese, 511 mg of zinc, and 249 of magnesium per animal
per day. The treated group received the same amount of manganese, zinc, copper,
and magnesium in its pelleted supplements, except the sources of the minerals were
amino acid chelates. Individual supplementation, as described, commenced as each
cow freshened with her calf and continued through the first two trimesters of her
lactation period. At the end of the second trimester, all supplementation was dis-
continued until the cow calved again. At that point, the supplement program recom-
menced as described above. The same supplement program was repeated following
the third calving. In total, the study was 40 months in duration. Since each of the
cows was milked three times a day, the daily ration of pelleted supplement from
either source of minerals was divided into three equal parts and hand fed to each
cow at each of her milkings.
Other than the sources of the minerals in the two supplements, all of the cows
were treated similarly. They all received the same quantity and type of feed and
water daily and were housed on the same farm.
TABLE 12.3
Mean Milk Volume, Milk Fat, and Milk Protein by Lactation for Dairy
Cows Receiving Inorganic Minerals or Metal Amino Acid Chelates as
Nutritional Supplements
Milk Production (kg/day) Milk Fat Milk Protein
Treatment/ Mean ± Corrected % Mean (%) Mean
Lactation SEM 3.5% Fat ± SEM (kg/day) ± SEM (kg/day)
Inorganic
1 32.5 ± 0.3 35.1 3.78 ± 0.03 1.23 3.09 ± 0.01 1.00
2 35.1 ± 0.4 37.3 3.72 ± 0.04 1.31 3.11 ± 0.02 1.09
3 36.0 ± 0.3a 36.4c 3.54 ± 0.03 1.27c 3.01 ± 0.01 1.08c
Metal Amino Acid Chelates

1 32.6 ± 0.3 33.8 3.63 ± 0.03 1.18 3.08 ± 0.01 1.00
2 36.0 ± 0.3 37.9 3.68 ± 0.04 1.32 3.09 ± 0.02 1.11
3 40.1 ± 0.3b 41.3d 3.60 ± 0.03 1.45d 2.97 ± 0.01 1.19d
Source: Redrawn from Ashmead, HD, and Samford, RA, “Effects of metal amino acid chelates or
organic minerals on three successive lactations in dairy cows,” Int J Appl Res Vet Med
2:181–188, 2004.
Note: Values within variables and lactation having different superscripts are significantly different
(a/b p < 0.005; c/d p < 0.05).
At 1 week postcalving (day 7 of lactation), a 30-cc composite milk sample was

obtained from each cow in the study by taking a 10-cc sample at the commencement
of each of her three daily milking periods. The three 10-cc samples were then mixed
together to create the 30-cc composite sample. A second 30-cc composite milk sam-
ple was obtained at 90 days of lactation from each cow as described above. The third
30-cc composite milk sample was obtained in the same manner at 180 days of lacta-
tion. Each composite milk sample was analyzed for the percentage of milk fat and
percentage of milk protein. The total milk production from Dairy Herd Information
Association (DHIA) records for each animal was also recorded. Table 12.3 sum-
marizes the results.
The milk production, butterfat production, and milk protein production from
cows receiving the metal amino acid chelates were all significantly greater by the
third lactation period compared to the production of their herdmates. That indicated
increased tissue retention of the minerals from the amino acid chelates from one lac-
tation period to the next and the resultant positive effect of those amino acid chelates
on milk and milk component production. If the production differences in the first
lactation are compared, these differences are not particularly large. Nevertheless,
they tended to increase in disparity in the second lactation, and by the third lacta-
tion they were significantly different.
There is a second aspect to this study. Like the pigs described previously, succes-
sive lactations brought about by continuous calving in high-producing dairy cows
4.00
Metal amino acid chelates
3.90
Inorganic mineral salts
3.80
Body Condition Score
3.70
3.60
3.50
3.40
3.30
3.20
3.10
1 2 3 4 5 6 7 8 9 10 11 12
Month
Start lactation Mineral level End

& supplements decreased supplements
FIGURE 12.1 Body condition scores starting at the commencement of the third lactation
period for groups of dairy cows given two different forms of nutritional supplements dur-
ing the first 180 days of three lactations. (Redrawn from Ashmead, HD, and Samford, RA,
“Effects of metal amino acid chelates or organic minerals on three successive lactations in
dairy cows,” Int J Appl Res Vet Med 2:181–188, 2004.)
results in increasingly poorer body conditions that ultimately lead to culling. As

shown in Figure 12.1, cows receiving the amino acid chelates started their third
lactation with better body scores and continued to maintain significantly better body
scores throughout the lactation period (p < 0.01). This also demonstrated that more
of the amino acid chelates was retained in the tissues of the cows for a longer period
compared to the retention of the inorganic mineral sources in herdmates. That reten-
tion played a role in improving the body condition of the cows just as it did in the
pigs previously described.
It is obvious from the data summarized in Table 12.3 that greater mineral-related
metabolic activity occurred in the group of cows receiving the amino acid chelates.
Not only was more milk produced, but also the percentage of milk fat (butterfat)
was significantly higher, as was milk protein. This is all a result of increased
mineral-dependent metabolic activity. The minerals, and perhaps even the amino
acid ligands from the amino acid chelates, had to stimulate that metabolic activity
for those observed production increases to occur. More of the amino acid-chelated
mineral sources was deposited and retained in the tissues that were involved in that
greater metabolic activity because these minerals remained attached to the amino
acid ligands. The metabolic activity that resulted in greater milk production and
milk component production as well as improved body scores all demonstrated that
the amino acid chelates were able to participate in their assigned metabolic activities
at higher rates for a longer period of time than were the minerals that were sourced
from inorganic metal salts. While a different metabolic response occurred in the
cows, the metabolic responses evidenced by greater production for a longer period
of time along with improved body score were similar to the previous observations
related to the swine reproductive performance.
Both of these studies point to the possibility that at least two different forms of
the same mineral were being transferred to the plasma from the mucosal cells. The
absorbed iron, manganese, copper, magnesium, and zinc from the ingested inorganic
metal salts that reached the plasma were attached to carrier molecules designed to
transport these inorganic metal ions to the target tissues. When the same minerals were
absorbed into the mucosal cells as either amino acid chelates or chelate/complexes, as
was discussed previously, only a portion of these absorbed molecules probably hydro-
lyzed within the mucosal cell and were treated similarly to absorbed metal ions.
The portion of the amino acid chelates or chelate/complexes that was not hydro-
lyzed in the absorptive cell probably entered the plasma as the same amino acid
chelates originally ingested or as chelates/complexes. As the dairy cow study dem-
onstrated, once the “smuggled” minerals in the amino acid chelates were taken up by
the tissues, the bodies of the animals ingesting them responded differently compared
to the responses to the inorganic form of minerals. The mineral-dependent meta-
bolic activity associated with greater milk and milk component production as well as
improved body scores all increased. To be sure, the minerals from both sources were
used for the same mineral-dependent metabolic process, but because the minerals
in the amino acid chelates were bonded to amino acid ligands, their metabolic use
by the cows occurred at different rates and quantities compared to the inorganic min-
eral salts. Further, the amino acid chelate form of minerals appeared to be employed
preferentially when compared to the inorganic form.
The body’s turnover of absorbed metal ions that are sourced from metal salts
appears to be greater than is the turnover of the same metals from that portion of
the amino acid chelate that remains intact as it is moved from the lumen through
the mucosal cells into the plasma and finally into the tissues. This is evidenced by
greater tissue retention of the chelated mineral source with resultant increased meta-
bolic activity. This lower turnover has to be the result of bonding the mineral to the
amino acids. If an amino acid chelate or chelate/complex is delivered to the tissue
or organ as an intact molecule containing both the essential metal and the essential
amino acid ligand, the complete amino acid chelate or chelate/complex molecule
may be able to participate in certain metabolic activities in a balanced and somewhat
synergistic relationship, which would then cause increased retention of both com-
ponents. If a high amount of a certain mineral from an ionized source is delivered
to this same “metabolic center” without a corresponding amount of an essential (for
the metabolic process) amino acid, the excess metal will be endogenously excreted
until a balance (for metabolic activity) is achieved between the metal and whatever
amount of amino acid was available.
In another study involving 88 registered Holstein dairy cows, the researchers matched
and then divided the animals into two equal groups. Both groups were hand fed pelleted
supplements that supplied 364 mg of manganese, 727 mg of zinc, and 182 mg of copper
per animal per day for 190 days of lactation. The source of minerals for one group was
inorganic salts, whereas the source for the other group was amino acid chelates. Other
than the different sources of supplemental minerals, both groups were treated identically.
The 310-day milk production from cows receiving the amino acid chelates was
10.64% higher (p < 0.001) with 13.03% more milk fat (p < 0.001), 3.24% more milk
protein (p < 0.001), 11.46% more urea nitrogen (p < 0.01), 1.86% more milk lactose
(p < 0.10), and 2.51% more milk solids (p < 0.10). Concurrently, those same cows
excreted 6.81% less manganese (p < 0.05), 14.43% less zinc (p < 0.05), and 32.36%
less copper (p < 0.10) into their milk compared to the cows receiving the inorganic
minerals in their supplement.
Manganese, copper, and zinc are essential for milk production.18–20 Yet, this study
demonstrated that cows receiving these minerals from inorganic metal salt sources
did not retain as much of these required minerals in their tissues as did their herd-
mates provided exactly the same minerals in the amino acid-chelated form. One
of the mediums for endogenous excretion of excess minerals is milk.21 When the
cows are fed the inorganic salts of manganese, zinc, and copper, they endogenously
excrete significantly more of these minerals into their milk, resulting in less milk
and milk component production because the minerals are not retained and used in
the tissues for milk production. The metal amino acid chelates are held in the tissues
in greater quantities (less excretion into the milk) and influence increased metabo-
lism in the body (greater milk and milk component production).
While the dairy cow study dealt with the effect of endogenous excretion of miner-
als into the milk, the same concept has also been demonstrated in the isotope studies
discussed below. These studies examined the endogenous excretion of minerals from
different sources into the urine.1,22 In the first of the two experiments, a group of adult
male Sprague-Dawley rats was partially anesthetized with ether following a 12-hour
fast. While anesthetized, each was administered, by gastric gavage, a single dose of
32 mg of manganese sourced from either an amino acid chelate (8.96% methionine,
19.3% glycine, 34.1% aspartic acid, and 37.4% glutamic acid) or from manganese
chloride. Prior to administration, both manganese sources were dissolved in distilled
and deionized water. Each source contained 35 microcuries of 54Mn. For 14 consecu-
tive days following administration, the urine from each animal was collected and
assayed for 54Mn. At 14 days postdosing, all of the animals in both groups were sac-
rificed and their major tissues and organs assayed for 54Mn. Table 12.4 summarizes
the results of the tissue analysis.
Based on whole-body counts, the group of rats receiving the amino acid-chelated
source of manganese absorbed and retained approximately 3.39 times more 54Mn
in their organs and tissues than did the animals receiving the manganese chloride
source (p < 0.05). Even though the chelate group absorbed more manganese, their
mean 14-day total urinary excretion of 54Mn was approximately 30% less than that
of the group administered the manganese chloride. This is contrary to what one
would predict. In theory, if the animals absorbed more manganese, then logic would
dictate that their subsequent urinary excretion of that manganese should have also
been higher. The group administered the manganese chloride retained significantly
less manganese in their tissues and organs, but even so, they endogenously excreted
more manganese into their urine, suggesting uptake and excretion without retention.
It also points to the probability that two different molecular forms of manganese
were presented to the tissues from the plasma. If both sources of manganese had
been delivered to the tissues in the same chemical form, then the subsequent urinary
TABLE 12.4
Mean 54Mn in Selected Rat Tissue and Organs (corrected
counts/per minute/per gram wet tissue) Following
Administration of Mn Amino Acid Chelate or MnCl2
Tissue Mn Amino Acid Chelate MnCl2 Increase (%)
Heart 107a 36b 197
Liver 106a 56b 104
Kidney 97 80 21
Spleen 397a 190b 109
Lung 56 54 4
Small intestine 141 89 58
Muscle 28a 22b 27
Bone 266a 112b 138
p < 0.05.
a/b
excretion of the manganese, as a percentage of the absorbed dose, should have been
identical. The fact that the urinary excretion of manganese from the chloride source
was significantly greater, as a percentage of the absorbed dose, demonstrates that
this form of manganese was chemically different from at least part of the manganese
from the amino acid chelate at the time that the two manganese sources were taken
up by the tissue from the plasma.
Following its absorption into the mucosal cells, the chloride source of manga-
nese was presumably attached to cellularly produced ionic transport molecules.
The amino acid-chelated source, while utilizing those same transport molecules for
translocation of that portion of the absorbed dose that was rendered ionic after being
hydrolyzed within the mucosal cells, also delivered a second form of manganese into
the plasma, which was probably in the form of an intact amino acid chelate or chelate/
complex molecule. It is this second form of manganese that would have resulted in
the increased manganese retention within the tissues and organs and the lower rate
of endogenous urinary manganese excretion. Presumably, there was also a different
rate of manganese-related metabolic activity (based on the swine and bovine studies
discussed), but manganese metabolism was not measured in this study.
A second isotope study using laboratory rats also tended to reinforce these con-
clusions.1,22 Under light ether sedation, 18 fasted male Sprague-Dawley rats were
each administered 5 mg of zinc as either the amino acid chelate (methionine) or zinc
chloride by gastric gavage. Each dose of zinc from either source contained 5 micro-
curies of 65Zn as a tracer. Urine was collected for 7 days postdosing and assayed for
65Zn. At the conclusion of that 7-day urine collection period, all of the animals in
both groups were sacrificed and their tissues and organs removed and assayed for
65Zn. Table 12.5 summarizes the results for the tissue analysis.
The analysis of the tissues and organs in Table 12.5 for 65Zn indicated that the
animals receiving the amino acid-chelated form of zinc retained 11.6% more zinc in
their tissues and organs than did the rats administered the zinc chloride. In spite of
TABLE 12.5
Mean 65Zinc Levels in Certain Rat Tissues and
Organs (corrected counts/per minute/per gram
wet tissue) Following Administration of Zinc
Amino Acid Chelate or Zinc Chloride
Tissue Zn Amino Acid Chelate ZnCl2 Increase (%)
Muscle 186a 153b 22
Heart 1,457 1,433 2
Liver 10,250a 7,529b 36
Kidney 8,629 7,797 11
Brain 541a 444b 22
p < 0.05.
a/b
the greater zinc retention by the group receiving the amino acid chelate source, the
group administered the zinc chloride excreted 2.35 times more endogenous 65Zn into
their urine during the 7-day collection period than did the amino acid chelate group
(p < 0.05).
This lower zinc urinary excretion coupled with higher retention of zinc in the
group receiving zinc amino acid chelate compared to the group administered zinc
chloride suggested different metabolic responses to the two zinc sources. If all of
the zinc from both sources had been delivered to the tissue bonded to the same ionic
transport molecules, then the zinc excretion as a percentage of absorption of the two
sources of zinc would have also been the same regardless of how much was absorbed
or not absorbed. Instead, the endogenous excretion of zinc from the chloride source
was higher, indicating that, in this study, a different form of zinc was being taken
up from the plasma following zinc chloride administration compared to the amino
acid-chelated source.
It is interesting to note that in the two isotope studies with the rats that a different
amino acid chelate ligand resulted in a different level of endogenous metal excretion.
Both rat studies were conducted similarly. The combination of ligands employed
in chelating the manganese had a higher calculated stability constant than did the
single methionine ligand used to chelate the zinc.23 Amino acid chelates formed
with ligands having lower stability constants generally exhibit increased mucosal
cell hydrolysis following uptake.24 With increased mucosal cell hydrolysis, more of
the metal from the hydrolyzed chelate will be released and subsequently attached to
ionic transport molecules produced by the body. Thus, an amino acid chelate with a
lower stability constant would potentially have absorption, tissue uptake, and reten-
tion characteristics that would more closely mimic that of a metal ion than would
an amino acid chelate having a higher stability constant. This is possibly being seen
when the two rat studies discussed are compared. When the manganese was chelated
to the combination of amino acid ligands, the resulting chelate had a higher stabil-
ity constant compared to the chelate formed with methionine. When the stability
constant was increased, more metal, as a percentage of the dose, was retained in the
tissues. A comparison of these two studies provides a possible explanation for why
more of the zinc methionine chelate underwent hydrolyzation in the mucosal tissue
and resulted in a greater percentage of the dose from that source of zinc behaving
similarly to that of the absorbed ionic zinc from an inorganic salt source.
On a cellular level, what happens to the amino acid chelate once it arrives at the
tissue site is still open to debate. Cellular uptake from the plasma of minerals che-
lated to amino acids is reported to be similar to absorption of those same minerals
into the intestinal absorptive cells.25 Once delivered to the body cells by the plasma,
hydrolysis of an intact amino acid chelate or chelate/complex molecule could occur
within the tissue/organ cells if there is a cellular ligand with a potentially higher
binding constant requiring that mineral and if the cellular pH favors the hydrolysis
of that chelate/complex.
Another alternative, which was alluded to previously, is that the amino acid che-
late or chelate/complex continues to remain as an intact molecule following its uptake
by the tissue or organ cells, and the entire molecule thus participates in the meta-
bolic activities within those tissues or organ cells that require the presence of both
the metal and amino acid. This is the balance theory described above. A balance
hypothesis is not an unreasonable concept to consider, particularly if the metabolic
process within the cell were to require a chelate/complex similar to the chelate/
complex absorbed from the plasma.
Outside the structural requirements of minerals, a major role is their involvements
in enzymatic processes, many of which frequently require chelating or complexing
the needed mineral to amino acid ligands, all of which become part of the enzyme.
Depending on the enzyme examined, it frequently necessitates that both the amino
acid and the metal be present for the enzymatic system to function. If both are sup-
plied concurrently, the result may be greater metabolic activity, which, as noted in
both the pig and dairy cow studies, resulted in increased performance.
It will be recalled from Liebig’s law of the minimum that the nutrient in the rela-
tive minimum determines the rate of growth.26 The same law could be applied to
enzymatic or metabolic activity: The nutrient in the relative minimum determines
the rate of enzymatic or metabolic activity within the body. Perhaps by providing
both the required mineral and the amino acid together in a single molecule, the
homeostatic mechanism that would normally remove excess metal taken up from an
ionic source would instead be arrested because the amino acid ligand in the intact
chelate would balance out the higher metal uptake, and the two nutrients could func-
tion together in an enzymatic or metabolic process.
Homeostasis attempts to maintain balance within the body. Suppose a metabolic
process, such as a metal-activated enzyme, requires a specific mineral as well as
certain amino acids for it to function. If the metal from the inorganic salt source
were delivered to the enzyme requiring it, but concurrently there were deficiencies
of another enzymatic component, such as a specific amino acid, the cell will not
generally store the metal while it waits for the delivery of an appropriate amount or
type of amino acid. Instead, it will eliminate the metal, which at the moment it sees
as being in excess, resulting in increased endogenous excretion of the metal into the
urine, the saliva, the milk, the perspiration, the lower bowel, and so on.21 Thus,
the retention and increased utilization of the metal may, in part, depend on whether
it was originally chelated to an amino acid or amino acids prior to ingestion. While
not all of the absorbed amino acid chelate remains intact as a chelate or chelate/com-
plex molecule as it travels through the body to the locus of need, some of it probably
does,27 and it is that intact portion that modifies the mineral retention and rate of
metabolic activity associated with that mineral.
The proof that more of a metal chelated to a specific amino acid can be taken up
by certain tissues or organs that require both the metal and the amino acid is demon-
strated by the following study28: Zinc is not only essential for sexual maturation but
also for the functioning of the male sexual tissues and organs.29 In the male rat, the
testis and prostate contain significantly higher concentrations of zinc than do most
other tissues or organs assayed.30–32 The amino acid arginine is also essential for
proper sexual functioning in the male. In conjunction with zinc, it reduces rat ventral
prostate weight and dihydrotestosterone without affecting testicular function.32,33
Eighteen adult male Sprague-Dawley rats of similar age and weight (280 g) were
selected for the study.28 They were arbitrarily divided into three groups of six ani-
mals each. All had been raised on the same rat chow. None was considered zinc
deficient, based on the composition of their feed, which met NRC requirements for
zinc.34 On the day of the study, while under light anesthesia, each animal received a
single intravenous dose of 0.06 mg of zinc containing 9.2 microcuries of 65Zn. The
sources of zinc were as follows: for group 1, zinc chloride; group 2, zinc chelated to
glycine; and group 3, zinc chelated to arginine. At 24 hours postinjection, all of the
animals were sacrificed followed by removal of their testes, epididymis, and seminal
vesicles, and each of those organs or tissues was assayed for 65Zn. Table 12.6 sum-
marizes the results of those assays.
In theory, if the zinc uptake by the tissues and organs required that the metal be in
a single molecular configuration for uptake, the deposition of zinc into the analyzed
organs and tissues should have been the same for all three zinc sources because the
same quantity of zinc was injected intravenously into the rats, thus bypassing any
potential intestinal absorption hindrance. As noted, all of the rats had been fed a
zinc-sufficient diet, so none of their organs or tissues should have been zinc deficient.
And yet, depending on the source of zinc and its molecular form when injected into
TABLE 12.6
Mean Deposition of 65Zn from an
Intravenous Dose of Three Zinc Sources into
Male Sexual Organs and Tissues (corrected
counts/per minute/per gram tissues)
Tissue Arginine Glycine Chloride
Testes 1.26a 0.96c Negligibleb
Epididymis 1.03a 0.73c Negligibleb
Seminal vesicle 2.53a 1.98c Negligibleb
p < 0.05.
a/b
p < 0.10.
a/c
the animals, there were significant differences in the amounts of zinc uptake by the
sexual tissues and organs.
Both the glycine and the arginine chelates resulted in significantly greater uptake
of zinc into the testes, epididymis, and seminal vesicle compared to the uptake of zinc
from the chloride source. One could argue that zinc chloride is absent in the plasma;
thus, this zinc source would not have been compatible with the tissues, which would
have been the reason for it not being absorbed into the tissue and organs. While
in theory that is true, the zinc that was injected into the rats was no longer zinc
chloride per se. Prior to injection, the zinc chloride was dissolved in sterile deion-
ized water, resulting in the creation of ionic zinc. As it was slowly injected into the
blood of the rats, the zinc ions that were introduced into the plasma would have been
complexed to specific plasma zinc transporters that normally circulate in the blood.
These plasma zinc transporters would have taken the injected zinc ions to areas of
the body requiring zinc, including the sexual tissues and organs. If any tissue or
organ had a zinc need, in terms of a deficiency, the injected zinc ions from the chlo-
ride source would have been delivered to that tissue or organ. If an acute need did not
exist, the metal would probably have been excreted as endogenous zinc. Apparently,
because there was no acute zinc deficiency in the sexual organs (based on all of the
rats receiving a zinc-sufficient diet) and, concurrently, since these same tissues and
organs probably did not contain excess arginine, glycine, or other amino acid that
would have been required to balance with the zinc and metabolically function with
it, there was little or no need for additional tissue uptake of the ionic source of zinc.
Liebig’s law of the minimum came into play.
On the other hand, this study further demonstrated that when the metal was che-
lated to amino acid ligands and then injected into the plasma, the tissue uptakes of
those sources of zinc were greater compared to zinc chloride. The responses were
probably not so much responses to zinc, but instead they were responses to the amino
acid ligands that were bonded to the zinc. As the amino acid chelates entered the
plasma, either zinc was more likely to be taken up by the tissue or organ, particularly
if its amino acid ligands were also required. In the case of the dairy cows discussed
previously, this increased uptake would have been manifest as greater milk produc-
tion, more fat in the milk, more protein in the milk, and so on. In the case of the pigs, it
was improved reproductive performance. In the case of the male rats, it was increased
zinc content of their sexual organs and tissues to enhance sexual functioning.
This rat study also demonstrated that the quantity of tissue or organ uptake of a
metal from the plasma is also influenced by the choice of the amino acid ligand to
which the metal is chelated. In the case of the male sexual organs and tissues, argi-
nine was in greater demand for various metabolic activities within those organs and
tissues than was glycine. When the zinc was chelated to arginine, the uptake from
the blood of zinc arginate increased significantly compared to uptake of the zinc che-
lated to a glycine. This difference demonstrated that the initial response of the body
to an amino acid chelate is to the ligand, not the metal. Thus, while zinc glycinate
was absorbed in reasonable amounts, its uptake was still appreciably less than was
the absorption of the zinc chelated to arginine.
The law of the minimum also applies when comparing the uptake of zinc from the
two ligands. A certain amount of glycine is required by the male sexual tissues and
organs; so, as the glycine was absorbed into those tissues and organs, the zinc, which
was chelated to that glycine, went along for the ride. But in this case, the uptake of
zinc glycinate from the plasma was limited by the maximum glycine required by
those organs and tissues. When that glycine requirement was met, there was little
additional uptake of either the zinc or the glycine. Assuming the zinc glycinate was
not absorbed by other tissues within the body (which is not a realistic assumption
in vivo), then the remaining zinc glycinate that had been injected into the blood
would have been excreted as endogenous zinc similarly to the zinc chloride.
Because the demand for arginine by the sexual organs and tissues was greater than
was the demand for glycine, more zinc chelated to the arginine was taken up from
the blood. In the case of the zinc arginate, assuming an unlimited supply, uptake
from the plasma would have ceased only when the tissues/organs had achieved their
optimum arginine/zinc levels in a balanced ratio for the required metabolic purpose
that the two nutrients influence. When that point was reached, any excess zinc would
have been endogenously excreted. If the zinc arginate were intracellularly hydro-
lyzed, the excess arginine, if any, would have been used for growing new body tissue
or used to help generate more energy.
This leads to some interesting possibilities as far as oral intake of amino acid che-
lates is concerned. If the desired outcome resulting from the ingestion of a mineral
supplement is to effect a change in a metabolic rate within a specific organ or tissue,
the mineral should probably be chelated to an amino acid ligand that also partici-
pates in a major way in that desired metabolic process. The outcome of ingesting
such an amino acid chelate could be enhanced enzymatic activity, which would lead
to greater metabolic activity.35
REFERENCES
amino acid chelates and inorganic metal salts,” in Subramanian, KS, Iyengar, GV, and
2. Ashmead, HD, “A peptide dependent intestinal pathway for absorption of essential
minerals,” in Southgate, D, Johnson, I, and Fenwich, GR, eds,, Nutrient Availability:
Chemical and Biological Aspects (Norwich, UK: Royal Chemical Society) 122–124,
1989.
3. Darneley, AH, “Improving reproductive performance with iron amino acid chelate,” in
Ashmead, HD, ed., The Roles of Amino Acid Chelates in Animal Nutrition (Park Ridge,
NJ: Noyes) 251–267, 1993.
4. Ashmead, H, Jensen, N, and Graff, D, Iron Metabolism as Related to Hemoglobin in
Pregnancy, Albion Laboratories Monograph, Albion Laboratories, Clearfield, UT, 1974.
5. Brady, PS, “Evaluation of an amino acid iron chelate hematinic,” Report of Swine
Research 1975 (East Lansing: Michigan State University) 4–6, 1975.
6. Kühn, LC, Schulman, HM, and Ponka, P, “Iron-transferrin requirements and trans-
ferring receptor expression in proliferating cells,” in Ponka, P, Schulman, HM, and
Woodworth, RC, eds., Iron Transport and Storage (Boca Raton, FL: CRC Press)
149–191, 1990.
7. Ashmead, HD, “The role of iron amino acid chelate in pig performance,” in Ashmead,
HD, ed., The Roles of Amino Acid Chelates in Animal Nutrition (Park Ridge, NJ: Noyes)
217–218, 1993.
8. Cagliero, G, and Ashmead, D, “Use of amino acid chelates of iron for increased placen-
tal transfer in swine,” Vet Med/Small Anim Clin 78:753–757, 1983.
9. Achibald, JG, and Linquist, HG, “Manganese in cows’ milk,” J Dairy Sci 26:325–330,
1943.
10. Ballantine, HT, Socha, MT, Tomlison, DJ, Johnson, AB, Fielding, AS, Shearer, JK, and
VanAmstel, SR, “Effects of feeding complexed zinc, manganese, copper and cobalt to
late gestation and lactating dairy cows on claw integrity, reproduction and lactation per-
formance,” Prof Anim Sci 18:211–218, 2002.
11. Miller, WJ, Clifton, CM, Fowler, PR, and Perkins, HF, “Influence of high levels of
dietary zinc on zinc in milk, performance and biochemistry of lactating cows,” J Dairy
Sci 48:450–453, 1965.
12. Stevenson, JW, and Earle, IP, “Zinc in sows colostrums and milk,” J Anim Sci 23:300–306,
1964.
13. Campbell, MH, Miller, JK, and Schrick, FN, “Effect of additional cobalt, copper, man-
ganese and zinc on reproduction and milk yield of lactating dairy cows receiving bovine
somatotrophin,” J Dairy Sci 82:1019–1025, 1999.
14. Slama, AAK, Cajat, G, Abanell, E, Such, X, Casals, R, and Plaizats, J, “Effects of dietary
supplements of zinc-methionine on milk production, udder health and zinc metabolism
in dairy goats,” J Dairy Res 70:7–9, 2003.
15. Ashmead, HD, Ashmead, SD, and Samford, RA, “The effects of metal amino acid che-
lates on milk production, reproduction and body condition in Holstein first calf heifers,”
Int J Appl Res Vet Med 2:252–260, 2004.
made from various protein sources into rat jejunal slices in vitro,” Proc Utah Acad Arts
Lett Sci April 1970.
17. Ashmead, HD, and Samford, RA, “Effects of metal amino acid chelates or organic min-
erals on three successive lactations in dairy cows,” Int J Appl Res Vet Med 2:181–188,
2004.
18. Zin, Z., Waterman, DF, Neinken, RW, and Harmon, RJ, “Copper status and requirement
during the dry period and early lactation in multiparous Holstein cows,” J Dairy Sci
76:2711–2716, 1993.
19. Kirchgessner, M, and Weigand, M, “Optimal zinc requirements of lactating dairy cows
based on various dose-response relationships,” Arch Tierenahr 32:569–578, 1982.
20. Bentley, OG, and Phillips, PH, “The effect of low manganese rations upon the dairy
cows,” J Dairy Sci 34:396–403, 1951.
21. Miller, WJ, “New concepts and developments in metabolism and homeostasis of inor-
ganic elements in dairy cattle. A review,” J Dairy Sci 58:1549–1560, 1975.
22. Jensen, NL, “Biological assimilation of metals,” U.S. Patent 4,167,564, Washington,
DC, September 11, 1979.
26. McCullum, EV, A History of Nutrition (Boston: Houghton Mifflin) 92, 1957.
28. Ashmead, HH, and Graff, DJ, “Zinc amino acid chelates and male sex organs,” unpub-
lished monograph, Albion Laboratories, Clearfield, UT, July 1987.
29. Prasad, AD, “Deficiency of zinc in man and its toxicity,” in Prasad, AS, ed., Trace-
Elements in Human Health and Disease (New York: Academic Press) v1, 3–12, 1976.
30. Tipton, IH, Schroeder, HA, Perry, HM, and Cook, MJ, “Trace elements in human tis-
sue. Part III. Subjects from Africa, the Near and Far East and Europe,” Health Phys
11:403–451, 1965.
31. Gilbert, IGF, and Taylor, DM, “The behavior of zinc and radio-zinc in the rat,” Biochem
Biophys Acta 21:546–551, 1956.
32. Mawson, CA, and Fischer, MI, “Zinc and carbonic anhydrase in human semen,”
Biochem J 55:696–700, 1953.
33. Fahim, MS, Wang, M, Sutcu, MF, and Fahim, Z, “Zinc arginine, a 5a-reductase inhibi-
tor, reduces rat ventral prostate weight and DNA without affecting testicular function,”
Andrologia 25(6):369–375, 1993.
34. National Research Council, Nutrient Requirements of Laboratory Animals (Washington,
DC: National Academy Press) 1978.
35. Ashmead, HH, Ashmead, HD, and Graff, DJ, “Amino acid chelated compositions for
delivery to specific biological sites,” U.S. Patent 4,863,898, September 5, 1989.
13 Some Metabolic
Responses of the Body
to Amino Acid Chelates
Absorption is the process by which simple nutrient components resulting from

digestion, or in some cases, ingestion alone, move out of the digestive tract into the
absorptive cells and from there are sent to the bloodstream or, to a lesser extent,
the lymphatic system. Some of the steps in the absorptive process require a specific
metabolic activity to facilitate or enhance the absorption. Ultimately, this metabolic
activity can also influence the bioavailability of certain absorbed nutrients. For
example, as has been discussed, once in the mucosal cell, part of the amino acid che-
late may be hydrolyzed before transferring any of its components or the remaining
intact chelate or chelate/complex molecules into the plasma. Following hydrolyza-
tion, the released metal ion is able to participate in several metabolic activities that it
is excluded from being involved in while still in the amino acid chelate state. While
those metabolic activities could be considered to be part of this particular bioavail-
ability process, for the purposes of this discussion they are included in the absorptive
process only. Thus, based on this, absorption is not the same as bioavailability.
Bioavailability requires that the absorbed nutrients be taken up by the tissues and
organs from the blood and in most cases metabolized anabolically or catabolically
before participating in metabolic processes that provide energy, facilitate growth and
maintenance of body tissue, regulate body processes, or facilitate sexual reproduction.
Thus, using iron ethylenediaminetetraacetic acid (EDTA) as an example, its absorp-
tion does not generally equate to bioavailability of that particular chelate. While it
is readily taken up from the gastrointestinal tract, most of that form of iron is sub-
sequently eliminated into the urine as the original intact iron EDTA molecules.1–3
Significant metabolism of that source of chelated iron does not generally occur
because its stability constant is considerably higher than the potential formation con-
stants of most of the ligands created by the body.
Most of the data discussed in the previous chapters has focused on the absorp-
tion of various amino acid chelates. Very little attention has been directed toward
their bioavailabilities following absorption. Actual direct evidence of participation
of a specific mineral in a biochemical process is difficult to generate. One can see
the involvement of a mineral in a given biological process, but is the mineral that is
initiating that activity the same metal that was ingested, or was the mineral removed
from body reserves to meet the body’s biochemical need? Without the employment
of isotopes, it is difficult to ascertain this. Nevertheless, if a known biochemical
process requires the participation of a specific metal, and if the ingestion of that
201
metal, in the form of an amino acid chelate, results in a greater metabolic response
compared to the response following ingestion of the same metal as an inorganic salt,
then it can be said with reasonable certainty that the bioavailability of the amino acid
chelate was greater when compared to that of the metal salt. The greater bioavailabil-
ity (that is, increased metabolic activity) that one observes can initially result from
greater intestinal absorption of the amino acid-chelated mineral, but ultimately, it
must rely on greater tissue or organ uptake and retention of the amino acid chelate/
complex after it exits from the mucosal cell.
Numerous studies, including the ones that follow, suggest that when compared
to inorganic metal salts, greater bioavailability, as measured by metal-related meta-
bolic activity, generally occurs when the nutritional metals are sourced from amino
acid chelates. Bioavailability studies do not usually focus on absorption alone, but
instead, they look at a specific biological activity within the body that is dependent
on the presence of a certain metal.
To illustrate, 40 human infants, ranging in age from 6 to 36 months, with diagnosed
iron deficiency anemia were matched by age, sex, and hemoglobin concentrations
and subsequently assigned to one of two groups, each containing 20 subjects.4 Each
child received a daily dose of 5 mg of iron per kilogram of body weight for 28 days.
The source of iron provided to one group was ferrous sulfate, while the other group
received ferrous bisglycinate chelate. As shown in Table 13.1, treatment with either
source of supplemental iron significantly (p < 0.001) raised the hemoglobin levels in
all of the infants when compared to their mean initial hemoglobin concentrations, indi-
cating that both iron sources were bioavailable; however, that is only part of the story.
When the blood ferritin levels from both groups were measured, only those
infants who had received the supplemental amino acid chelate source of iron showed
a significant increase (p < 0.05) in their iron stores, thus demonstrating, in a differ-
ent metabolic way, significantly greater bioavailability of the amino acid chelate. As
seen in Table 13.2, there was no significant difference in the mean basal levels of the
serum concentrations of ferritin between the two groups at study commencement.
TABLE 13.1
Effect of 28 Days of Treatment with Ferrous Bisglycinate
Chelate or Ferrous Sulfate on Hemoglobin Levelsa
Mean ± SD (g/dL)
Group n Basal Posttreatment Changeb
Ferrous sulfate 20 8.7 ± 1.64 10.5 ± 0.81 1.8 ± 1.59
Ferrous bisglycinate chelate 20 8.0 ± 1.49 10.5 ± 0.22 2.5 ± 1.31
Source: Data from Pineda, O, and Ashmead, HD, “Effectiveness of treatment of

iron deficiency anemia in infants and young children with ferrous bis-
glycinate chelate,” Nutrition 17:381–384, 2001.
a Analysis of variance showed no difference within basal or treatment groups but
significant differences between basal and treatment values (p < 0.001).

b Posttreatment minus basal values.
Some Metabolic Responses of the Body to Amino Acid Chelates 203
TABLE 13.2
Effect of 28 Days of Treatment with Ferrous Bisglycinate
Chelate or Ferrous Sulfate on Serum Ferritin Levelsa
Mean ± SD (mg/L)
Group n Basal Treatment Change
Ferrous sulfate 20 43.9 ± 23.20 70.4 ± 45.71 26.5 ± 53.98
Ferrous bisglycinate chelate 20 53.6 ± 36.18 128.2 ± 86.94 74.6 ± 75.90
Source: Data from Pineda, O, and Ashmead, HD, “Effectiveness of treatment of iron
deficiency anemia in infants and young children with ferrous bis-glycinate
chelate,” Nutrition 17:381–384, 2001.
a Analysis of variance showed no difference within basal groups. There was a signifi-
cant treatment effect for iron bis-glycinate chelate (p < 0.005) but not for FeSO4.
There was also a significant difference (p < 0.05) between the groups after 28 days.
Twenty-eight days later, however, the mean serum ferritin level of the infants receiv-
ing the ferrous bisglycinate was significantly higher when it was compared to the
mean ferritin level of the group of infants administered exactly the same amounts
of iron as ferrous sulfate (p < 0.005). Further, only the group taking the amino acid
chelate supplement had a significant mean increase in serum ferritin compared to the
basal level (p < 0.05).
What this study suggested, in practical terms, is that when the iron was chelated
to glycine and formed a bisglycinate molecule, enough of that source of iron was
absorbed and metabolized to both correct the present and prevent future iron defi-
ciency anemia in those infants. When 5 mg of iron per kilogram of body weight was
provided for 28 days, only the amino acid-chelated source of iron was able to replenish
the depleted ferritin stores in the body. The bioavailability of the amino acid-chelated
source of iron was obviously greater than that of ferrous sulfate. The apparent bio-
availability for either source of iron was calculated using the following formula:
[body weight {kg}X 0.075][∆Hb {g/L}X

% AB = 3.44 + [∆ferritin {μg/L}X 10] ×100
total Fe intake [mg]

When the apparent bioavailability of the ferrous sulfate in these anemic infants
was calculated, it was 26.7%. The apparent bioavailability of the chelated iron bis-
glycinate was 90.9%.4
This greater bioavailability suggests that part of the chelated iron was proba-
bly transferred to the plasma still attached to its original amino acid ligands, and
this may explain why there was a significant increase in ferritin concentrations.
Regression analysis demonstrated that iron absorption from either source, as mea-
sured by hemoglobin, was similarly regulated, suggesting that a majority of the
ferrous bisglycinate was absorbed by facilitated diffusion.4 As hemoglobin levels
increased in either group, iron absorption decreased. The similar biological response
to the two forms of iron, in terms of hemoglobin increases, suggests that most of the
iron from the amino acid chelate that was incorporated into hemoglobin molecules
was in a similar (ionic) form as the iron sourced from ferrous sulfate. Otherwise, the
regulation of the two sources of iron would not be similar. Thus, there was signifi-
cant mucosal cell hydrolysis of the absorbed iron bisglycinate before transferring the
iron for hemoglobin production.
To the contrary, however, since there was a significantly different biological
response as it related to ferritin concentration, this suggests that the iron used to
replenish the iron stores in the group receiving the amino acid-chelated source may
have been in a different form since only the ferritin in the group receiving the amino
acid-chelated source of iron demonstrated an increased ferritin concentration. This
suggests that the iron that was delivered to the ferritin may have been in the form of
intact chelate or chelate/complex molecules since the response in the bodies of the
group ingesting the iron bisglycinate was different. If that occurred, this could mean
that the greater biological response in that group was probably due to the infants’ bod-
ies responding to that intact portion of the amino acid chelate as if it were an amino
acid. In a metabolic sense, however, all that can really be said is that much of the
absorbed iron from the ferrous bisglycinate source that was not initially used to build
hemoglobin was, instead, employed in replenishing the low body stores of iron in the
infants rather than being treated similarly to the ferrous sulfate that was absorbed.
In another metabolic study, the effect of supplemental copper on diamine oxidase
(DAO) activity was measured. DAO activity is responsible for the oxidative removal
of the amino group from organic compounds that contain two amino groups, such as
putrescine, cadaverine, and histamine. Plasma DAO has been reported to be low in
individuals who are suffering from a copper deficiency; in fact, the measurement of
plasma DAO activity can easily differentiate between copper deficiency and copper
sufficiency in the body.5
Using plasma DAO activity as a measure of the bioavailability of two copper
sources, a double-blind, crossover study was designed. It employed 12 males and
12 females ranging in age from 22 to 45 years. The study consisted of six 3-week
supplementation regimens in which 3 mg of daily copper supplementation, as cop-
per sulfate, was administered to the subjects for the first 21 days. This was followed
by a 21-day placebo period during which time none of the individuals received a
copper supplement before crossover and the commencement of administration of
3 mg copper from chelated copper bisglycinate for another 3 weeks. That second
supplementation period was followed by another 21-day placebo period. At the con-
clusion of this placebo period, all of the subjects in both groups were given 6 mg of
copper from the amino acid-chelated source daily for 21 days, followed by a final
third 3-week placebo period. At the end of every supplemental and placebo period,
blood samples were obtained from each individual in the study. The samples were
subsequently analyzed for DAO activity. Table 13.3 summarizes the results.5
From a metabolic point of view, it is first important to note, in this study, that
plasma DAO activity was almost 22% higher following supplementation with 3 mg
of copper as the amino acid chelate compared to the DAO activity when the same
amount of copper from copper sulfate was administered. When the supplemental
Some Metabolic Responses of the Body to Amino Acid Chelates
TABLE 13.3
Diamine Oxidase (DAO) Responses to Copper Status at Different Levels of Copper Supplementation in the
Total Group and in Men and Women Separatelya
Crossover 1 Crossover 2 Crossover 3
3 mg CuSO4/day Placebo 1 3 mg CuGC/day Placebo 2 6 mg CuGC/day Placebo 3
Mean SE Mean SE n Mean SE Mean SE n Mean SE Mean SE n
Serum diamine oxidase (U/L)
Total 1.34b 0.31 0.30 0.29 16 1.63b 0.48 0.44 0.11 20 2.00b 0.11 0.89 0.22 17
Men 0.87b 0.27 0.27 0.17 7 1.21b 0.36 0.47 0.01 9 1.55b 0.49 0.44 0.28 7
Women 1.64b 0.47 0.33 0.17 9 1.92b 0.41 0.42 0.23 11 2.37b 0.79 1.16 0.32 10
Source: Data from Kehoe, CA, Turley, E, Bonham, MP, O’Connor, JM, McKeown, A, Faughnan, MS, Coulter, JS, Howard, AN, and Strain, JJ,
“Response of putative indices of copper status to copper supplementation in human subjects,” Br J Nutr 84:151–156, 2000.
Note: CuGC/day, copper-glycine chelate.
a Mean values with standard errors.
b Mean value was significantly different from placebo group, p < 0.01.
205
intake of copper as the amino acid chelate was doubled to 6 mg, the mean DAO
activity increased 49% compared to the DAO activity resulting from the copper sul-
fate supplementation and 22% compared to when the 3 mg of copper supplementa-
tion as copper bisglycinate was provided.
While this study underscored increased copper bioavailability in the amino acid-
chelated form, the DAO activity should have been twice as high following supple-
menting with 6 mg copper compared to when 3 mg of copper was provided when
both copper sources were administered in the amino acid-chelated form. This lower
DAO activity, as a percentage increase, following the 6 mg copper supplementation
period points to a regulatory mechanism within the body that appears to control
the uptake of the amino acid-chelated form of copper. If there were no regulatory
mechanism in place, then doubling the amount of copper provided during the third
supplemental period should have resulted in a doubling of the DAO activity. Since
that did not occur, this study demonstrated that if more copper continued to be made
available at a cellular level, then greater uptake, which in this case was reflected
as greater enzymatic activity specific to copper, can potentially occur up to a cer-
tain point. That point of maximum activity is probably dictated by several factors,
including the balance theory. Copper is required not only for DAO activity but also
for other enzymatic systems. When all of the enzymes are replete with copper or
its ligand, then the enzymatic activity associated with the copper or its ligand will
not generally increase past that point, even when more of the bioavailable chelated
source of copper is provided.
A second observation also arose from this study. When comparing the DAO
activity at the end of each of the 3-week placebo periods following the two 3-mg
copper supplementation periods, there was always greater DAO activity resulting
from an amino acid chelate supplementation period compared to copper sulfate
intake period. This suggested greater tissue uptake and retention of the copper from
the amino acid chelate source was occurring. This indicated greater copper bioavail-
ability from the amino acid chelate source.
Dietary magnesium has been reported to have several benefits for individuals who
are suffering from asthma.6,7 These include inhibiting the contraction of vascular and
bronchial smooth muscles, inhibition of the release of both acetylcholine from cho-
linergic nerve terminals and histamine from mast cells, and finally, the promotion
of nitric oxide prostacyclin generation. Consequently, using these activities as mea-
surements for a research study, a randomized, double-blind, placebo-controlled trial
involving 17 asthmatic subjects was designed to evaluate the effects of magnesium
amino acid chelate on the frequency and duration of asthmatic symptoms.
During the first 7 days of this asthmatic study, each enrolled individual recorded
all of his or her daily food and beverage consumption. These food diaries were
analyzed, and each subject was advised of any minor dietary alterations that he or
she would be required to adhere to throughout the study period to reduce the total
magnesium intake from his or her diet to 100 to 200 mg per day. Each subject was
then required to follow this new low-magnesium diet for two periods of 3 weeks
each, plus one additional week during which time either magnesium or placebo sup-
plementation was provided. After the initial 3-week washout period, every person
enrolled in the trial received a capsule daily containing either a placebo or 400 mg
of magnesium, as an amino acid chelate, for 1 week. This was followed by another
3-week washout period, during which time no magnesium supplementation was
taken by either group. As noted, except for magnesium supplementation, the low-
magnesium diet was adhered to throughout the entire 7-week study period that fol-
lowed the initial week of dietary analysis.
Clinical measurements of all subjects enrolled in the trial included forced expi-
ratory volumes for 1 second (FEV1) and the amount of methacholine required to
produce a 20% fall in FEV (PD20, FEV1) from baseline. These measurements were
recorded at the beginning and the end of each period, that is, the first 3-week wash-
out period, the 1-week supplementation period, and the second 3-week washout
period. The variations in the peak expiratory flow (PEF) rate, bronchodilator use,
and asthma symptom scores were also recorded.
While there were no significant differences in FEV1, PD20, FEV1, or mean PEF
variations between the group receiving the magnesium treatments compared to the
group provided the placebo, asthma symptom scores were significantly lowered dur-
ing the time that the treatment group ingested the magnesium amino acid chelate
supplement compared to the placebo group (p < 0.05), suggesting bioavailability
of the magnesium amino acid chelate. The investigators were able to demonstrate
that the intake of magnesium amino acid chelate was associated with improvements
in asthmatic patients’ symptom scores.6 The results were not clear enough, however,
to truly demonstrate magnesium bioavailability.
In a second, 60-day, double-blind study, also involving magnesium amino acid
chelate and individuals afflicted with asthma, 37 patients (aged 7 to 19 years) were
randomly divided into two groups.7 One group (n = 18) received 300 mg of magne-
sium daily as chelated magnesium bisglycinate, and the other group (n = 19) received
glycine daily, which was considered the placebo. The only treatment difference in
the study was the supplemental magnesium. Both groups consumed normal diets
that contained magnesium. Unlike the previous study, the diets were not controlled.
Both groups received inhaled fluticasone (250 mg twice a day) and salbutamol as
needed. Bronchial reactivity was evaluated with a methacholine test at the com-
mencement of the study and again at 60 days postinitiation of the study.
The patients receiving the supplemental magnesium amino acid chelate exhibited
fewer asthma exacerbation episodes (p < 0.05) and required less inhaled salbutamol
(p < 0.05), which Figure 13.1 summarizes.7
Bronchial reactivity to methacholine was also significantly reduced (p < 0.05).
The skin responses to recognized antigens were decreased in those patients receiv-
ing the magnesium amino acid chelate supplement. After 60 days of continuous
magnesium bisglycinate supplementation, the required dose of methacholine neces-
sary to induce a 20% fall in the FEV1 was significantly higher in the magnesium
group compared to the placebo control group (p < 0.05) (Figure 13.2).7
This study demonstrated not only the bioavailability of the magnesium amino
acid chelate but also that a long-term supplemental period is required to elicit a
measurable asthmatic response in terms of statistics. In the first study, the magne-
sium treatment period was only 1 week in duration. While there was a significant
reduction in overall asthmatic symptoms in that first study, FEV was not affected.
When the magnesium amino acid chelate supplement period was extended to
25 *p=0.0024
20
Number of Days
*p=0.0001
15
Mg
10
Placebo
5
0
Asthma Inhaled
Exacerbation Salbutamol
FIGURE 13.1 The clinical evaluation of asthmatic children and adolescents who received
300 mg of supplemental magnesium as a bisglycinate chelate or placebo (glycine) supplemen-
tation for 60 days. (Redrawn from Gontijo-Amaral, C, Ribeiro, M, Gontijo, L, Condino-Neto,
A, and Ribeiro, JD, “Oral magnesium supplementation in asthmatic children: A double-blind
randomized placebo-controlled trial,” Eur J Clin Nutr 1–7, June 2006.)
*p<0.05
0.8
Methacholine (µM)
0.6
Mg
0.4
Placebo
0.2
0
Before After
FIGURE 13.2 The amount of methacholine required to induce bronchial challenge in

patients receiving either magnesium bisglycinate chelate or placebo (glycine). (Redrawn from
Gontijo-Amaral, C, Ribeiro, M, Gontijo, L, Condino-Neto, A, and Ribeiro, JD, “Oral magne-
sium supplementation in asthmatic children: A double-blind randomized placebo-controlled
trial,” Eur J Clin Nutr 1–7, June 2006.)
2 months, FEV was significantly reduced. This suggests that when individuals are
suffering from a particular mineral deficiency that is related to a clinical condi-
tion, such as in the case of asthma and magnesium deficiency, the correction does
not generally respond to a “quick fix” because the magnesium deficiency probably
goes much deeper than the asthmatic symptoms of the deficiency would suggest.
Further, there are probably other magnesium requirements within the body that
are also deficient and need to be addressed concurrently to those associated to
asthma. A 1-week supplementation period, while having an impact on asthmatic
symptoms, appeared to be an insufficient period of time to totally replete the body
with a sufficient amount of magnesium to correct the deficiency experienced by the

entire body. Thus, in the second study, when the magnesium-supplemented period
was extended, the asthmatic symptoms associated with the deficiency were more
fully addressed, and the differences between the treatment and placebo groups
became significant.
This same concept was also suggested in the study involving the anemic infants.
Repleting low hemoglobin levels with iron supplementation is the first priority of the
body in cases of iron deficiency anemia. Low serum ferritin levels are subsequently
addressed, but at a different priority level, particularly if the iron source is not as bio-
available, as was the case of the ferrous sulfate compared to the ferrous bisglycinate.
In a double-blind, crossover study,8 which also involved magnesium amino acid
chelate, thrombin clotting times and platelet adhesion were measured following daily
supplementation of either 2 g per day of fish oil or 160 mg of magnesium per day
sourced from an amino acid chelate. Fish oil has been reported to be useful in reduc-
ing platelet adhesion.9 Seven volunteers were first given the fish oil regime for 7 days,
followed by a 7-day washout period. They then received the magnesium amino acid
chelate supplementation for 7 days. After each supplementation period, blood samples
were obtained from each subject and analyzed. The tests included the baseline and test
period serum magnesium levels, thrombin clotting times, and platelet adhesiveness.
The results showed no significant difference in the serum magnesium concentra-
tions following supplementation with either the fish oil or the magnesium amino
acid chelate. The absence of an increased serum magnesium level even after supple-
mentation suggested that either the magnesium from the amino acid chelate was
not absorbed or the serum magnesium level was tightly controlled by the body and
any excess magnesium entering the plasma was rapidly transferred from the blood
serum into the tissues. The fact that this study subsequently demonstrated several
significant metabolic responses following ingestion of the supplemental magnesium
indicates that the magnesium from the amino acid chelate was in fact absorbed,
but instead of that uptake being reflected by increased magnesium serum levels, it
was quickly taken up by the cells and tissues, where it enhanced magnesium-related
metabolic activity, including those activities described next.
The thrombin clotting time increased from 10.6 ± 0.3 seconds to 10.8 ± 0.3 sec-
onds (p = 0.05) following fish oil supplementation. When the magnesium amino acid
chelate was supplemented, the mean thrombin clotting time increased from 10.9 ±
0.2 seconds to 12.0 ± 0.06 seconds. This was an 11% increase in the amount of time
required to clot the blood of individuals receiving the magnesium amino acid chelate
compared to those same subjects following ingestion of the fish oil. The difference
between the two treatments was significant (p = 0.006). Furthermore, supplementing
with fish oil did not reduce platelet adhesiveness, which actually increased from 33.86
± 8.4% to 42.29 ± 6.4%. When the magnesium amino acid chelate was supplemented,
platelet adhesion decreased from 35.17 ± 5.9% to 21.17 ± 9.1%. Supplementing the
magnesium amino acid chelate resulted in almost a 40% decrease in platelet adhe-
sion compared to a 24.9% increase following fish oil supplementation, which was
a significant difference between the two treatments (p = 0.02). This study demon-
strated significant bioavailability of the magnesium amino acid chelate. When it was
ingested, thrombin clotting times increased, and platelet adhesion decreased.
Non-insulin-dependent diabetes is frequently associated with a moderate zinc

deficiency. Individuals affected with this disease tend to exhibit lower-than-normal
serum zinc concentrations, higher zinc excretion, changes in taste acuity, and an
impaired immune function.10 A zinc deficiency also affects the enzymatic activity
of zinc-dependent 5′-nucleotidase, as well as the insulin-like growth factor I (IGF-I)
and lipoprotein oxidation.10 With this in mind, a study was designed to examine the
effect of zinc amino acid chelate supplementation on type 1 diabetes.11
Sixty postmenopausal women (mean age 60 ± 12 years) were divided into two
groups of 30 individuals each. One group received 30 mg of zinc daily as a che-
lated zinc bisglycinate for 21 days. The other group received a placebo consisting
of maltodextran for the same period. Each subject enrolled in the study was taking
either sulfonylurea or insulin to control her diabetic symptoms, but none was tak-
ing any form of zinc supplementation. Blood was drawn from each woman on day 0
(at the commencement of the study) and again on day 22 (24 hours following receiv-
ing the final dose of zinc). No food was consumed for at least 3 hours prior to each
blood draw. The blood samples were assayed for plasma zinc, 5′-nucleotidase activ-
ity, lipoprotein oxidation rates, and plasma IGF-I concentrations.
As expected, the plasma zinc levels were low in all of the subjects at the com-
mencement of the study. When the zinc amino acid chelate supplement was pro-
vided, it significantly elevated the zinc serum levels in all of the women receiving
the chelate (p < 0.05). Nevertheless, their plasma zinc concentrations remained lower
than the mean level observed in nondiabetic controls taken from another recently
conducted study that was similar in population and design.11 This suggested, as was
the case with the magnesium studies previously described, that perhaps a longer
zinc supplement period was required to totally replete zinc body stores. Or perhaps,
as in the case of iron and hemoglobin and serum ferritin, certain mineral deficien-
cies in the body are addressed before others, and plasma zinc has a lower prior-
ity. The initial zinc plasma 5′-nucleotidase activity was also extremely low in all
of the diabetic patients, but following zinc amino acid chelate supplementation, as
seen in Figure 13.3, it was significantly increased (p < 0.05).11 While the reasons
for the lower-than-normal zinc plasma levels and lower-than-expected zinc-related
metabolic activity were outside the scope of this study, they do point to greater zinc
excretion or increased zinc requirements of individuals suffering from diabetes even
when zinc amino acid chelate supplementation is employed compared to normal
healthy individuals. Nevertheless, the significantly higher plasma 5′-nucleotidase
activity demonstrated that increased zinc activity occurred following zinc amino
acid chelate supplementation.
The ingestion of the zinc amino acid chelate also resulted in significantly increased
concentrations of IGF-I (p < 0.01) in the study population,11 as seen in Table 13.4.
The mean oxidation lag time for low-density lipoprotein and very low-density lipo-
protein increased following the supplementation with zinc amino acid chelate, but
these increases were not significant. The investigators concluded that zinc status may
not be a major factor in lipoprotein oxidation, whereas it is directly involved in the
IGF-I and 5′-nucleotidase activities.
There are two forms of superoxide dismutase (SOD) found in mammalian tissue.
One is Cu/Zn SOD, which is located in the cytoplasm of most cells. This enzyme
p<0.05
0.60
Pre
5´ nucleotidase Activity (µ/L)

0.48
Post
0.36
0.24
0.12
0.00
Placebo Zinc
Treatments
FIGURE 13.3 The effect of daily supplementation of 30 mg of zinc from zinc amino acid
chelate on plasma 5′-nucleotidase activity in subjects with non-insulin-dependent diabetes
mellitus. (Redrawn from Bolstein-Fujii, A, DiSilvestro, RA, Frid, D, Katz, C, and Malarkey,
W, “Short-term zinc supplementation in women with non-insulin dependent diabetes mel-
litus: Effects on plasma 5′-nucleotidase activities, insulin-like growth factor I concentrations,
and lipoprotein oxidation rates in vitro,” Am J Clin Nutr 66:639–642, 1997.)
TABLE 13.4
Effects of Zinc Supplementation on Plasma Insulin-Like
Growth Factor I (IGF-I) Concentrations
Insulin-like Growth Factor I (mg/L)
Supplementation Group Presupplementation Postsupplementation
Placebo
Median (n = 20) 112 116
Below median (n = 10) 76 ± 10a 92 ± 12
Above median (n = 10) 199 ± 26 209 ± 13
Zinc
Median (n = 20) 157 156
Below median (n = 10) 104 ± 9 153 ± 16b
Above median (n = 10) 187 ± 8 154 ± 20
Source: Data from Bolstein-Fujii, A, DiSilvestro, RA, Frid, D, Katz, C, and

Malarkey, W, “Short-term zinc supplementation in women with non-
insulin dependent diabetes mellitus: Effects on plasma 5′-nucleotidase
activities, insulin-like growth factor I concentrations, and lipoprotein
oxidation rates in vitro,” Am J Clin Nutr 66:639–642, 1997.
a x ± SEM.
b Significantly different from presupplementation, p < 0.01.

is activated by copper and stabilized by zinc. The other type of SOD is Mn SOD.
The manganese-activated SOD is present in cell mitochondria. Regardless of their
locations, both enzymes have a similar function, that of converting superoxide into
hydrogen peroxide and oxygen.12
– cu/zn
2 O 2 + 2 H+ SOD or Mn SOD H2 O2 + O2

Mn SOD is a largely prokaryotic enzyme that contains 4 moles of manganese in

each mole of that enzyme. While the metal catalyzes the SOD activity by undergo-
ing valency changes during the catalytic process, the reaction is more complex than
a simple oxidation-reduction reaction. When the manganese is removed from the
enzyme, its activity ceases. Replacement of the manganese with other metals does
not restore the activity. This makes Mn SOD specific for manganese.12 Thus, a con-
tinuous supply of manganese either from the diet or from body stores is essential for
optimal functioning of Mn SOD.
The bioavailability of manganese amino acid chelate, as it relates to Mn SOD
activity, was examined in a double-blind study13 involving 22 nonsmoking females
with a mean age of 23.9 years. The volunteers were randomly divided into two
groups. One group of 9 individuals was administered 15 mg of manganese daily
as an amino acid chelate. The other group of 13 subjects was given a placebo. The
supplementation period was 120 days. Blood was drawn from each person at 20, 60,
90, and 120 days following study initiation and the lymphocytes from each sample
analyzed for Mn SOD activity. As seen in Figure 13.4, the women receiving the man-
ganese amino acid chelate supplement demonstrated significantly greater Mn SOD
activity in their lymphocytes compared to the lymphocyte Mn SOD activity in the
group receiving the placebo (p < 0.01).13
0.7
Changes in Lymphocyte Mn SOD Activity
0.6 Mn Amino
Acid Chelate
0.5
Group
0.4
(U/mg protein)
0.3
0.2
0.1 Placebo
0
–0.1
–0.2
–0.3
0 20 40 60 80 100 120
FIGURE 13.4 Actual changes in Mn SOD activity with their trend lines showing the change
in Mn SOD activity in lymphocyte during 120 days of daily supplementation with 15 mg of
manganese as an amino acid chelate or placebo. (Redrawn from Davis, CD, and Greger JL,
“Longitudinal changes of manganese-dependent superoxide dismutase and other indexes of
manganese and iron status in women,” Am J Clin Nutr 55:747–752, 1992.)
The trend line for Mn SOD lymphocyte activity in Figure 13.4 is almost flat in
the group of women receiving the placebo. Those women were probably obtaining
sufficient manganese from their diets, body stores, or both to support the level of
lymphocyte Mn SOD activity seen in Figure 13.4 but not to significantly increase
that activity. On the other hand, the trend line for the lymphocyte Mn SOD activity
in the group of women receiving manganese amino acid chelate increased over time,
indicating not only absorption but also manganese retention as a function of time.
This led to the increased Mn SOD activity over time. Since the only measured vari-
able in the study was the supplementation of the manganese amino acid chelate, and
since manganese is essential to activate certain forms of SOD, one could conclude
that the amino acid chelate was responsible for the increased lymphocyte Mn SOD
activity and thus was a measure of its bioavailability. This study did not address
what the consequences of increased Mn SOD activity were, only that it did increase.
Nevertheless, it is well known that increased Mn SOD activity does have a positive
impact on overall health.
In addition to Mn SOD, Cu/Zn SOD is found in the body. As stated, both forms
are involved in converting superoxide, a toxic by-product produced by the body, to
hydrogen peroxide and water. Unlike Mn SOD, which is in the mitochondria, Cu/Zn
SOD is found elsewhere in the cytoplasm of the cells. Its activity is dependent on
the presence of copper to activate the enzyme and zinc to stabilize it.12,14 Cu/Zn
SOD activity can be influenced by supplementation of both metals, as the following
study demonstrates.
The 250-head Oklahoma State University beef cattle herd was divided into
two groups.14 At the time of the division, a 7-cc blood sample was obtained from
each animal by jugular puncture and analyzed for erythrocyte Cu/Zn SOD activ-
ity. Following the measurement of Cu/Zn SOD activity, the feed for both groups of
animals was supplemented with 226 mg of copper per day (as either copper sulfate
alone or a combination of 199 mg from copper amino acid chelate and 27 mg from
copper sulfate) and 821 mg of zinc per day (as either zinc sulfate alone or 699 mg
from zinc amino acid chelate and 122 mg from zinc sulfate). The supplementation
program was continued for 60 days, after which time a second blood sample was
obtained from each cow. These blood samples were also analyzed for erythrocyte
Cu/Zn SOD activity.
There was a significant increase in erythrocyte Cu/Zn SOD activity in those
animals receiving the supplement containing the amino acid-chelated minerals
(p < 0.05). Their herdmates, which had received the same quantity of copper and
zinc but in a completely inorganic salt form, exhibited decreased erythrocyte Cu/Zn
SOD activity during the same 60-day period. Table 13.5 summarizes these results.14
Since the animals were treated identically, it was speculated that the reason for the
decline in the Cu/Zn SOD activity in the cattle receiving the inorganic metal salts
alone was due to the minerals being administered during a period of rapid growth in
the animals coupled with the lower bioavailability of the salt form of minerals. When
providing a supplement consisting exclusively of inorganic metal salts, the bioavail-
ability of that source of minerals appeared to be insufficient to support the total opti-
mal needs of the growing animals, whereas in the group fed the amino acid chelates,
mineral bioavailability was sufficient to meet the needs of the growing animals in
TABLE 13.5
Mean Changes in Cu/Zn SOD Activitya
in Bovine Erythrocytes
Difference AAC/Salt
Bleeding AAC Inorganic (%)
Initial 52.9b 59.0 10.3
Final 59.1c 56.8 4.1
Increase (%) 11.7 –2.2 —
Source: Data from Kropp, JR, “Reproductive performance of

first calf heifers supplemented with amino acid chelated
minerals,” in Oltijen, J, Horn, G, Geisert, R, and Yates,
L, eds., Animal Science Research Report (Stillwater,
OK: Oklahoma State University) 35–43, 1990.
a Units/milligram protein-RBC lysate.
b/c Significant at p < 0.05.
that group in addition to increasing their Cu/Zn SOD activities. Regardless of the
reason for the decline in Cu/Zn SOD activity in the group of animals receiving
the inorganic salt form of minerals, this study underscored the greater bioavailability
of copper and zinc when they were sourced from amino acid chelates compared to
the inorganic salt.
Cu/Zn SOD activity has also been reported to play a role in reducing or pre-
venting ethanol toxicity (alcoholism). Ingested ethanol is normally oxidized by
zinc-activated alcohol dehydrogenase (ADH). When ADH activity is overwhelmed
by excessive ethanol intake, increased production of superoxide radicals coupled
with reduced tissue levels of copper and zinc result. In cases of inadequate copper
and zinc intake or where there are insufficient body stores of these metals, Cu/Zn
SOD activity is reduced. With the lower Cu/Zn SOD activity and increased levels of
superoxide radicals, the consequence is increased physical dependency on ethanol.15
Ashmead and Graff published a study15 in which a group of 20 ethanol-depen-
dent (alcoholic) adult male rats were randomly divided into two groups of 10 animals
each. Each animal in the treatment group received 100 mcg of supplemental zinc and
45 mcg of supplemental copper, both as amino acid chelates, on a daily basis for a
period of 21 days. The zinc and copper amino acid chelates were dissolved in dis-
tilled water and administered by gastric gavage. The control group did not receive any
supplemental zinc or copper, but based on analysis of their diet, they had free-choice
access to adequate amounts of both zinc and copper. Like the treated group, their food
contained 20 ppm of zinc (ZnCO3) and 15 ppm of copper (CuCO3). This amount of
zinc and copper in the rat chow met the National Research Council (NRC) dietary
requirements for these two metals.16 During the 21-day period that the zinc and copper
amino acid chelates were being administered to the treatment group, both groups of
ethanol-dependent animals had free access to food and water as well as to a solution
containing a mixture of 5% ethanol and 95% water.
Water/Ethanol Mixture Consumption No supplementation

20 Cu/Zn supplement
(ml)/animal/day
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Day
FIGURE 13.5 Mean daily ethanol/water mixture consumption for 21 days in ethanol-
dependent rats. (Redrawn from Ashmead, HD, and Graff, DJ, “Decreasing ethanol con-
sumption in ethanol-dependent rats through supplementation of zinc and copper amino acid
chelates: A preliminary study,” J Appl Res 6:19–27, 2006.)
Figure 13.5 summarizes the mean ethanol/water mixture consumption of the

two groups of rats commencing 24 hours following initiation of the copper and
zinc supplementation in the treatment group.15 The average ethanol/water mixture
intake of the control group was 2.3 ± 0.72 mL per day compared to 0.1 ± 0.02 mL
in the group of animals receiving the amino acid chelate supplement. This 96%
decrease in ethanol/water mixture intake by the treatment group over the 21-day
period was significant (p < 0.001). In the group of rats that did not receive the sup-
plemental zinc and copper amino acid chelates, voluntary consumption of ethanol/
water mixture averaged 11.35 times more per day. The investigators postulated that
the forced ingestion of the zinc and copper amino acid chelate supplementations led
to an increased Cu/Zn SOD activity in that group of rats, which in turn manifested
itself as a reduced physical need to consume the ethanol/water mixture.15
Figure 13.6 shows the water consumption for the two groups of rats.15 The ani-
mals receiving the zinc and copper amino acid chelate supplement consumed a daily
mean amount of 33.9 ± 7.9 mL of water compared to 12.9 ± 9.5 mL of water by the
nonsupplemented group (p < 0.001). This did not include the ethanol/water mixture.
When the total liquid consumption per day was calculated, the nonsupplemented
group consumed 24.3 ± 10.3 mL of liquid per day compared to 34.8 ± 7.8 mL of
liquid for the group of rats administered the copper and zinc amino acid chelates
(p < 0.05). The difference between the two groups was a result of the nonsupple-
mented group obtaining more of its liquid from the water/ethanol mixture and less
from the water alone compared to the chelate-supplemented group. When the rats
consumed the ethanol/water mixture, there seemed to be less desire for water alone.
This study demonstrated the effect of increased zinc and copper bioavailability on
ethanol toxicity.
As reported previously, copper, manganese, zinc, magnesium, and potassium all
contribute metabolically to dairy milk production as well as butterfat (fat) and protein
60
Water Consumed (ml)/animal/day

50
40
30
20
10
Cu/Zn supplementation
No supplementation
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Day
FIGURE 13.6 Mean daily water consumption for 21 days in ethanol-dependent rats.
(Redrawn from Ashmead, HD, and Graff, DJ, “Decreasing ethanol consumption in ethanol-
dependent rats through supplementation of zinc and copper amino acid chelates: A prelimi-
nary study,” J Appl Res 6:19–27, 2006.)
in that milk.17 Several studies, previously reported, have demonstrated this, including
one in which two groups of Holstein first-calf heifers from the same herd and with
the same genetics were given a daily supplement of these minerals as either amino
acid chelates or metal salts. The supplements commenced at calving and continued
through 180 days of lactation. (Since, by definition, potassium cannot be chelated,
it was supplied as an amino acid complex defined by the Association of American
Feed Control Officials [AAFCO] to the animals receiving the chelate supplement.)
As shown in Table 13.6, for the total lactation period of 305 days, milk production
was 5.2% greater, with 9.3% (p < 0.01) more total fat and 8.1% (p < 0.05) more total
milk protein produced by those heifers that had received the metal amino acid che-
late supplement compared to their herdmates supplemented with the same metals
as inorganic salts.17 The increased milk production and increased milk component
production continued through the 305-day lactation period even through the amino
acid chelate supplementation was only administered for 180 days.
While milk and milk component production demonstrated greater mineral bio-
availability of the amino acid chelates, these increases are not the whole picture.
Before the conclusion of a lactation cycle, dairy cows must rebreed and become
pregnant to repeat the lactation cycle. The same minerals that influence milk pro-
duction can also affect reproduction. The mean number of days from parturition to
estrus in the group receiving the inorganic mineral supplement averaged 130.5 days
compared to 89.0 days for the chelate group (p < 0.01). From a commercial as well as
a biological point of view, the 31.8% reduction in open days was highly significant.
The animals receiving the inorganic mineral sources required an average of 2.58
services (by artificial insemination) for each successful conception compared to an
average of 1.50 services per conception in the cows fed the amino acid chelates. This
Some Metabolic Responses of the Body to Amino Acid Chelates
TABLE 13.6
Milk and Milk Component Production Data in First Calf Heifers Given Supplements of Either Inorganic
Metals (IOMs) or Amino Acid Chelates (AACs) in a 305-Day Milking Period
Mean Milk Mean Production/ Mean Total
Production, Day for 305 Days Mean Milk Mean Total Mean Milk Milk Protein
Study Groups 305 Days (kg) (kg) Fat (%) Milk Fat (kg) Protein (%) (kg)
IOM group 10,047 ± 1,456.36 32.94 3.57 ± 0.48 355.9 ± 46.11a 3.01 ± 0.21 302.1 ± 40.76c
AAC group 10,568 ± 911.76 34.65 3.76 ± 0.35 389.1 ± 47.02b 3.06 ± 0.17 326.4 ± 29.89d
Difference IOM/AAC (%) 5.2 5.2 5.32 9.3 3.3 8.1
Source: Data from Ashmead, HD, Ashmead, SD, and Samford, RA, “Effects of metal amino acid chelates on milk production, reproduction
and body condition in Holstein first calf heifers,” Int J Appl Res Vet Med 2:252–260, 2004.
a/b Significant difference at p < 001.
c/d Significant difference at p < 0.05.
217
41.9% reduction in the number of times a cow required rebreeding was also highly
significant (p < 0.01).17
In terms of mineral bioavailability, this study demonstrated that the minerals con-
tained in the amino acid chelates were not only absorbed in greater quantities but
also were translocated more rapidly and in greater amounts to the organs and tissues
that were initially responsible for lactation and subsequently to those organs and
tissues that were involved in reproduction. This resulted in the cows being able to
produce more milk and milk components while concurrently, over time, obtaining
sufficient mineral nutrients also to repair their sexual organs and tissues following
the normal damage resulting from the previous calving. Thus, with more rapid heal-
ing they were able to be rebred sooner. Both biological responses indicated signifi-
cantly greater bioavailability resulting from the amino acid chelates. From this and
other studies reported in previous chapters, it has been seen that not only are the
inorganic mineral salt sources of minerals absorbed in lesser amounts, but further,
smaller amounts of the absorbed minerals appear to be deposited in organs and tis-
sues. Thus, it took a longer period of time for cows receiving the inorganic salt form
of minerals to heal following calving, and until they healed, they could not return to
estrus or conceive when rebred.
Ingestion of metal amino acid chelates does not always guarantee increased min-
eral bioavailability. There are numerous factors that can affect mineral bioavailabil-
ity, including health and diet.18–20 As shown in Figure 13.7, when the body is exposed
to an infectious disease, one of its very first responses is to depress plasma amino
acids, iron, and zinc. This is followed by urinary retention of zinc. At the height of
the fever that is associated with the illness, there is a negative balance of nitrogen
(amino acids), magnesium, and zinc that will not be resolved until well into the con-
valescent period.18 Thus, during an illness, many nutrients, even if they are absorbed,
are sequestered and are thus not as available for essential metabolic activity as they
would have been in times of good health.
The following clinical study illustrates the effect of one such disease on the bio-
availability of iron amino acid chelate: A group of 100 gestating sows that were the
same age and breed were randomly divided into two groups of 50 pigs each. During
the last 28 days of their pregnancy, one group received 125 mg per day of iron as an
amino acid chelate, while the sows in the other group received the same amount of
iron but in the form of ferrous sulfate. At farrowing, the number of stillborn pigs was
11.7% higher in the pigs born to mothers that had received the ferrous sulfate. The
greater livability of piglets from mothers receiving the iron amino acid chelate was
presumably due in part to their 1.2% heavier mean birth weight. Heavier pigs at birth
result in lower piglet mortality rates both prenatally and postnatally.20 A heavier
birth weight also results in heavier pigs at weaning and at market.20
Contrary to expectations, however, following the births of the pigs in both of
these groups, there was no difference in their postnatal mortality rates. Furthermore,
instead of the pigs from the chelate group continuing to grow at a faster rate, their
mean 25-day weaning weights were also about the same as the pigs farrowed from
the sows fed ferrous sulfate. In reviewing why the expected differences between the
two groups did not occur as it had in other studies, it was discovered that the pig-
lets from the mothers that were fed the amino acid-chelated iron were infected with
Phagocytic activity
Depression of plasma amino acids, Fe and Zn
Saluresis Retention of urinary PO4 and Zn
Increased secretion of glucocorticoids and growth hormone
Increased deiodination of thyroxine
Increased synthesis of hepatic enzymes
Secretion of “acute phase” serum proteins
Carbohydrate intolerance
Increased dependence on lipids for fuel
Increased secretion of aidosterone and ADH
Negative Balances Begin – N, K, Mg, PO4, Zn, and SO4
Retention of body salt and water
Increased secretion of thyroxine
Diuresis Return to positive balances
Fever
Incubation Illness Convalescent

Period Period
Moment of Exposure
FIGURE 13.7 Schematic representation of the sequence of nutritional responses that evolve
during the course of a “typical” generalized infectious illness. (Redrawn from Beisel, WR,
“Magnitude of the host nutritional responses to infection,” Am J Clin Nutr 30:1236–1247,
1977.)
subclinical swine erysipelas, a disease that occurs frequently in suckling pigs. (The
other group of baby pigs was disease free.) The infected animals exhibited fevers that
were accompanied by high temperatures (104–108°F/40–42°C). Based on the find-
ings of Beisel,18 the natural biological response of the sick baby pigs would have been
to sequester the extra iron they had received from the amino acid chelate source fol-
lowing farrowing but which was not utilized until the pigs returned to health.
The baby pigs from mothers fed amino acid chelates commenced life with the
proper fetal iron nutrition foundation for faster growth and reduced mortality. The
heavier birth weight of the piglets in that group indicated that the amino acid chelate
had accomplished its assigned role during gestation by being transferred from mater-
nal plasma to the fetus in greater amounts when compared to the transfer of iron from
ferrous sulfate. Iron is essential for the synthesis of purines, which becomes the basis
for the pigs to manufacture body protein for growth both prenatally and postnatally.
During the period between birth and weaning, however, the increased storage of iron
in the tissues that was received during gestation from the mothers consuming the
chelate was prevented from being utilized to maximize the growth due to the baby
pigs’ biochemical response to erysipelas. Iron demand from body reserves in new-
born pigs is very high. When the iron is sequestered in the hemosiderin molecules
rather than being available to assist in protein production, growth is retarded, which
is exactly what happened in this baby pig study. Thus, as this example demonstrated,
a disease can reduce amino acid chelate bioavailability even if the absorption is high.
The second example shows how dietary components are also able to affect the bio-
availability of an amino acid chelate.21 A group of nine Angus steers was fed an iden-
tical diet for 120 days. Starting with a balanced ration, the investigators added 750 mg
of iron, 0.35% sulfur, and 10 mg of molybdenum per kilogram of feed. All of these
are known absorption and metabolism antagonists of copper, and when combined,
they are particularly aggressive in affecting copper uptake from the feed. In addition
to feeding these copper antagonists, each animal received 10 mg of supplemental
copper, as an amino acid chelate, per kilogram of feed. This supplemental copper
was in addition to the 20 mg of copper per kilogram of feed copper in the basal diet.
Liver and blood samples were taken at study commencement and again at 60, 90,
and 120 days following study initiation. The liver samples were assayed for copper
by plasma emission spectroscopy. The blood serum was also assayed for copper
by plasma emission spectroscopy in addition to having ceruloplasmin activity mea-
sured. Table 13.7 summarizes the results.21
The feeding of the copper antagonists had an impact on both copper amino acid
chelate absorption and bioavailability. The copper in the blood serum declined rap-
idly during the first 90 days of the study, suggesting that the dietary antagonists
were probably affecting its transfer from the mucosal tissue into the plasma. Copper
will initially bind to thiomolybdenate to form an insoluble and unavailable copper
compound. Once the copper that is available is absorbed into the mucosal tissue, it
continues to be affected by the dietary antagonists that are also absorbed into the
mucosal tissue. This results in less copper being transferred to the plasma. This
lower copper uptake was accompanied by increased ceruloplasmin activity, which
was partially associated with the growth of the animals. Much of the copper required
to support that ceruloplasmin activity appeared to be sourced from the liver, which
reflected a significant decline in heptic copper coupled with little copper replace-
ment. The inability of the livers to maintain their initial copper concentrations also
indicated lower bioavailability of the copper amino acid chelate.
TABLE 13.7
Changes in Copper and Copper-Related Activity
in Liver and Blood
Initial 60 Days 90 Days 120 Days
Liver 174.34 62.59 53.38 41.32
Blood serum (mg/dL) 0.89 0.73 0.63 0.74
Ceruloplasmin activity 42.44 51.67 49.78 61.44
Source: Data from Ashmead, HD, and Ashmead, SD, “The effects of
dietary molybdenum, sulfur and iron on absorption of three
organic copper sources,” J Appl Res Vet Med 2:1–9, 2004.
Obviously, the overall bioavailability of the copper amino acid chelate was
affected by the combination of the very high iron, sulfur, and molybdenum intake
from the feed. Much of the antagonistic activity probably occurred in the mucosal
tissue, where the amino acid chelate would have been subject to the normal hydro-
lytic activity that occurs in this tissue. It is also in the mucosal cells where the combi-
nation of iron sulfur and molybdenum have their initial negative impact on copper.21
Once the copper from the chelate was released into the cytoplasm, it would have
immediately been acted on by the iron, sulfur, and molybdenum in such a way that it
prevented its transfer and utilization by the steer’s bodies.21
These studies have been summarized simply to underscore the importance of
mineral bioavailability to promote metabolic performance. This book has focused on
how an amino acid chelate is absorbed from the gastrointestinal tract into the blood.
It has provided a small amount of data indicating that the uptake of the amino acid
chelates from the plasma by the body tissue is more efficient than the uptake of the
same metal when it is presented to those tissues, as a cation, via a transport molecule
sourced from the plasma, and that is a true measure of bioavailability.
Without attempting to identify every biochemical or metabolic pathway open to
a particular amino acid chelate or the metal derived from an amino acid chelate
that has been hydrolyzed in the mucosal cell, once the metal from either derivative
arrives in the tissue, this chapter has demonstrated that certain metabolic activities
are enhanced as a result of having ingested those amino acid chelates. Thus, not only
is their absorption from the gastrointestinal tract increased, but also the bioavailabil-
ity of the minerals from those amino acid chelates is greater when compared to the
ingestion of nonchelated minerals.
REFERENCES
1. Bates, GW, Boyer, J, Hegenauer, JC, and Saltman, P, “Facilitation of iron absorption by
ferric fructose,” Am J Clin Nutr 25: 983–986, 1972.
2. Rubin, M, Houlihan, J, and Princiotto, JV, “Chelation and iron metabolism I: Relative
iron binding of chelating agents and siderophilin in serum,” Proc Soc Exp Biol Med
103:663–666, 1960.
3. Rubin, M, and Princiotto, JV, “Synthetic amino acid chelating agents and iron metabo-
lism,” Ann NY Acad Sci 88:450–459, 1960.
4. Pineda, O, and Ashmead, HD, “Effectiveness of treatment of iron deficiency anemia in
infants and young children with ferrous bis-glycinate chelate,” Nutrition 17:381–384,
2001.
5. Kehoe, CA, Turley, E, Bonham, MP, O’Connor, JM, McKeown, A, Faughnan, MS,
Coulter, JS, Gilmore, WS, Howard, AN, and Strain, JJ, “Response of putative indices
of copper status to copper supplementation in human subjects,” Br J Nutr 84:151–156,
2000.
6. Hill, J, Micklewright, A, Lewis, S, and Britton, J, “Investigation of the effect of short-
term change in dietary magnesium intake in asthma,” Eur Respir J 10:2225–2229, 1997.
7. Gontijo-Amaral, C, Ribeiro, M, Gontijo, L, Condino-Neto, A, and Ribeiro, JD, “Oral
magnesium supplementation in asthmatic children: A double-blind randomized
placebo-controlled trial,” Eur J Clin Nutr 1–7, June 2006.
8. Weaver, K, and Speigel, KM, “Platelet functional response to oral doses of magnesium
and fish oil in human volunteers,” J Am Coll Nutr 7:5, 1988.
9. Baum, SJ, The Total Guide to a Healthy Heart (New York: Kensington Books) 1999.
10. Prasad, AS, “Zinc deficiency in human subjects,” in Prasad, AS, Caudar, AO, Brewer,
GJ, and Aggett, PJ, eds., Zinc Deficiency in Human Subjects (New York: Liss) 1–33,
1983.
11. Bolstein-Fujii, A, DiSilvestro, RA, Frid, D, Katz, C, and Malarkey, W, “Short-term zinc
supplementation in women with non-insulin dependent diabetes mellitus: Effects on
plasma 5′-nucleotidase activities, insulin-like growth factor I concentrations, and lipo-
protein oxidation rates in vitro,” Am J Clin Nutr 66:639–642, 1997.
12. Diplock, AT, “Antioxidants and free radical scavengers,” in Rice-Evans, C, and Burdon,
R, eds., Free Radical Damage and Its Control (Amsterdam: Elsevier) 114–115, 1994.
13. Davis, CD, and Greger JL, “Longitudinal changes of manganese-dependent superoxide
dismutase and other indexes of manganese and iron status in women,” Am J Clin Nutr
55:747–752, 1992.
14. Kropp, JR, “Reproductive performance of first calf heifers supplemented with amino
acid chelated minerals,” in Oltijen, J, Horn, G, Geisert, R, and Yates, L, eds., Animal
Science Research Report (Stillwater, OK: Oklahoma State University) 35–43, 1990.
15. Ashmead, HD, and Graff, DJ, “Decreasing ethanol consumption in ethanol-dependent
rats through supplementation of zinc and copper amino acid chelates: A preliminary
study,” J Appl Res 6:19–27, 2006.
16. National Research Council, Nutrient Requirements of Laboratory Animals (Washington,
DC: National Academy Press) 2002.
17. Ashmead, HD, Ashmead, SD, and Samford, RA, “Effects of metal amino acid chelates
on milk production, reproduction and body condition in Holstein first calf heifers,” Int J
Appl Res Vet Med 2:252–260, 2004.
18. Beisel, WR, “Magnitude of the host nutritional responses to infection,” Am J Clin Nutr
30: 1236–1247, 1977.
19. Abrams, S, “Where do all the minerals go? Finding the causes of long term mineral
deficiency in children with serious illness,” Nutrition 17:259–269, 2001.
20. Wayne Feeds, The Different Chelated Iron, Wayne Feeds Monograph, Allied Mills,
Chicago, IL, 1976.
21. Ashmead, HD, and Ashmead, SD, “The effects of dietary molybdenum, sulfur and iron
on absorption of three organic copper sources,” J Appl Res Vet Med 2:1–9, 2004.
14 Toxicity of Amino
Acid Chelates
Although the electronic properties of a mineral ion are important in effecting certain
metabolic changes, metals are not generally in an ionic state when stored in cells.
Thus, when the mineral ion is released from the plasma transport molecule and taken
up by the cells composing a tissue or organ, it is complexed to various intracellular
ligands that either store the metal or move it to other loci within the cell for usage or,
if the metal ion is in excess, they facilitate its elimination.1
When a metal ion is absorbed into the mucosal cell and transferred to the plasma,
less of that form of mineral is taken up by the tissues from the plasma compared to
the uptake of an amino acid chelate that contains the same metal.2,3 Further, once a
metal ion is absorbed into the tissues or organs, a higher percentage appears to be
returned to the plasma and eliminated from the body as an endogenous metal com-
pared to an amino acid chelate or chelate/complex containing the same metal.2–4 It
was postulated that one reason for the higher tissue and organ removal rates of ionic-
sourced metals may be due to inadequate quantities of specific amino acids that may
be required to work with that cation and synergistically balance with that metal for
metabolic purposes.
The different removal rates of two sources of the same metal (ionic and amino
acid chelate) from tissues or organs can be seen in a study involving milk production.
As discussed previously, dairy cow milk production is influenced by several factors,
including the types and amounts of mineral nutrients consumed by the cow.5 When
the lactating animal is deprived of adequate amounts of the essential minerals for
milk production, the cow’s initial response will be to cannibalize the required miner-
als from her own body. When cannibalization commences, milk production declines
rapidly as the body reserves are depleted. Thus, a continuing supply of dietary min-
erals is absolutely required to support the high milk production that many breeds of
cow are genetically capable of producing.
Depending on the supplemental mineral source, milk and milk component pro-
duction in dairy cows can be augmented.3 Several studies have been published that
compared milk production in cattle consuming the same mineral sources as either
inorganic metal salts or as other organic minerals, including amino acid chelates.6–9
Certain of these studies reported that when the mineral is provided as an amino acid
chelate, total milk production, the percentage of protein, and the percentage of fat
in the milk were all significantly increased compared to milk production from herd-
mates that ingested the same minerals as inorganic metal salts.3,9 These data also
provided evidence that amino acid chelates not only have an immediate and positive
impact on milk production, but also, due to their greater tissue retention, appear
to have a longer-term positive residual effect on milk production.3 In Chapter 13,
223
studies with both pigs and dairy cows demonstrated the long-term effects of supple-
menting their diets with amino acid chelates. In both cases, animal production and
performance continued to be enhanced long after amino acid chelate supplementa-
tion ceased.
As discussed previously, it is the chelating of those minerals with amino acids that
seems to be responsible for augmenting the uptake of both the metal and the ligand.
This is presumably due to the body initially responding to an amino acid chelate as
if it were a proteinaceous molecule rather than a metal.
It has been hypothesized further that the resultant increased tissue retention that
is observed is due to achieving a mineral-amino acid balance that is primarily a
consequence of the amino acid ligands being taken up by the tissues while the met-
als bonded to them are simply smuggled into the cells as part of the amino acid
ligands. While there is a tissue turnover of metals sourced from amino acid che-
lates, this turnover is slower, thus leading to increased tissue concentration of those
metals. Greater mineral retention from the amino acid chelate, however, suggests
a potential for toxicity, particularly if the chelates are consumed daily. The studies
described in the following discussion indicate, however, that this concern is gener-
ally unfounded.10–12
The first of these reported studies involved two groups of pigs, each initially
containing six sows.11 As each sow gave birth, the female offspring from her lit-
ter were retained, raised, and bred. This program continued until four additional
generations of adult female pigs were ultimately included in each of the two groups.
Thus, the two groups of pigs expanded to encompass the original six mothers in each
group plus their daughters, granddaughters, great-granddaughters, and great-great-
granddaughters. The animals in each generation and in each group were all from
the same genetic breed. Each successive generation received similar health care,
feed, and so on as the original six sows had received, so all generations were treated
similarly. At breeding, each animal was bred to the same boar by artificial insemina-
tion. At the time of sacrifice, each generation of pigs was of comparable age, health,
and the like. The only variable between the two groups of pigs was the sources of the
minerals provided in their supplement programs.
The female pigs assigned to the treatment group received supplemental magne-
sium, iron, manganese, copper, zinc, and cobalt, all as amino acid chelates. The
female pigs in the control group received the same minerals in the same quantities
but in the form of inorganic metal salts. While younger at the time of sacrifice, the
pigs in the fifth filial generation in both groups received these minerals prenatally as
well as postnatally all of their lives, as had their mothers, grandmothers, and great-
grandmothers. The original six great-great-grandmothers had been fed the same
minerals from either source all of their postnatal lives, but they had not been enrolled
in the study until the time of their births, so they did not receive them prenatally.
These minerals were provided as a supplement in addition to the minerals normally
fed as part of a daily ration that met the NRC recommended mineral levels.13
All of the animals in both groups were taken to the same abattoir and killed on
the same day. Prior to sacrifice, each pig was subjected to a premortem examination
by a licensed veterinarian, and all were found to be disease free and in normal condi-
tion. Following sacrifice and while still on site at the abattoir, a pathologist examined
Toxicity of Amino Acid Chelates 225
the viscera and carcass of each animal for any gross teratogenic effects. None was
noted in any of the animals in either group, so excised tissues were taken from the
liver, kidney, heart, duodenum, jejunum, ileum, large bowel, spleen, muscle, mes-
enteric lymph node, ovary, brain, and bone marrow of every pig in both groups; the
samples were all submitted for histopathologic analysis. In addition, a blood sample
was obtained from each animal at the time of slaughter and submitted for hematol-
ogy and blood chemistry studies. The histopathologist and hematologist were both
kept blind regarding which tissues or blood came from which group.
Following microscopic examination of the submitted samples, the histopatholo-
gist reported that there were no differences in the health or condition between the
tissues in any of the pigs. Further, he could not discover any histopathologic tissue
alterations that could be attributed to any of the supplemented minerals from either
source. The hematologist reported that there were no differences in the hematologi-
cal or blood chemistry measurements between the animals in either group. It was
concluded that long-term feeding of supplemented amino acid chelates did not result
in any cumulative teratogenic effects or cause chronic morbidity at either a histologi-
cal level or at a level of practical observation. The condition of the tissues and organs
of the animals appeared to be the same whether the animals consumed amino acid
chelates or inorganic metal salts.
A similarly designed follow-up study was conducted on another group of sows to
assess whether there was an accumulation of minerals in the tissues and organs of
the pigs receiving amino acid chelates and, if so, if that accumulation would contrib-
ute an added risk of toxicity.12 Again, two groups of six sows each were selected to
initiate the study. One group received magnesium, iron, manganese, zinc, and cop-
per as amino acid chelates, while the other group received the same minerals in the
same amounts as inorganic metal salts. All of the minerals were supplied as supple-
ments to both groups and were fed in addition to the minerals normally provided in
the feed of both groups. The minerals that were contained in the commercial feed
rations met NRC requirements.13
The first generation commenced receiving a daily mineral supplement of these
amino acid chelates or inorganic metal salts as soon after birth as they could con-
sume solid food. The study continued until there were four subsequent generations
of female pigs in each group, all of which had received the supplemental minerals,
as either amino acid chelates or as inorganic metal salts on a daily basis, throughout
their pre- and postnatal lives. (During lactation, the iron came from their mothers’
milk prior to weaning.) Thus, the pigs in the fifth filial generation, while the young-
est in age, came from mothers, grandmothers, and great-grandmothers that had been
administered the amino acid chelates or inorganic mineral salts as daily supple-
ments all of their pre- and postnatal lives and from great-great-grandmothers that
had received the same chelates or inorganic metal salts all of their postnatal lives.
The total amounts of minerals consumed as either amino acid chelates or inorganic
metal salts were the greatest in the earlier generations because these animals had
lived the longest. Unfortunately, the first-generation great-great-grandmothers in the
control group, which had received the same feed and inorganic minerals supplement
as all of the later generations, were butchered early. Consequently, they were not
available for histopathology and hematology analysis at the time that the sacrifice of
the remainder of the pigs in both groups occurred. Nevertheless, since none of these
original sows was rejected by government inspectors working at the abattoir, it was
assumed that these initial sows in the control group did not exhibit any untoward
histopathological or hematological effects resulting from the supplementation of the
inorganic mineral salts.
With the exception described, the remaining test and control groups of female
pigs were all delivered to the abattoir on the same day. Each was examined by a
veterinarian before slaughter. All were determined to be healthy and normal. Blood
samples were taken immediately following sacrifice and delivered to a clinical labo-
ratory for hematological analysis. The viscera and carcasses were each examined on
site by a pathologist for any abnormalities, followed by excising samples of the duo-
denum, jejunum, ileum, large bowel, mesenteric lymph node, bone marrow, muscle,
brain, heart, liver, kidneys, spleen, and ovaries from each pig. All of these tissue
samples were submitted for histopathologic examination followed by a mineral anal-
ysis of each sample by atomic absorption spectrophotometry. The histopathologist,
the hematologist, and the laboratory conducting the mineral analysis on the tissues
were all kept blind regarding which samples came from which group or parity.
The tissues from the animals receiving the amino acid chelates contained signifi-
cantly (p < 0.05) higher concentrations of magnesium, iron, manganese, zinc, and
copper when compared to the tissues taken from pigs fed the inorganic minerals. In
spite of the greater amounts of metals in the tissues taken from the chelate group,
there was no significant difference in the health or condition of any of the tissues
from either group based on the gross pathological examinations. When the blood of
each pig was examined, the hematological examinations did not reveal any abnor-
malities related to the administration of the amino acid chelates. The histopatho-
logical comparison of the tissues of the test animals to the tissues from the control
sows of similar ages and parities (except the first generation) revealed no observ-
able microscopic untoward effects of a cumulative or chronically morbid nature that
could be attributed to the administration of the amino acid chelates. Finally, the tis-
sues from the first-generation test sows that had received the amino acid chelate
the longest time did not present any gross or microscopic untoward effects in those
tissues resulting from being fed the amino acid chelates even though they were not
compared to tissue from a first-generation control sow for the reasons stated.
The administrator of this study concluded the following:
The results of this long term study demonstrate that Albion® magnesium, iron, manga-
nese, zinc and copper amino acid chelates are devoid of chronic toxicity and are safe
for continued feeding as nutritive dietary supplements when administered at recog-
nized and recommended levels for nutritive minerals.14
These two swine studies suggest that a paradigm shift in thinking is required as
it relates to amino acid chelate toxicity. To a substantial degree, uptake of the amino
acid chelate or chelate/complex into tissues and organs may be equally, or even more,
dependent on the uptake of the amino acid ligand than on the metal. A certain por-
tion of the two components is absorbed together as an amino acid chelate or chelate/
complex into the body cells via amino acid absorption pathways. Following uptake,
they appear to be retained within these cells and participate together at a higher level
in their assigned cellular metabolic processes. This is probably due to the balance
theory previously described.
If there is any subsequent intracellular hydrolysis of the amino acid chelate or
chelate/complex within the tissues of the body, and there probably is, some of the
metal that is released from these molecules may ultimately be returned to the plasma
and excreted as endogenous minerals similarly to cations sourced from inorganic
salts that have been taken up by the tissue. Further, if the uptake of the amino acid
ligand were excessive in terms of cellular requirements, it also could be returned to
the plasma following intracellular hydrolysis of the amino acid chelate or chelate/
complex. Or, another possibility may be that the amino acid chelate or chelate/com-
plex, acting as an amino acid (for tissue uptake purposes), may be in excess in cyto-
plasm of the cell and the entire molecule returned to the plasma. In the latter case,
the metal that was chelated to the amino acid ligands would continue to go along
for the ride both in and out of the cell. In any of these possibilities, it appears that
the presence of the amino acid ligand seems to be the dominant determining factor
that dictates how the body will respond to the higher absorption and tissue uptake of
amino acid chelates.
These potential scenarios seem to be more clearly described in the following toxi-
cology study, which was initially performed to assess the safety of a copper amino
acid chelate. In this case, it was previously determined that the LD50 (lethal dose
for 50% of the test animals) resulting from an intravenous injection of copper was
5.86 mg Cu/kg of body weight from copper sulfate and 8.78 mg Cu/kg of body
weight from copper amino acid chelate.15 Based on these LD50 levels, a study was
designed in which a group of 10 Sprague-Dawley albino rats was divided into two
groups of five animals each. All were anesthetized with ether, and the LD50 dose for
copper sulfate (5.86 Mg Cu/Kg) was injected intravenously into the animals in one
group and the LD50 dose for copper amino acid chelate (8.78 mg Cu/Kg) injected
intravenously into the rats composing the other group. Both copper sources were dis-
solved in sterile, deionized water prior to being injected. Thirty minutes following
the injections, while still anesthetized, all of the animals were sacrificed and gross
necropsies performed. The results for the organs most affected by the copper are
shown in Table 14.1.15
In addition to performing gross necropsies on the livers, spleens, and intestines,
these same organs, along with several other organs, were subjected to a copper anal-
ysis by atomic absorption spectroscopy. As seen in Table 14.2, many of the organs
from the rats injected with the copper amino acid chelate generally accumulated
more copper from the amino acid chelate source compared to the rats that were
injected with the sulfate source; yet, the organs from the chelate group all retained
a normal appearance.15 This strongly suggested that the molecular form of the cop-
per taken up from the blood by the organs analyzed in Table 14.1 was in a different
form following injection of the copper amino acid chelate compared to the injection
of copper sulfate. The ionized copper from the sulfate source presumably had to be
picked up by transport molecules sourced from the blood prior to delivering that
source of copper into the organs. Even though less copper, based on a milligram per
kilogram of body weight, was injected into the rats composing the sulfate group, this
TABLE 14.1
Condition of Selected Organs Following Gross Necropsies
of Rats Receiving an LD50 Intravenous Injection of Copper
Amino Acid Chelate or Copper Sulfate
Organ Type Cu Amino Acid Chelate Cu Sulfate
Liver Normal Mottled and white
Spleen Normal Mottled and dark with red areas
Intestine Normal Inflamed
Source: Data from Larson, AE, “Tissue accumulation and relative safety of
copper amino acid chelate vs. copper sulfate,” unpublished research
report, University of Utah, Salt Lake City, 1982.
TABLE 14.2
Copper Levels in Organs from Rats Receiving an
Intravenous Injection of Either Copper Amino Acid
Chelate or Copper Sulfate
Cu Cu Amino Acid
Organ Type Sulfate* Chelate*
Skeletal muscle (right quadricep) 15.539 19.635
Heart 13.789 12.769
Kidney 12.026 9.923
Liver 10.642 6.408
Brain (left cerebrum) 12.500 16.893
Lung 7.649 9.905
Intestine 12.364 6.359
Spleen 23.225 16.740
Blood 11.891 13.443
Source: Data from Larson, AE, “Tissue accumulation and relative safety of
copper amino acid chelate vs. copper sulfate,” unpublished research
report, University of Utah, Salt Lake City, 1982.
* mg Cu per gram of wet body tissue.
form of copper seemed to overwhelm the organs and resulted in the toxic response15
summarized in Table 14.2.
It was obvious that the amino acid-chelated source of copper was handled dif-
ferently by the bodies of the animals receiving that source of copper. The chelating
of the copper with amino acids seemed to have resulted in the presentation of a
source of copper that was more compatible with the tissues and organs, even when
the copper concentration within the organ was greater. Because the copper in the
dissolved amino acid-chelated copper was not in ionic form, it did not require trans-
port molecules to be carried throughout the plasma and into the organs. Since the
animals’ responses to the two sources of copper were obviously not the same, it was
concluded that, for the most part, the copper amino acid chelate was probably taken
up by the organs as the same molecule that was injected into the rats rather than
first being hydrolyzed into the amino acid ligands and ionic copper. Had hydroly-
zation occurred, the released copper would have been bonded to plasma transport
ligands that would have been identical to the ligands that picked up the ionized cop-
per from the sulfate source and resulted in a similar toxic response. Since there was
no observed toxic response, such as inflammation, in any of the organs following
injection of the copper amino acid chelate, that source of copper had to be in a dif-
ferent molecular form at the time that the copper from that source was taken up.
This experiment was not designed to determine copper bioavailability from either
source of copper. Each copper source was injected intravenously into the rats rather
than being administered orally. The purpose of the study was to examine the physi-
ological response to the two copper sources. The circulatory system and organs in
the group of rats receiving the copper sulfate appeared to be ill prepared to ward off
the onslaught of the large bolus of copper when presented in the ionic form that was
injected at the intravenous LD50 level.
When the three organs that were the most severely compromised by the intrave-
nous injection of inorganic copper were analyzed in the sulfate group and compared
to the same organs from the amino acid group, the livers from the animals receiv-
ing the copper as copper sulfate actually contained 66% more copper than did the
hepatic tissue from animals given the amino acid chelate. Since the livers containing
the most copper were from rats administered the lower amount of copper from the
sulfate source, it appeared that the livers from these animals were aggressively try-
ing to detoxify the body. As previously suggested, the increased compatibility of the
copper amino acid chelate with the rats’ bodies, which was reflected by reduced
amounts of copper in the livers of those animals receiving this copper source, may
be due to balancing the copper with the amino acid ligands. The lower amounts of
copper from the amino acid chelate in their livers may have been in response to the
presence of the amino acid ligands rather than the copper itself. The balancing may
have resulted in less reactivity14 and reduced the need of the livers in that group to
aggressively detoxify the bodies.
The spleens from the rats receiving the sulfate source also appeared to be involved
in this same detoxifying effort since they contained 39% more copper than did the
spleens in the animals receiving the chelated source. As a large lymphoid organ, one
of its major functions is removal of cells and cellular detritus. It acts as a type of fil-
ter for the blood.16 While no measurement was made in this study, it is possible that
when the bolus of copper from copper sulfate was taken up, in part, by the red blood
cells, it immediately killed many of those cells. The spleen further extracts the iron
from dead red blood cells and directs it to the bone marrow for incorporation into
new red blood cells. In a biological sense, iron and copper are very similar metals,
and the spleen may have also collected the copper from dead red blood cells concur-
rently with its collection of the iron. Both copper and iron share many of the same
pathways that mediate their dietary assimilations, storage, and distributions.17 Or, the
spleens from the group receiving copper sulfate may have been collecting red blood
cells that had been killed by the copper bolus, and the analysis of those spleens may
have actually included the copper that was originally taken up by the red blood cells.
The reaction of the spleens from the rats receiving the bolus of copper amino
acid chelate was very different when it was compared to the reaction from the group
receiving copper sulfate. Even though more copper per kilogram of body weight
was injected into the animals in the chelate group (to achieve the LD50 level), their
spleens did not attempt to collect or remove as much of that form of copper from the
circulatory system or red blood cells. This suggested that more of that source of cop-
per was absorbed into the other tissues of the rats’ bodies or was at least deemed not
to be toxic to the animals. It appeared not to be as lethal to the red blood cells as was
copper sulfate. Based on the data in Table 14.1, it is obvious the amino acid-chelated
form of copper was more compatible with the organs and tissues of the rats receiving
it than was the ionized copper sulfate.
It was also interesting to note that the rats receiving the copper sulfate injec-
tion had 94% more copper in their intestinal tissue than did the rats injected with
copper amino acid chelate. This suggested that active endogenous excretion of the
injected copper from the sulfate sourced back into the gastrointestinal tract was
being attemped.18 The inflamed intestine, coupled with almost twice as much copper
concentration in that tissue from the rats receiving the bolus of copper sulfate, sug-
gested that their bodies were much more aggressive in trying to endogenously elimi-
nate that source of copper compared to bodies of the animals receiving the amino
acid chelate source. This also suggested less compatibility and greater toxicity from
the sulfate source compared to copper amino acid chelate.
Additional LD50 studies19 and acute and subacute toxicity studies20,21 and other
long-term toxicity studies22 all affirmed the safety and compatibility of the amino acid
chelates with the tissues of the body. In his discussion concerning iron amino acid che-
late, Jeppsen summarized the general conclusion that can be related to all of the
amino acid-chelated forms of the minerals when he wrote that it
may be the best form for supplementation and fortification of iron into human diets. It
presents iron in ferrous state, resists cleavage once ingested, is absorbed intact in its
protective state, repletes iron levels at relatively small doses, has been proven remark-
ably nontoxic, even at relatively high daily doses, demonstrates the characteristics of
being physiologically regulated, and increases iron absorption over that obtainable
from ferrous sulfate, the current standard for iron intervention.21
REFERENCES
2. Ashmead, HD, “Comparative intestinal absorption and subsequent metabolism of
metal amino acid chelates and inorganic salts,” in Subramanian, KS, Iyengar, GV, and
3. Jensen, NL, “Biological assimilation of metals,” U.S. Patent 4,167,546, Washington
4. Ashmead, HD, and Samford, RA, “Effects of metal amino acid chelates or inor-
ganic minerals on three successive lactations in dairy cows,” Int J Appl Res Vet Med
2:181–188, 2004.
5. McCullogh, ME, Optimum Feeding of Dairy Animals (Athens, GA: University Georgia
Press) 135, 1976.
6. Smith, MB, Amos, HE, and Froetschel, MA; “Influence of ruminally undegraded pro-
tein and zinc methionine on milk production, hoof growth and composition, and selected
metabolites of high producing dairy cows,” Prof Anim Sci 15:268–277, 1999.
7. VanHorn, HH, Shearer, JK, Wilcox, CJ, and de Groot, W; “Utilization of heterogene-
ity of regression to delineate effects of Zn-, Mn-, and Cu-proteinates on milk somatic
cell counts, milk yields, and cow mobility in research conducted on farms,” J Dairy Sci
77:156(Suppl.), 1994.
8. Bonomi, A, Quarantelli, A, Sabbioni, A, and Superchi, P, “I complessi oligodinamic: che-
late nell’alimentazione dell bovine da latte,” II Nuevo Progresso Vet 41:673–682, 1986.
9. Gupte, SS, Dickie, MG, Patil, CN, Kank, VD, and Patil, MB, “Results of MAAC® in
dairy cattle in India,” Proceedings of Albion International Animal Nutrition Conference,
Salt Lake City, UT, February 1997.
10. Jeppsen, RB, “Assessment of long term feeding of chelated amino acid minerals in
sows,” unpublished research report, Albion Laboratories, 1997.
11. Jeppsen, RB, “Assessment of long term feeding of Albion® magnesium, iron, manga-
nese, zinc, copper and cobalt amino acid chelates,” unpublished research report, Albion
Laboratories, Clearfield, UT, 1987.
12. Jeppsen, RB, “Examination of long term feeding of Albion® magnesium, iron, man-
ganese, zinc and copper amino acid chelates in sows achieving six or greater parity,”
unpublished research report, Albion Laboratories, Clearfield, UT, 1990.
13. National Research Council, Nutrient Requirements of Swine (Washington, DC: National
Academy Press) 1988.
proof of chelation,” Proceedings of the Albion International Conference on Human
Nutrition, Salt Lake City, UT, January 1995.
15. Larson, AE, “Tissue accumulation and relative safety of copper amino acid chelate vs.
copper sulfate,” unpublished research report, University of Utah, Salt Lake City, 1982.
16. Conley, CL, “The Blood,” in Mountcastle, VB, ed., Medical Physiology (St. Louis, MO:
Mosby) 1034, 1974.
17. Collins, JF, Prohaska, JR, and Knutson, MD, “Metabolic crossroads of iron and copper,”
Nutri Rev 68:133–147, 2010.
18. Davis, GK, and Mertz, W, “Copper,” in Mertz, W, ed., Trace Elements in Human and
Animal Nutrition (San Diego, CA: Academic Press) 325, 1987.
19. Larson, AE, “L.D.50 studies with chelated minerals,” in Ashmead, D, ed., Chelated
Minerals Nutrition in Plants, Animals and Man (Springfield, IL: Thomas) 163, 1982.
20. Jeppsen, RB, and Berzelleca, JF, “Safety evaluation of ferrous bisglycinate chelate,”
Food Chem Toxicol 37:723–731, 1999.
21. Jeppsen, RB, “Toxicology and safety of Ferrochel and other iron amino acid chelates,”
Arch Latin Am Nutr 51:26–34 (Suppl.), 2001.
22. Jeppsen, RB, “An assessment of long term feeding of amino acid chelates,” in Ashmead,
HD, ed., The Roles of Amino Acid Chelates in Animal Nutrition (Park Ridge, NJ: Noyes)
106–113, 1993.
15 The Absorption
and Metabolism of
Amino Acid Chelates
When Alfred Warner published his 1893 landmark paper describing chelation, there
was probably no thought regarding the implications or impact his observations would
have on mineral nutrition. He wrote, “If we think of the metal ion as the center of the
whole system, then we can most simply place the molecules bound to it at the corners
of an octohedron.”1 He viewed the chelate molecule as the metal ion being bonded to
two ligands with the metal ion being the closing member of each chelate ring.1
While subsequent research has established that a single metal ion can chelate with
one, two, three, four, five, and possibly more ligands,2 the Association of American
Feed Control Officials (AAFCO) has defined one type of chelate, an amino acid
chelate, as
the product resulting from the reaction of a metal ion from a soluble metal salt with
amino acids with a mole ratio of one mole of metal to one to three (preferably two)
moles of amino acids to form coordinate covalent bonds. … The resulting molecular
weight of the chelate must not exceed 800.3
The AAFCO has added that
at least one ligand (electron pair donor) forms two or more bonds with the central metal
ion through different atoms of the ligand. A distinctive feature of a metal chelate is the
presence of a heterocyclic ring(s) in which the metal is a member of the ring.3
As early as 1957, the employment of amino acid ligands as a way to increase min-
eral bioavailability was reported,4 but extensive research on this type of chelate was
not conducted until the 1960s. Both Cardon and Scott suggested that amino acids
were excellent potential ligands and should be used to chelate minerals for nutri-
tional use. Neither investigator, however, conducted significant research to prove
the hypothesis, preferring instead to focus on other synthetic chelating ligands.5,6 In
1963, under Harvey Ashmead’s direction, a study involving gestating rats was initi-
ated to evaluate the biological effects of amino acid chelates on those animals.7 The
results were favorable and led to more extensive research with several additional spe-
cies of animals.8,9 Ashmead later reported that by 1970, to study the effects of feed-
ing this form of minerals, he and his team had administered amino acid chelates to
over 200,000 animals, including poultry, cattle, swine, sheep, horses, and fish. Their
general observations included better absorption, faster growth, improved production,
reduced morbidity and mortality, and improved reproduction.10
233
When their findings were published, controversies arose for several reasons. First,
as with any new scientific advancement, no peer-reviewed reference material existed
substantiating Ashmead et al.’s findings. Without that secondary scientific support
data, it was difficult for many to accept the research conclusions immediately.
This led to a second reason for controversy. The published reports contradicted
some of the more widely accepted theories relating to mineral nutrition, particu-
larly when it focused on bioavailability. These contradictions led to criticism and the
unwarranted conclusion by some that amino acid chelates were simply an expensive
fraud.11,12 While disappointing to Ashmead et al., these reactions from the scientific
community should have been expected. Most advances in any scientific field are usu-
ally met with skepticism, particularly if they advocate new theories or observations
that contradict old ones.
The third major problem preventing wholesale acceptance of these early reports
was that, at the time, very few people comprehended the chemical or molecular
requirements necessary to create a true amino acid chelate. For several years, there
was no peer-reviewed analytical procedure that could prove amino acid chelates even
existed. Later, when analytical methods proving amino acid chelation were validated
through an American Organization of Analytical Chemists (AOAC) program,13
these procedures were rarely employed as a prerequisite to a research study involv-
ing AAFCO-defined amino acid chelates. Many investigators continued to work as
they had in the past and assumed they were studying amino acid chelates without
first analyzing the molecules to confirm their assumption. Consequently, the results
ascribed to an amino acid chelate were frequently derived from metallic molecules
that were not proven amino acid chelates. To confuse the issue even more, many
investigators mixed data from studies involving AAFCO-defined amino acid com-
plexes and metal proteinates with those of amino acid chelates.14–16 All of this pro-
duced conflicting absorption and bioavailability data that were confusing at best.17
In their monograph, Leach and Patton wrote the following:
Acceptance of chelates has long been hindered by the lack of an assay. A buyer sim-
ply had no way of testing his purchase, having to rely exclusively on the reputation of
his supplier and subjective feedback from the field. The final determination is almost
solely influenced by the cost per kilogram. Their frustration has been that they could
not determine if the high price one paid for a copper chelate was in actuality inexpen-
sive copper sulfate.18
While both the scientific community and industry alike recognized the need for an
analytical method to substantiate that the mineral being evaluated was truly che-
lated, no such method existed until amino acid chelate crystals could be grown and
analyzed by x-ray diffraction crystallography. This analytical procedure proved that
these crystals were actually amino acid chelates, and ultimately these crystals were
used as reference samples to validate other analytical tests.19,20
Prior to having reference amino acid chelates, several investigators attempted to
develop analytical procedures that would measure a characteristic of a known che-
late in the mistaken belief that if such a characteristic existed in the product being
analyzed, it would prove that product was chelated, even though certain nonchelated
The Absorption and Metabolism of Amino Acid Chelates 235
mineral sources exhibited the same characteristics. One simple technique was to
mix a suspected chelate with water, filter the slurry, and then analyze the filtrate
solution for metal content. The belief was that the unbound metal would ionize in
solution and be recovered in the filtrate solution, whereas the insoluble chelated min-
eral would be trapped in the insoluble precipitate.21,22 It was ultimately concluded,
however, that there was no way of determining if the bound mineral in the precipitate
was chelated or attached to some other type of molecule that also resulted in insol-
uble mineral products.21 Further, depending on the ligand used for chelation as well
as the temperature of the water mixed with the suspected chelate and the resulting
pH of the solution, more or less of the amino acid chelate could potentially solubilize
and thus would become part of the filtrate solution, which would then be erroneously
included as part of the ionized metal content.
Other investigators tried to develop a more sophisticated modification of this tech-
nique. They subjected both the filtrate solution and the precipitate not only to a metal
analysis but also to a nitrogen analysis using the Kjeldahl method.23 The assumption
was that analysis for nitrogen would indicate how much of the amino acid ligand
was present in each of the components. While this method could demonstrate the
percentages of metal and nitrogen in both the filtrate solution and the precipitate, it
did not show how these two essential components were related to each other. The
metal may have indeed been chelated to nitrogenous matter (the amino acid), as the
investigators tried to advocate with this analytical technique, but the metal could just
as easily have been complexed to a protein molecule or even exist as an inorganic
metal salt that had simply been admixed with the protein. This analytical method did
not ascertain the type of bonds that existed between the ligand and the metal ion that
was the defining feature of an amino acid chelate.
Still other researchers attempted to employ an ultrafiltration technique to deter-
mine chelation.24 Ultrafiltration required solubility. In an attempt to solubilize the
product under investigation, these investigators added ethylenediaminetetraacetic
acid (EDTA) to the solution containing the insoluble mineral product they suspected
of being an amino acid chelate. While they were somewhat successful in putting the
mineral product into solution, the EDTA, with its higher potential formation con-
stant, had a tendency to pull the metal away from the proteinaceous ligands and form
EDTA chelates instead. Since an EDTA chelate is a small molecular weight chelate,
it was able to pass through the filter and be recovered in the filtered solution. The
remaining precipitate, which would not go into solution and thus not pass through
the ultrafilter, was supposed to be an amino acid chelate or metal proteinate, but the
precipitate could not be proven to be a chelate since the EDTA had removed much of
the original mineral from the product.
Paper chromatographic assays also failed to prove chelation because, as indi-
cated, not all of the analyzed chelates were soluble.25,26 When EDTA was utilized to
solubilize the product just as in the case of the ultrafiltration procedure, it removed
the metals from amino acid chelates or complexes and thus destroyed the products
being targeted for analysis.16
Polarography was another laboratory procedure that tried to prove chelation.26,27
While polarographic analysis could demonstrate whether a reaction had occurred
between the metal and the ligand in the samples being analyzed, polarography could
not definitely prove what type of molecule that reaction had produced, if it occurred
at all. The product could have been an amino acid chelate, or it could have just as
easily been an amino acid complex. Or, it may have been the metal ion reacting with
an inorganic anion.27 In forming any of these molecules, a chemical reaction would
likely have taken place, and the polarographic analysis would have reported the
occurrence of that reaction. Unfortunately, it did not identify the type of reaction
observed as a result of the polarographic analysis.
Like polarography, nuclear magnetic resonance (NMR) spectroscopy could also
demonstrate that a reaction had taken place between the metal and the ligand in the
samples assayed.28 It was the consensus of the investigators who developed this ana-
lytical technique that NMR could demonstrate chelation, but the analysis required
that the sample submitted for analysis be soluble. Unfortunately, not all amino acid
chelates are equally soluble. Thus, NMR at that time was not a viable analytical
procedure to prove chelation.
Another analytical procedure, electron paramagnetic resonance (EPR) spectros-
copy, could demonstrate that a sample was chelated by comparing the EPR spectra
of the mineral product to the spectra of a known amino acid chelate that occurred
in nature.29 Use of this procedure illustrated one essential requirement for a peer-
reviewed analytical procedure that proved chelation. The spectra of known amino
acid chelates had to be used as reference spectra. At the time that EPR was studied,
reference spectra for all amino acid chelate metals did not exist. Further, EPR was
only applicable for analysis of transitional metals, so as an analytical tool to prove
chelation, EPR was not particularly viable.
Following several years of investigation, Hartle and Ashmead published a sum-
mary of their team’s research. Titled, “Bonds Important for Amino Acid Chelates,”
these authors wrote:
Physical characteristics alone (solubility, percent nitrogen and approximate molecu-

lar weights) do not confirm the existence of chelation. … The only way to verify that
an amino acid chelate exists is to look at the bonds [between the metal and the amino
acid ligand].30
These investigators, along with others on their team, had been able to grow
pure amino acid chelate crystals. The crystals were subsequently analyzed by x-ray
crystallography, which determined the angles of the bonds between the amino acid
ligands and the metal, the lengths of those bonds, and finally, the steric orientations
of the various atoms within the chelate molecule.31 That analysis proved that the
crystalline metallic molecules that they subjected to x-ray crystallography analysis
were truly in the form of amino acid chelates. Further, these same crystals were later
subjected to electrospray mass spectroscopy, which expanded the elucidation of their
structures besides confirming chelation.20
Once it was proven that these crystals were true amino acid chelates, additional
samples from the same amino acid chelate crystals were submitted for analysis by
Fourier transform infrared (FT-IR) spectroscopy.30 In this way, FT-IR reference
spectra developed from known amino acid chelates could be compared against other
suspected amino acid chelates to prove whether the unknown sample was a chelate
or some other type of metallic molecule. This FT-IR procedure was submitted to the
AOAC using their single-laboratory validation methology.13
Knowing whether one was employing a true amino acid chelate in a research
project is extremely important, particularly so in studies describing how an amino
acid chelate is absorbed or metabolized because “amino acid chelate bonding char-
acteristics are what set it apart as a highly bioavailable mineral source.”30 If one does
not absolutely know if he or she is working with a true amino acid chelate, how then
can one ascribe the absorption characteristics that are observed as being due to the
chelation of that metal? The same is true as it relates to the unique metabolism of an
amino acid chelate. To claim that the metabolism of a particular metal source is due
to it being chelated with amino acids, one must ensure that he or she is working with
a true amino acid chelate and not a complex or some other metal form.
One of the characteristics of a true amino acid chelate is that it must have a coor-
dinate covalent bond between the metal ion and the nitrogen on the amino group.
This bond will form at certain pH values and dissociate when subjected to other pH
values.32 Further, this forming and breaking of the bond between the amino group of
the ligand and the metal ion in the chelate molecule will occur at different pH val-
ues depending on the amino acid ligand employed in the chelation process.33 If the
stability constant of the amino acid chelate is lower due to the selection of an amino
acid ligand with a lower potential formation constant, then that amino bond will
break in a less-extreme acidic or alkaline environment compared to the amino bond
in an amino acid chelate formed with the higher stability constant. This assumes, of
course, that the only variable is the amino acid ligand. Other environmental vari-
ables can and do influence the stability of the chelate.34
This effect of the pH environment on an amino acid chelate is illustrated in
Figure 15.1. When a low-acid pH causes the bond between the metal ion and the
nitrogen in the amino moiety to be broken, the reconfigured molecule acquires a pos-
itive charge on each of its amine termini due to a donation of a hydrogen ion from the
acid environment to the nitrogen. The metal ion continues to remain attached to the
amino acid ligand through the carboxyl bond even though the amino bond has been
broken. In this state, the metal is no longer chelated, but it still remains complexed
to the amino acid ligand and has been termed an amino acid chelate/complex in this
treatise to distinguish it from the amino acid complex that is defined by AAFCO.3
The formation of the amino acid chelate/complex is essential for active transport
of this intact molecule across the mucosal cell membrane from the lumen to occur.
While facilitated diffusion across the membrane of an amino acid chelate can also
take place, the formation of an amino acid chelate/complex is not always a pre
requisite for that absorption to take place. Further, there may occasionally be a small
portion of an ingested dose of an amino acid chelate that is hydrolyzed in the lumen
and the released metal ion absorbed similarly to metal ions from ingested metal
salts. Most of the amino acid chelate, however, that is ingested is absorbed by either
facilitated diffusion or active transport.35 Both facilitated diffusion and active trans-
port require carrier molecules to translocate the chelate and some chelate/complexes,
in the case of facilitated diffusion. Neither transfer process can happen unless the
amino acid chelate or chelate/complex is able to be attached to the carrier molecule
O
Alkaline pH of NH2
approximately 9 and above
O Zn O
H2N O
O
Neutral pH ±
Zn
approximately 2
O
O
NH3+
O Zn O
Acidic pH of
+H N approximately 5 or below
3
O
FIGURE 15.1 The effect of pH on the amino bond of a zinc bisglycinate chelate.
through an electronic attraction between the two molecules. The carrier molecules
for facilitated diffusion or active transport are different. When the amino acid che-
late/complex is attached to an active transport molecule, the two are taken into the
cell through an enzymatic process that requires energy. No energy is required for
facilitated diffusion to occur. Movement into the mucosal cell is dependent on a
concentration gradient. The carrier molecule may be water, a protein molecule, or
some other ion or molecule. In either case, the amino acid chelate is carried into the
mucosal cell as an intact chelate or chelate/complex molecule.
Active transport requires an acid pH environment of approximately 5 or below. At
that pH, the amino bond between the metal ion and the ligand is broken, as described,
and the amine moiety on the terminal end of the amino acid ligand picks up a hydro-
gen ion to form NH3+.32 It is only then that the ligand of the chelate/complex molecule
acquires a positive charge and can be bonded to the negative terminus of the active
transport molecule. When the negative charge on the transport molecule is balanced
by the positive charge, as a result of attaching the amino acid chelate/complex to
the transport molecule, it initiates the enzymatic reactions, including the utilization
of cellular-produced energy, that are necessary to translocate the chelate/complex
across the mucosal cell membrane.36,37 The amino acid chelate/complex that is
absorbed by active transport follows the same active transport pathway employed
by amino acids or peptides because amino acids and peptides behave similarly to an
amino acid chelate/complex in an acid environment.
Amino Acid
Chelate/Complex
Lumen (glutathione)
γ-glu-cysH-gly ADP + Pi
ATP
γ-glutamyl Transpeptidase Gsh Synthetase
glycine
cysH-gly
Membrane γ-glutamyl-amino acid chelate/complex Peptidase γ-glu-cysH
γ-glutamyl Cyclotransferase
cysteine ADP + Pi
γ-glu-cysh Synthetase
ATP
5-Oxoprolinase
Cytoplasm
ATP ADP + Pi
Amino Acid Chelate or Amino Acid Chelate/Complex
FIGURE 15.2 The steps for the absorption of an amino acid chelate/complex across the
intestinal cell membrane by active transport. (Redrawn from Ashmead, HD, “Summary and
conclusions,” in Ashmead, HD, ed., The Roles of Amino Acid Chelates in Animal Nutrition
(Park Ridge, NJ: Noyes) 464, 1993.)
Meister et al. elucidated this active transport mechanism.36 As Figure 15.2 sug-

gests,37 through the positive charge on its amino moiety, the amino acid chelate/
complex is bonded to the negative charge on the glutathione molecule to form
γ-glutamyl-amino acid chelate/complex plus cysteinyl-glycine. This initiates the
commencement of the enzymatic reactions necessary to move the chelate/complex
by active transport across the mucosal membrane. The cysteinyl-glycine that results
from the transpeptidation reaction of the glutathione is cleaved by dipeptidase to
cysteine and glycine, which later are utilized in the re-formation of the glutathione.
The γ-glutamyl transpeptidase, which is located in the brush border of the mucosal
cell membrane, then acts on that glutathione and the γ-glutamyl amino acid chelate/
complex. The resulting γ-glutamyl amino acid chelate/complex is imbued with
energy from the degradation of adenosine triphosphate (ATP) to adenosine diphos-
phate (ADP). This energy is released as a result of removing the cysteinyl-glycine
from the molecule. The released energy is expended moving the γ-glutamyl amino
acid chelate/complex across the cell membrane into the cytoplasm. Once in the cyto-
plasm, the γ-glutamyl amino acid chelate/complex is hydrolyzed into 5-oxoproline
plus the amino acid chelate/complex by the action of γ-glutamyl cyclotransferase.
The γ-glutamyl amino acid chelate/complex that was previously formed by the
transpeptidation is subsequently converted to 5-oxoproline (which is also known
as pyroglutamate or pyrrolidone carboxylate) plus the amino acid chelate/complex
by the action of γ-glutamyl cyclotransferase. At the conclusion of these enzymatic
reactions, the amino acid chelate/complex will have completed its active transport
crossing of the mucosal cell membrane and will be released into the cytoplasm
of the cell, where it is then free to enter into certain assigned metabolic activities
within the mucosal cell or be transferred, intact, to the plasma, while the 5-oxo-
proline becomes the starting material to re-form another molecule of glutathione.
The remainder of the reactions in the γ-glutamyl cycle, which are illustrated in
Figure 15.2, are directed toward resynthesizing the degraded glutathione and renew-
ing the energy source for the active transport of another amino acid, peptide, or
amino acid chelate/complex molecule.36,38
The absorption of an amino acid chelate/complex by active transport can poten-
tially occur at almost any point in the stomach or the small intestine. Just as the
stomach is capable of absorbing free amino acids,38 it is also capable of absorbing
amino acid chelates/complexes.39,40 When introduced into the small intestine, the
absorption of the amino acid chelate/complex can continue in the acidic environ-
ment of the duodenum. As the luminal pH changes from an acid to an alkaline
environment near the common bile duct, the amino acid chelate/complex generally
reconfigures back into an amino acid chelate. At a neutral pH of 7 plus or minus
approximately 2, the amino acid chelate does not carry a charge because the mol-
ecule is in a chelated form.32,33 If intestinal motility moves the chelate near the
luminal wall where the pH remains acidic, the chelate may again reconfigure into
an amino acid chelate/complex.41 If the pH environment continues to remain close
to neutral and does not favor the formation of an amino acid chelate/complex, the
amino acid chelate can still be absorbed into the mucosal cell by solute flow or by
facilitated diffusion if the concentration of the amino acid chelate becomes high
enough to initiate diffusion. Diffusion, in this case, is based on the amino acid
concentration on both sides of the mucosal cell membrane. The metal ion within
the chelate molecule is unreactive36 and does not appear to play a major role in the
diffusion of an amino acid chelate.
As the amino acid chelate enters the alkaline environment of the small intes-
tine (the jejunum and ileum), the higher pH environment in that area of the lumen
can potentially cause the amino bond of each amino acid ligand to again be bro-
ken. Because the pH is alkaline, the amino moiety cannot acquire a positive charge
by picking up an extra hydrogen ion. Thus, the necessary ion attraction between
the amino acid chelate/complex molecule and an active transport molecule cannot
occur in an alkaline environment.32 The bonding of the two molecules can only take
place in an acid environment where the terminal end of the amino moiety gains a
positive charge by picking up an extra H+.
Since most of the lumen is alkaline, this could create a potential problem for
the active transport or some types of facilitated diffusion of the amino acid che-
late/complex across the absorptive cell membrane. The problem is resolved in the
microenvironment near the membrane on the luminal side of the mucosal cell. The
positive charges on integral and peripheral proteins embedded in the membrane of
the mucosal cell41 can potentially attract the amino acid chelate/complex molecules
via ion-dipole forces. As the amino acid chelate/complex formed in the alkaline
environment of the lumen approaches the mucosal cell membrane, it will enter a
pH environment between the microvilli of the mucosal cell in the unstirred water
layer that remains acidic in spite of the alkaline pH of the lumen as a whole.42 This
microacidic environment can cause the terminal end of the amine moiety to acquire
a positive charge as a result of gaining a hydrogen ion, thus allowing the chelate/
complex to bond to an active transport molecule that is embedded in the mucosal cell
membrane and initiate active transport.
Once the amino acid chelate or amino acid chelate/complex has been taken up by
the absorptive cell by either active transport or facilitated diffusion, its fate is depen-
dent on the degree of need of the body for the metal that is attached to the ligand
as well as the amino acids in the ligand itself. Some intracellular hydrolysis of the
amino acid chelate or chelate/complex molecule into the metal ion and amino acid
ligands can, and generally does, occur in the mucosal cell. If the chelate or chelate/
complex is partially or totally hydrolyzed, the released metal can be sequestered into
an intracellular storage molecule, such as apoferritin, or the metal ion can be bound
to a cellularly produced transporter, such as transferrin, both in the case of iron, and
moved to a site of need throughout the body.35
Besides the stability constant of the amino acid chelate or chelate/complex that
enters the mucosal cell and the pH environment in the cytoplasm, the rate of intracel-
lular hydrolysis of the chelate/complex in the absorptive cell partially depends on the
need of the body for that specific metal at the time that the molecule is absorbed. If
the need of the body is substantial, more of the absorbed metallic molecules will be
rapidly transferred, as an intact amino acid chelate or chelate/complex, directly into
body tissues and organs. If the mineral need is not as great, less of the intact mol-
ecule will be transferred to the plasma, and more will be hydrolyzed in the mucosal
cell with the resultant storage of the metal ion. Movement of most of the metal into
the plasma from the absorptive cell is controlled by metabolic reactions that occur
within the mucosal cell. When the metal is removed from the amino acid ligand, the
freed metal ion is subject to the same conditions as is an inorganic metal ion that has
been absorbed into the absorptive cell.43
If the metal ions are separated from their amino acid ligands following cellular
uptake and hydrolysis, in times of greater mineral need more transporter molecules
will be produced in the cytoplasm of the absorptive cell to facilitate increased move-
ment of the released cations into the plasma. When the need for that metal is less,
an increased amount of the metal ions will remain sequestered within the mucosal
cell due to a lower production of transporter molecules.44 Further, the movement of
metal ions from hydrolyzed amino acid chelates and chelate/complexes out of the
mucosal cell into the plasma will occur at about the same rate as would the transfer
of metal ions that were absorbed from an inorganic metal salt source. The rate of
movement in both cases is determined by the need for that particular metal by the
body (Figure 15.3).45
Figure 15.3 summarizes a previously described study in which 40 infants and
young children, all suffering from severe iron deficiency anemia, were matched
for sex, age, and hemoglobin values before assignment to two groups. Each child
received a daily dose of 5 mg of iron as either ferrous sulfate or iron bisglycine che-
late per kilogram of body weight for 28 days. During the test period, both groups
exhibited significant improvements in their hemoglobin levels (p < 0.001). The
regression analysis shown in Figure 15.3, however, indicated that the uptakes of both
sources of iron into hemoglobin were similarly regulated by the body according to
7
Ferrous Bis-Glycinate chelate R = –0.6611
6 Ferrous Sulfate R = –0.8627
Change in Hemoglobin, g/dL
0
4.5 5 6 7 8 9 10 11
Basal Hemoglobin, g/dL
FIGURE 15.3 Regression analysis demonstrating that iron absorption values are inversely
proportional to hemoglobin levels, regardless of the iron source. (Redrawn from Pineda, O,
and Ashmead, HD, “Effectiveness of treatment of iron-deficiency anemia in infants and
young children with ferrous bis-glycinate chelate,” Nutrition 17:381–384, 2001.)
changes in hemoglobin. This suggested that while absorbed in greater amounts, a

substantial portion of the amino acid chelate that was taken up by the mucosal cells
was probably hydrolyzed while still in the mucosal cells before the iron was utilized
to produce more hemoglobin molecules. This is the only way that the regulatory
process illustrated in Figure 15.3 could occur.45 As the hemoglobin levels increased,
the uptake of iron from either source decreased similarly.
It should also be remembered that, in this example, the hemoglobin values
controlled the rates of transfer of both sources of iron. The significant difference
between the two groups is not that hemoglobin repletion was similarly regulated
in both groups. The infants and young children used in this study also had ferritin
levels that were severely depleted. The ferritin values were only repleted in those
subjects receiving the amino acid-chelated source of iron. Iron uptake into the hemo-
globin molecules was controlled by the rate of iron saturation in the hemoglobin.
When hemoglobin values attained sufficiency, iron transfer to the plasma from the
mucosal tissue became very low (more ionic iron storage in the mucosal cells with
resulting lower transfer of that form of iron to the plasma). Concurrently, however, it
appeared that additional iron, absorbed into the mucosal tissue from the amino acid
chelate source, was transferred to the ferritin molecules to replete their depleted sta-
tus, possibly as intact amino acid chelate/complex molecules. The apparent bioavail-
ability of the iron from the amino acid chelate was calculated to be 90.9% compared
to 26.7% for the iron from the ferrous sulfate. Thus, in a time of acute need, more
iron from the amino acid chelate was available to address all of the iron deficiency
issues, including ferritin levels, compared to ferrous sulfate because some of the
amino acid chelate was not hydrolyzed in the mucosal cell but transferred intact to
the area of need.
Guthrie has suggested that nonheme iron ions must be chelated with amino acids
that are sourced from the chyme prior to their being absorbed into the mucosal cell.44
If the body has an immediate need for that absorbed iron, transferrin in the plasma
will come in contact with the mucosal cell. The iron will be complexed, taken up
from the intestine, and carried across the mucosal cell by the same amino acids that
initially complexed the iron in the lumen. The iron will immediately be released on
the other side of the basement membrane to combine with the plasma transferrin.
The iron-transferrin complex can carry two atoms of iron per molecule of trans-
ferrin to the body cells requiring the iron. When the available transferrin is about
33% saturated with iron, it will not pick up additional iron from the mucosal cells.
Instead, the ferrous iron is released from the amino acids that initially chelated it in
the lumen, oxidized to ferric iron, and stored within the apoferritin molecule within
the mucosal cell.44 While there may be some disagreement with the idea that all non-
heme iron ions must first be chelated to amino acids from chyme prior to mucosal
cell absorption, the rest of the model proposed by Guthrie could easily apply to the
description of the fate of a significant portion of the amino acid chelate/complex that
is absorbed into the mucosal cell.
The foregoing is not meant to imply that the entire dose of the amino acid chelate
or chelate/complex that is taken up by the mucosal cell must be hydrolyzed prior to
the metal being transferred to the plasma. Some of the absorbed amino acid che-
late or chelate/complex remains intact and is transferred into the plasma as such. If
one ignores the discussion relating to the need for a metal by the body, the percentage
of amino acid chelate or chelate/complex not hydrolyzed in the mucosal cells is, to
an extent, dictated by the stability constant of the molecule. The higher the stability
constant of the chelate or chelate/complex is, the smaller will be the percentage of
mucosal cell hydrolyzation of that molecule. Thus, using iron as an example, more of
an iron methionate chelate will be hydrolyzed within the mucosal cells than will be
iron lysinate.35 This is because a bisamino acid chelate formed from iron and methio-
nine has a stability constant of 6.4810 compared to 9.010 for an iron bislysinate.46
Stability constants can also be dictated by the size of a chelate ring. The larger
the number of members in the chelate ring, the lower the stability constant of that
chelate and the greater is its potential for intracellular hydrolyzation, assuming that
the chelate/complex is absorbed intact. Thus, small amino acid ligands tend to form
stronger chelates than do larger amino acids. The number of rings of the re-formed
chelate will also influence its stability constant. As the number of chelate rings
increases, so does the stability of the chelate and the lower the rate of mucosal cell
hydrolysis will be. And finally, the basicity of the amino acid will affect its stability.
The more basic the ligands forming the chelate, the stronger the chelate will be.34
From the discussion, one could conclude that more of an amino acid chelate that is
created with more than one highly basic smaller amino acid ligand would likely be
transferred to the plasma intact compared to a chelate of the same metal created with
larger and less basic amino acid ligands.
There are several studies suggesting that some portion of an ingested amino
acid chelate will arrive in the plasma intact regardless of the amino acid ligand
employed.47–49 Indeed, in an in vitro double-isotope study, in which both the metal
and the amino acid ligand were tagged,35 it was estimated that, in that particular
study, approximately 50% of the absorbed amino acid chelate was transferred to the
serosa intact.50
The delivery of intact amino acid chelate or chelates/complexes to the tissue or
organ results in a different biological response compared to delivery of the same
minerals ingested as inorganic metal salts. While these differences can be partially
attributed to the greater absorption of the amino acid-chelated form of minerals, that
does not account for all of the difference. For reasons not completely elucidated,
tissue retention of metals from ingested amino acid chelates is greater than is tissue
retention of those same metals when they are ingested as inorganic metal salts.51–53
This increased retention allows for greater metabolic activity related to the metal
atom in the chelate and probably its amino acid ligand. As additional amino acid
chelate or chelate/complex molecules are delivered, retained, and begin to initiate
metabolic activity, the overall performance of the organism increases over time. In
a previously discussed example, Ashmead and Samford conducted a 3-year study
in dairy cows from the same herd.54 Sixty-four animals received a formula contain-
ing magnesium, zinc, manganese, and copper as amino acid chelates, and 65 other
cows matched for age, initial milk production, parity, and genetics received the
same mineral supplement formula except the minerals were provided as inorganic
metal salts.54 The mineral supplementation program, the feeding program, and so on
remained the same for all of the animals in both groups throughout the entire three
lactations. The only difference between the two groups of animals was the source of
the minerals provided in their supplement program. The supplements were isomin-
eral and isonitrogenous.
As Table 15.1 summarizes,54 in each successive lactation period, the difference in
milk, milk fat, and protein production between the two groups became increasingly
greater. The initial biological response difference seen in Table 15.1 could only be
TABLE 15.1
Mean Percentage Increases in Milk and Fat
and Protein Production between Dairy Cows
Receiving Amino Acid Chelates versus
Inorganic Metal Salts
Lactation 1 Lactation 2 Lactation 3
Milk production +0.3% +2.6%a +11.4%a
Milk fat –3.2% +0.8%a +14.2%a
Milk protein N.D.* +1.8% +10.2%a
Source: Redrawn from Ashmead, HD, and Samford, RA,

“Effects of metal amino acid chelates or inorganic min-
erals on three successive lactations in dairy cows,” Int J
Appl Res Vet Med 2:181–188, 2004.
a Differences from previous lactation are significant (p < 0.05).
* Not determined.
due to the greater absorption resulting from ingestion of the chelated source of miner-
als and their subsequent increased tissue retention, but even that difference between
the two groups did not remain consistent from lactation period to lactation period. If
the initial difference in mineral uptake and tissue retention of the absorbed minerals
had remained the same for each successive lactation period, then one would have
predicted that the percentage differences in milk, fat, and protein production in lac-
tation periods 2 and 3 would have been similar to the differences observed in the first
lactation period. Instead, the differences between the two groups of cows became
increasingly greater with each successive lactation period, indicating increasingly
greater mineral retention in the animals receiving the amino acid-chelated source,
which allowed for the greater mineral-related metabolic activity.
Because a portion of the amino acid chelate/complex can be taken up by the
plasma as an intact molecule, there is another benefit attached to the ingestion of this
source of minerals: the ability to prepare amino acid chelates that can deliver more
of a desired metal to a target tissue or organ.55,56 Certain tissues or organs require
more of a specific amino acid for their metabolic purposes than do certain other
tissues or organs. For example, arginine and zinc are both required in significant
amounts for sexual activity in the male.57,58 When the zinc is chelated to the arginine,
more of that zinc can potentially be delivered to the male sex organs than when the
zinc is chelated to certain other amino acids.56
Certainly, some of the absorbed chelated zinc arginate in the example will be
hydrolyzed into zinc ions and arginine following uptake into the mucosal cells.
And just as certainly, some of that absorbed zinc arginate will survive that mucosal
hydrolysis and be delivered intact into the plasma. When that occurs, the surviving
zinc arginate will potentially be distributed to various tissues and organs throughout
the body as an intact molecule. But, since there is a greater requirement for arginine
by the male sex organs, and because the zinc remains attached to the arginine as the
zinc arginate is introduced into the plasma, more of the zinc will be delivered to
the male sex organs along with the arginine.56
Of the zinc arginate that was hydrolyzed in the mucosal cell, some of that free
arginine and some of that free zinc could also be sent to the male sex organs. The
difference is that these two individual components will be delivered separately and
not as a zinc arginate. One of the main problems arising from the separate delivery
of the two components is that, following mucosal cell hydrolysis of zinc arginate, the
rates of exit of the metal and the amino acid from the mucosal cells into the serosa
are not the same.35 Thus, the metal-ligand ratio in the plasma may not be as condu-
cive to the re-formation of a zinc arginine complex in the sexual organs and tissues.
If the metal and amino acid do not balance each other out for purposes of metabolic
activity, the nutrient that is in excess will probably be excreted as endogenous, with
resulting lower overall metabolic activity as it relates to these two nutrients.
In conclusion, the ingestion of a proven amino acid chelate will result in greater
absorption of that metal with consequentially enhanced mineral-related metabo-
lism due to the bonding of the metal to the amino acid ligands. Chelating the metal
with amino acids protects it or reduces the probability of the metal entering into
absorption-limiting reactions while residing in the gastrointestinal tract. Most
of the ingested amino acid chelate can be absorbed intact into the mucosal cells
from the lumen. Intracellular hydrolysis will probably destroy part of the chelate
or chelate/complex that has been delivered to the cytoplasm of the mucosal cell.
That metal derived from that intracellular hydrolysis will be treated as if it had been
absorbed similarly to the absorption of a metal ion. The remainder of the amino acid
chelate or chelate/complex that is not hydrolyzed in the mucosal cell will traverse the
mucosal cell and enter the plasma intact. Following entry into the plasma, the amino
acid chelate or chelate/complex will be carried to areas of need within the body.
There, the tissues and organs will take up the amino acid chelate or chelate/complex
as if it were a proteinaceous metallic molecule that originated from ingested food.
As a result of this process and the higher absorption rate of the amino acid chelate,
metabolic activity associated with the mineral and its ligand is augmented for a lon-
ger period of time.
REFERENCES
1. Werner, A, “Beitrag zur constitution Anaorganisiher Verbindungen,” Z Anorg U Allgem
Chem 3:267, 1983.
2. Mellor, DP, “Historical background and fundamental concepts” in Dwyer, FP, and
Mellor, DP, eds., Chelating Agents and Metal Chelates (New York: Academic Press) 20,
1964.
3. Krebs, S, ed., Official Publication Association of American Feed Control Officials
Incorporated (Oxford, IN: AAFCO) 310, 361, 2008.
4. Wasserman, RH, Comar, CL, Schooley, JC, and Lengemann, FW, “Interrelated effects
of L-lysine and other dietary factors on the gastrointestinal absorption of calcium 45 in
the rat and chick,” J Nutr 62:367, 1957.
5. Cardon, BP, “The importance of chelation in trace mineral nutrition,” paper presented at
Montana Nutrition Conference, Billings, February 11, 1963.
6. Scott, ML, “Some practical aspects of chelates in animal nutrition,” Feedstuffs 37:31–32,
February 13, 1965.
7. MacGregor, D, personal communication, December 7, 1963.
8. Hinze, P, and Ashmead, H, “The use of Metalosates™ in bovine nutrition,” unpublished
9. Hinze, PM, “Metalosates: Their electromotive potential in laying hens,” Feed Manag
20:28–30, 1969.
Spring-Summer, 1970.
11. Shores, A, and Jacobs, DL, “Canine disc disease: Charges and rebuttals,” Vet Med/Small
Animal Clin 72:1798–1804, 1977.
12. Brady, PS, “Evaluation of an amino acid iron chelate hematinic,” Research Report 341,
Michigan State University, East Lansing, 4–6, 1975.
13. Hartle, JW, and Ashmead, SD, “Single laboratory validation of the quantification of che-
at 118th AOAC International Meeting, St. Louis, MO, September 2004.
14. Patton, R, “Chelated minerals” What are they, do they work?” Feedstuffs 62:14–17,
February 26, 1990.
15. Ashmead, HD, “Reader disputes chelated minerals article,” Feedstuffs 62:12, April 16,
1990.
16. Brown, TF, and Zeringue, LK, “Laboratory evaluations of solubility and structural
integrity of complexed and chelated trace mineral supplements,” J Dairy Sci 77:181,
1994.
17. Pineda, O, “Iron bis-glycine chelate competes for the non-heme absorption pathway,”
J Am Clin Nutr 77:495–496, 2003.
18. Leach, GA, and Patton, RS, Evaluation of Chelated Mineral Analysis Techniques,
monograph, Chelated Minerals International, 1997.
19. Dalley, NK, “Report on x-ray diffraction crystallography of Albion® zinc amino acid
chelates,” unpublished research report, Brigham Young University, Provo, UT, 1985.
21. Parks, FP, and Harmston, KJ, “A field study of trace mineral complexes,” study pre-
sented to AAFCO for use as a quality control assay, Indianapolis, IN, 1995.
22. Parks, FP, and Harmston, KJ, “Judging organic trace minerals,” Feed Manage 45:35,
October 1994.
23. Leach, GA, and Patton, RS, “Analysis techniques for chelated minerals evaluated,”
Feedstuffs 701:13–15, March 31, 1997.
24. Hynes, MJ, Colleran, D, Fin, J, Headon, DR, and Lyons, TP, “Quantitative chemical assess-
ment of some mineral proteinates used in animal feeds,” J Animal Sci 68 (Supp1.):408,
1997.
25. Little, P, and Ashmead, H, “Metal ionization chromatography,” unpublished research
report, Albion Laboratories, Clearfield, UT, 1976.
26. James, H, “A differential pulse polarography study of zinc glycinate chelate,” unpub-
lished research report, Weber State University, Ogden, UT, 1987.
27. Holwerda, RA, Albin, RC, and Madsen, FC, “Chelation effectiveness of zinc proteinates
demonstrated,” Feedstuffs 19:12–13, June 19, 1995.
28. Olsen, JI, and Schweized, MP, “NMR analysis of Chelazome® products,” unpublished
research report, University of Utah, Salt Lake City, UT, 1987.
29. Rogers, K, and Landcaster, J, “Report on electron paramagnetic resonance spectra of
Albion® metal amino acid chelates,” unpublished research report, Utah State University,
Salt Lake City, UT, 1985.
30. Hartle, JW, and Ashmead, HD, “Bonds important for amino acid chelates,” Feedstuff
78:16–17, September 11, 2006.
31. Dalley, NK, “Report on x-ray diffraction crystallography of Albion® zinc amino acid
chelate,” unpublished research report, Brigham Young University, Provo, UT, 1985.
32. Ericson, C, and Ashmead, SD, “A novel approach in confirming dietary amino acid che-
34. Kratzer, FH, and Vohra, P, Chelates in Nutrition (Boca Raton, FL: CRC Press) 25, 1986.
37. Ashmead, HD, “Summary and conclusions,” in Ashmead, HD, ed., The Roles of Amino
Acid Chelates in Animal Nutrition (Park Ridge, NJ: Noyes) 464, 1993.
Hill) 633, 1973.
39. Kornegay, ET, Swinkels, JWGM, Webb, KE, and Lindermann, MD, “Absorption of
zinc amino acid chelate and sulfate during repletion of zinc depleted pigs,” in Anke,
M, Meissner D, and Mills, CF, eds., Trace Elements in Man and Animals—TEMA8
(Gersdof, Germany: Verlag media Touristik) 398–399, 1993.
40. Kim, SW, “Zinc amino acid chelates as upgraded zinc sources for monogastric animals,”
paper presented at Albion Animal Nutrition Conference, Midway, UT, January 2007.
41. Giese, AC, Cell Physiology (Philadelphia: Saunders College) 192, 1973.
158–165, 1975.
proof of chelation,” Proceedings of Albion Laboratories, Inc. International Conference
on Human Nutrition, Salt Lake City, UT, January 1995.
44. Guthrie, HA, Introductory Nutrition (St. Louis, MO: Times Mirror/Mosby College)
293–294, 1989.
45. Pineda, O, and Ashmead, HD, “Effectiveness of treatment of iron-deficiency anemia in
infants and young children with ferrous bis-glycinate chelate,” Nutrition 17:381–384, 2001.
47. Bovell-Benjamin, AC, Viteri, FE, and Allen LH, “Iron absorption from ferrous bisglyci-
71:1563–1569, 2000.
48. Ashmead, D, and Graff, D, “Placental transport of chelated iron,” Proceedings of the
International Pig Veterinary Society, Mexico City, Mexico, July 1982.
49. Martin, RB, “Complexes of g-amino acids with chelatable side chain donor atoms” in
Sigel, H, ed., Metal Ions in Biological Systems (New York: Dekker) v9, 1–37, 1979.
50. Pineda, O, personal communication, 2001.
51. Ashmead, HD, Ashmead SD, and Samford, RA, “The effects of metal amino acid che-
lates on milk production, reproduction and body condition in Holstein first calf heifers,”
Int J Appl Res Vet Med 2:252–260, 2004.
52. Jensen, NL, “Biological assimilation of metals,” U.S. Patent 4,167,564 Washington,
53. Ashmead, HD, “Comparative intestinal absorption and subsequent metabolism of
metal amino acid chelates and inorganic salts,” in Subramanian, KS, Iyengar, GV, and
54. Ashmead, HD, and Samford, RA, “Effects of metal amino acid chelates or inor-
ganic minerals on three successive lactations in dairy cows,” Int J Appl Res Vet Med
2:181–188, 2004.
55. Ashmead, HH, and Graff, DJ, “Zinc amino acid chelates and the male sex organs,”
monograph, Albion Laboratories, Clearfield, UT, July 1987 (unpublished).
56. Ashmead, HH, Ashmead, HD, and Graff, DJ, “Amino acid chelated compositions for
delivery to specific biological tissue sites,” U.S. Patent 4,863,898, Washington, DC,
September 5, 1989.
57. Fahim MS, Wang M, Sutcu MF, and Fahim, Z, “Zinc arginine, a 5 alpha-reductase inhib-
itor, reduces rat ventral prostate weight and DNA without affecting testicular function,”
Andrologia 25:369–375, 1993.
58. Prasad, AS, “Deficiency of zinc and its toxicity” in Prasad, AS, ed., Trace Elements in
Human Health and Disease (New York: Academic Press) v1, 312, 1976.

Amino Acid Chelation in Human and Animal Nutrition (PDFDrive)

Uploaded by

Copyright:

Available Formats

Amino Acid Chelation in Human and Animal Nutrition (PDFDrive)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Amino Acid Chelation in Human and Animal Nutrition (PDFDrive)

Uploaded by

Copyright:

Available Formats

AMINO ACID

Boca Raton London New York

CRC Press is an imprint of the

© 2012 by H. DeWayne Ashmead

No claim to original U.S. Government works

International Standard Book Number-13: 978-1-4398-9768-3 (eBook - PDF)

and the CRC Press Web site at

Chapter 1 The Fundamentals of Mineral Nutrition............................................... 1

Chapter 2 The Chemistry of Chelation................................................................ 19

Chapter 3 The History of Nutritional Chelates.................................................... 35

Chapter 4 The Requirements for a Nutritionally Functional Chelate.................. 49

Chapter 5 The Development of Analytical Methods to Prove Amino Acid

Chapter 6 Absorption of Amino Acid Chelates from the Alimentary Canal...... 81

Chapter 7 The Pathways for Absorption of an Amino Acid Chelate...................97

Chapter 8 The Absorption of Amino Acid Chelates by Active Transport........ 117

Chapter 9 The Absorption of Amino Acid Chelates by Facilitated Diffusion.... 135

Chapter 12 Tissue Metabolism of Amino Acid Chelates.................................... 185

Chapter 14 Toxicity of Amino Acid Chelates...................................................... 223

Chapter 15 The Absorption and Metabolism of Amino Acid Chelates............... 233

Wayne Askew, PhD

active transport. The relationship between absorption through passive diffusion,

Boyd R. Beck, PhD

PANTOTHENIC ACID SULFUR

THIAMIN Carbohydrate IODINE

VITAMIN B12 COPPER

A small number of enzymes, such as pepsin or trypepsin, are composed exclu-

Histidine Glutamic Acid

Pantothenic Acid Riboflavin

1. The formation of insoluble complexes between dissimilar ions

3. The complexing of ions by metal-binding proteins

MoO42- → MoO3S2- → MoO2S22- → MoOS32- → MoS42-

In animal diets, if molybdenum intake exceeds 10 mg/kg of dry matter, the

Source: Data from Ashmead, H, “Tissue trans-

A second example is the enzyme carboxypeptidase. This enzyme is activated by

Source: Data from Ashmead, HD, and Ashmead, SD, “The

The following theoretical example illustrates the possible consequence of ingest-

Vegetable Origin Animal Origin

Incubation Illness Convalescent

FIGURE 2.1 Platinum bis-ethylenediamine chloride structure as proposed by Alfred

The purpose of this chapter is to review, on an elementary level, the chemical

bonding. Further, bonding in the d orbital results in different characteristics com-

b. After the loss of a proton:

FIGURE 2.6 Zinc bisglycinate chelate as determined by x-ray diffraction spectrometry.

coupled plasma spectroscopy, electrospray mass spectroscopy, and combustion ele-

The Chemistry of Chelation

9. Nobelstiftelsen, A, Nobel Lectures Chemistry 1901–1920 (Hackensack, NJ: World

Source: Summarized from Ashmead, D, Beck, B, and Hopson, H, “A new prophylactic

Birth 1 Week 2 Week 3 Week

Miller expanded his theoretical views in subsequent publications.31–33

Source: Summarized from Ashmead, H, Jensen, N, and Graff, D, “Stability constants of

Source: Summarized from Ashmead, H, Jensen, N, and Graff, D, “The influ-

Source: Summarized from Ashmead, H, Ashmead, D, and Jensen,

body portion being analyzed (cc/min/g) average of rats.

1. the size of the chelate ring;

The digestive system is relatively efficient. It is estimated that several hundred

HI-PROT REGEN NH+3 CH2CH2CH2CH2CHNH+3 COO– HI-PROT (sl.)

8. Hughes, M, The Inorganic Chemistry of Biological Processes (London: Wiley) 71–72,

32. Johnson, B, “Molecular weights of several commercial chelates,” unpublished report,

As described previously, he reportedly produced the chelate in solution using a ratio

In order to correct this tending to poor absorption of … metals, it is necessary to com