See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/244554553

Speciation, quantification and stability of selenomethionine, S-

(methylseleno)cysteine and selenomethionine Se-oxide in yeast-based
nutritional supplements

Article  in  Journal of Analytical Atomic Spectrometry · July 2007

DOI: 10.1039/b703381h

CITATIONS READS

19 479

5 authors, including:

Chey Kahakachchi Peter C Uden

Mallinckrodt University of Massachusetts Amherst
12 PUBLICATIONS   358 CITATIONS    248 PUBLICATIONS   6,233 CITATIONS   

SEE PROFILE SEE PROFILE

Julian Tyson
University of Massachusetts Amherst
286 PUBLICATIONS   5,654 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Determination of inorganic arsenic in rice. View project

Phytoremediation of Environmental Contaminants View project

All content following this page was uploaded by Julian Tyson on 04 June 2014.

The user has requested enhancement of the downloaded file.

PAPER www.rsc.org/jaas | Journal of Analytical Atomic Spectrometry

Speciation, quantification and stability of selenomethionine,

S-(methylseleno)cysteine and selenomethionine Se-oxide in
yeast-based nutritional supplements
Prince O. Amoako, Chethaka L. Kahakachchi, Elena N. Dodova,
Peter C. Uden* and Julian F. Tyson
Received 6th March 2007, Accepted 8th June 2007
First published as an Advance Article on the web 29th June 2007
DOI: 10.1039/b703381h

The speciation and stability of organoselenium compounds present in the selenized yeast
evaluated in National Cancer Institute human intervention trials were examined along with those
of related supplements. Total selenium was measured by inductively coupled plasma optical
emission spectroscopy (ICP-OES), ICP-mass spectrometry (ICP-MS) and/or electrothermal
atomic absorption spectrometry (ET-AAS) after microwave digestion. Enzymatic extraction
enabled selenium speciation profiles to be obtained by ion-pair reversed phase high performance
liquid chromatography (HPLC) with ICP-MS detection. Extracts were also subjected to
chloroformate derivatization followed by Se and S specific GC analysis with atomic emission
detection (GC-AED). Currently available and archived supplements show major differences in
speciation for selenomethionine, selenomethionine Se-oxide and S-(methylseleno)cysteine with
respect to time and conditions of storage. The formation of the latter compound is reported
under different synthetic and elevated temperature conditions.

Introduction to be evaluated, this agent being chosen in part by reason of

the presence or formation in yeast of other organoselenium
The biochemistry, toxicology and nutritional importance of species in addition to SeMet. The marketing of selenium
selenium have been reviewed regularly.1–5 Selenium (Se) is an supplements in recent years has increased but regulation with
essential micronutrient for humans and animals being found in respect to quality, storage conditions, stability and content of
many foods, with grain products having higher concentra- selenium supplements is problematic. Further selenium specia-
tions. Concentration varies with geographical origin of the tion for selenized yeast and selenium supplements is needed to
agricultural products and on the selenium content of the soil. quantify active components and potential anticarcinogenic
Human selenium deficiency has been noted in some global components.16–23
regions as a result of low soil selenium levels and dietary Prior to the determination of selenium species in yeast by
sources, a Se plasma level at or below 85 mg L 1 being HPLC with ICP-MS detection, an extraction technique is
indicative.6,7 Selenium dietary deficiency has been linked to needed to liberate them from the yeast matrix. Water has been
heart diseases,7 arthritis,8 cancer,9 and AIDS.10 Dietary intake used to extract soluble low molecular-mass selenium species
of Se in the United States is typically 80 to 120 mg day 1 and from yeasts, but with poor efficiency.24 Higher extraction
the recommended dietary allowance (RDA) is 55 mg day 1; efficiencies have been achieved with proteolytic enzymes,25
this exceeds Se intake in many world regions.6 For those hydrochloric acid26 and methanesulfonic acid.27 The more
seeking to increase their selenium intake, selenium-enriched widely employed proteolytic enzymes for species liberation,
foods have been recommended.11,12 Clark et al. reported that most closely resemble the physiological conditions in the
dietary supplementation by 200 mg of selenium per day, human intestine, whereas harsher chemical hydrolysis may
administered in high selenium yeast, significantly reduced cause degradation or alteration of selenium species of physio-
lung, prostate, and colorectal cancer in the Nutrition Preven- logical importance. Enzymatic hydrolysis usually requires long
tion of Cancer (NPC) trial, but all Se species in these supple- extraction periods but a rapid extraction procedure using
ments were not characterized.13 Further studies are on-going focused ultrasound in conjunction with enzymatic hydrolysis,
in major clinical trials, the European Prevention of Cancer by enzymatic probe sonication (EPS), for total selenium and
Intervention with Selenium (PRECISE)7 trial and the Sele- selenomethionine determination in selenium-enriched yeast
nium and Vitamin E Cancer Prevention Trial (SELECT).14,15 has been reported.28 A multi-laboratory round-robin evalua-
In the PRECISE trial, selenized yeast is the intervention agent. tion of a proposed selenized yeast reference material (SELM-
In the SELECT trial, the effect of selenomethionine (SeMet) is 1) (NRC, Canada) has recently emphasized the challenges
inherent in the acquisition of accurate and precise data for
selenomethionine and methionine content.29 The need can be
Department of Chemistry, Lederle Graduate Research Tower A 701,
University of Massachusetts, 710 North Pleasant Street, Amherst, argued, however, for data for other selenium species, such as
MA 01003-9336, USA selenomethionine selenoxide and S-(methylseleno)cysteine,

938 | J. Anal. At. Spectrom., 2007, 22, 938–946 This journal is

c The Royal Society of Chemistry 2007
prior to any formalization of a standard reference selenized column (3.9 mm  15 cm) (Waters Corporation, Milford, MA,
yeast, a goal which has been in view for a number of years. USA) were used. The injected sample volume was 20 mL, and
Without such a more comprehensive investigation and analy- the flow rate of the mobile phase was 1 mL min 1. The mobile
sis, a fully speciated reference selenized yeast for method phase composition was 0.1% heptafluorobutanoic acid
validation remains an unaccomplished goal. Integrated chro- (HFBA), 1% methanol and 99% water (v/v). A PerkinElmer
matographic strategies of size-exclusion, anion exchange and Optima 4300 DV ICP-optical emission spectrometer was used
reversed-phase ion pair (RP-IP) in combination with ultra- for total selenium determination, operating conditions being
sonic nebulization, ICP-MS and electrospray MS–MS were shown in Table 1.
used to optimize speciation of selenized yeast and Se-(methyl) A PerkinElmer 4100ZL atomic absorption spectrometer
selenocysteine based supplements.30 Simultaneous speciation equipped with longitudinal Zeeman-background correction
of Se-containing glutathione species in selenized yeast has also and a transversely heated graphite furnace with a transversely
been recently reported in a similar study.31 heated graphite atomizer (THGA) tube, equipped with an
In the present study, selenium nutritional supplements, integral L’vov platform (PerkinElmer part number BO504053)
including those administered in human intervention trials, (ET-AAS) was used. A PerkinElmer AS-70 autosampler in-
and commercially available selenized yeast products were jected the standards and samples in to the graphite furnace.
examined. The goals of the study were to characterize and The selenium electrodeless discharge lamp (EDL) was pow-
quantify the selenium species present in the enzymatic hydro- ered by a PerkinElmer system 2 EDL and operated at 260 mA.
lysates of these materials and compare speciation profiles with A slit width of 2 nm and a wavelength of 196 nm were selected
regard to sample age, storage and subjected temperatures with for the experiments. The inert carrier gas was argon (Mer-
particular emphasis on the formation of S-(methylseleno) riam–Graves, high purity), the flow rate being 250 mL min 1
cysteine, the recently identified Se–S aminoacid metabolite of during all stages of the furnace program, except during the
selenized yeast.32,33 atomization step when the gas flow was stopped. For total
selenium determination the ET-AAS furnace heating pro-
grams were run under optimized conditions recommended
Materials and methods by PerkinElmer.
Instrumentation A Hewlett Packard (Agilent) HP 5921A atomic emission
detector interfaced to a HP 5890 II gas chromatograph (GC-
Elan 5000 and subsequently Elan DRC-e inductively plasma AED) was used with hydrogen plasma reagent gas. Eluted
mass spectrometers (Perkin–Elmer Sciex, Ontario, Canada) selenium species were monitored at the 196 nm emission line.
were used as HPLC detectors. Operating conditions for the Sulfur and carbon-containing species may be simultaneously
spectrometers are shown in Table 1. The HPLC system has monitored at the adjacent emission lines 181 nm (S) and
been described previously.32 A 5 mm Symmetry Shieldt RP-C8 193 nm (C), respectively. A 30 m SE-30 capillary column of
0.32 mm internal diameter and 0.25 mm film thickness was
Table 1 Operating conditions of the ICP-MS and ICP-OES instru- employed with helium carrier flow rate 1 mL min 1 and
ments
injection split ratio 30:1. The GC oven was held at 60 1C for
ICP-MS Perkin Elmer Elan DRC-e the first 10 min and then heated to 300 1C at a rate of 5 1C
RF power: 1100 W min 1 and held there for 2 min. The solvent vent on the
Plasma Ar flow rate: 15 L min 1
Nebuliser Ar flow rate: 0.98 L min 1 detector was set at 7 min to avoid the introduction of excess
Auxiliary gas flow rate: 1.20 L min 1 solvent into the plasma.
Cell gas (CH4) flow rate: 0.6 ml min 1
78
Isotopes monitored: Se, 80Se, 82Se
Dwell time 80 ms Chemicals
Nebulizer Cross flow
Spray chamber Scott All reagents used were of analytical reagent grade. Barnstead
E-pure 18 MO water (Boston, MA, USA), methanol (HPLC
ICP-MS Perkin Elmer Elan 5000
RF power: 1100 W
grade) and HFBA (Aldrich, Milwaukee, WI, USA) were used
Plasma Ar flow rate: 15 L min 1 throughout the experiments. Palladium (10 000 mg L 1) and
Nebuliser Ar flow rate: 0.88 L min 1 magnesium nitrate (10 000 mg L 1) modifier solutions for ET-
Auxiliary gas flow rate: 0.80 L min 1 AAS analysis were from PerkinElmer. For microwave diges-
82
Isotopes monitored: Se, 72Ge
Dwell time 1000 ms tions, nitric acid and 30% (v/v) hydrogen peroxide (Fisher
Nebulizer Cross flow Scientific) were used. Sodium selenite [Na2SeO3], sodium
Spray chamber Scott selenate [Na2SeO4], DL-selenomethionine [CH3SeCH2CH2CH-
(NH2)COOH] and Protease XIV were obtained from Sigma–
ICP-OES Perkin Elmer Optima 4300 DV
RF power: 1300 W Aldrich Chemical Company (St. Louis, MO, USA). Stock
Plasma Ar flow rate: 15 L min 1 standard solutions (1000 mg L 1 Se) were prepared in ultra
Nebuliser Ar flow rate: 0.80 L min 1 pure water (18 MO Barnstead E-pure) and solutions were
Auxiliary gas flow rate: 0.20 L min 1
Plasma view: Axial stored between 0 and 4 1C and in the dark. S-(methylseleno)-
Wavelength monitored: Se 196 nm cysteine [CH3SeSCH2CH(NH2)COOH] was synthesized and
Nebulizer Cross flow characterized as reported previously.32,33 Selenomethionine
Spray chamber Cyclonic Se-oxide (hydrate) [CH3Se(OH)2CH2CH2CH(NH2)COOH]

This journal is
c The Royal Society of Chemistry 2007 J. Anal. At. Spectrom., 2007, 22, 938–946 | 939
 98  1.2

97  1.0

2  0.1
35  0.4

13  0.2
Selenium content per tablet specified on label (mg). b Selenium species identified by HPLC-ICP-MS in the present study after enzymatic hydrolysis: 1: selenomethionine Se-oxide hydrate, 2:
S-(methylseleno)cysteine, 3: selenomethionine, 4: selenate. c The % selenium distributions (mean  standard error; n = 3) are expressed in terms of total selenium content eluting from the column based
was prepared by adding excess oxidizing agent, 50 mL of 30%
(v/v) hydrogen peroxide to a 500 mL aliquot of a 0.1 M HCl

—
4
solution of selenomethionine (1 mg L 1 Se) and left overnight

81  1.1

53  0.8
23  0.3

88  1.2
in the dark. This and other oxidized selenium species have
been characterized and authenticated in various selenized

—
yeast and selenomethionine supplements.34–36

3
Selenium species (%)b,c

2  0.10
18  0.4
7  0.2
Selenized yeast samples
A selenium yeast slurry sample (a liquid sample of live Se-

—
2
yeast) and Selenolife yeast tablets (Lot #0094; 200 mg Se per

1  0.1

5  0.2

22  0.3
34  0.6

1  0.1
tablet) were provided by LeSaffre Yeast Corporation-Red Star
Nutritional Yeast (Milwaukee, WI, USA). Selenium-enriched

—
1
yeast powder with a total selenium content of 1250 mg g 1, dry

Total
matter (10%), a minimum protein content of 50% (w/w),

200

200

200
Sea

20

50

50
and a phosphate content in the range of 2.5–3.4% (w/w)
(SelenoPreciset also designated SelenoExcellt) was obtained

Vitamins, minerals, herbs and botanical extracts among other ingredients. Selenium as L-selenomethionine
Natural source of selenium from Indian Mustard (Brassica juncea) stem and leaf. Also contains cellulose,
from Cypress Inc. (Fresno, CA, USA). The selenium supple-
ments (in Table 2) were obtained from local retail stores.

Vitamin E (as dl-alpha tocopheryl acetate), selenium from selenium yeast. Also contains gelatin,
cellulose (plant origin), vegetable stearic acid, vegetable magnesium stearate and croscarmelose
Procedure

Vitamin C, bioflavonoids, selenium yeast chelate, brewer’s yeast, citrus bioflavonoids, whey,
Microwave digestion. For the closed vessel microwave as-

Yeast free form of selenium as L-selenomethionine. Also contains dicalcium phosphate,

sisted acid digestion of the yeast samples a MDS 2100 micro-
wave oven (CEM, NC, USA) with PTFE vessels was used. To
determine the total selenium content, 0.2 g yeast samples were
digested with 2 mL of HNO3 and 1 mL of H2O2. The

High selenium yeast, rich source of natural and organic L-selenomethionine

microwave conditions used to digest the samples are reported
elsewhere.32 After this treatment, the yeast digests were trans-

vegetable oil, glycerin, soybean oil, yellow bees wax and carob extract
ferred and made up to volume with de-ionized water in 10 mL
volumetric flasks.

ascorbic acid, orange peel extract and magnesium stearate

Enzymatic digestion for HPLC and GC. Enzymatic extrac-
Commercial selenium supplements, selenium species determined by HPLC-ICP-MS

tion was as reported earlier and is briefly summarized.34

Representative tablets were crushed using mortar and pestle,
after manually removing titanium dioxide coatings if present.
Approximately 0.2 g of the powdered sample was added to
0.02 g of the enzyme (Protease XIV) in a 15 mL centrifuge
gelatin, silica and magnesium stearate

tube. After the addition of 5 mL of distilled deionized water to

the centrifuge tube, the sample was shaken for 24 h at room
temperature using LABQUAKEs Shakers and Rotators
(Barnstead/Thermolyne Corporation, Dubuque, IA, USA).
USP water, hydrogenated

For the extraction of the liquid sample of live yeast 5 mL of

the slurry mixture and 0.02 g of the enzyme were mixed and
shaken for 24 h. After the extraction the sample was centri-
fuged (Fisher Scientific, Fair Lawn, NJ, USA) at 3000 g for
Ingredientsa

20 min. Then the supernatant was filtered through a 0.45 mm

polypropylene filter (Whatmanns). The filtered supernatant
was once again filtered through a 10 000 Dalton molecular
weight cutoff Ultrafree-4-centrifugal filter and tube (Millipore
Vitamin World—naturally inspired

Corporation, Bedford, MA, USA) by centrifugation for

50 min. The clear filtrate (900 mL) was mixed with 100 mL of
Solaray Dietary Supplement

concentrated HFBA before HPLC-ICP-MS analysis.

Selenomax (Nutrition 21)

on Se-specific peak areas.

Schiff’s Antioxidant Plus
Vitamin E 400 IU with

Derivatization of standard selenoamino acids for GC-AED.

Bio-Active Selenium/

Selenoamino acid standards were derivatized with ethylchloro-

formate (ECF) using procedures reported by a number of
Commercial
brand name

workers.37–41 Stock solutions of the amino acids were prepared

Sundown

selenium

by dissolving 10 mg in 1 mL of 0.1 M HCl to make 10 000 mg

Table 2

mL 1 standard solutions. Then 1000 mg mL 1 derivatized

standard solutions were prepared as follows: 100 mL of the
a

940 | J. Anal. At. Spectrom., 2007, 22, 938–946 This journal is

c The Royal Society of Chemistry 2007
10 000 mg mL 1 selenoamino acid stock solution was treated ment tablets employed in the ‘Clark’ trials13 and stored
with 1 mL of water–ethanol–pyridine mixture (60 : 32 : 8 by subsequently for a number of years at ambient temperature.32
volume), 50 mL of ECF was added and the mixture shaken In order to gain a perspective as to the presence of the three
until gas evolution ceased. Then 1.0 mL of chloroform con- primary organoselenium species present in commercial sele-
taining 1% ECF was added, and the derivatives were extracted nized yeasts and supplements, four different brands of sele-
into the organic phase by shaking for 30 min, which was nium nutritional supplements along with two non-yeast
evaporated to dryness. The residue was redissolved in 1.0 mL supplements were obtained locally, and showed substantially
of chloroform and 1 mL was injected. different speciation content (Table 2). The reported distribu-
Yeast samples were derivatized with ethylchloroformate and tion numbers are percentage concentrations, relative to the
analyzed by GC-AED. The procedure for the amino acids was total selenium response to the eluting selenium compounds,
scaled up for their determination in the yeast sample. To a 2 including selenate. With the exception of the ‘‘Sundown’’
mL portion of the extract was added 5 mL of a water–ethanol– material, summed percent data for designated eluted selenium
pyridine mixture (60 : 32 : 8 by volume) followed by 1 mL of species were in the 96–99% range of total eluted selenium
ethyl-chloroformate, the sample was shaken and any evolved signal over the elution time measured. The exception for the
gas vented. Ethylated derivatives were extracted into 3 mL ‘‘Sundown’’ material suggests the presence there of ca. 15% of
chloroform containing 1% ECF by shaking for 30 min. The undefined eluted selenium species.
chloroform layer was separated, solvent was removed with a The data indicate that supplement labeling should be ap-
flow of nitrogen and the residue was redissolved in 100 mL proached with caution. The supplement ‘‘Vitamin World’’
chloroform or hexane. One mL was injected. indicating selenomethionine content contained inorganic sele-
nium (selenate) and ‘‘Bio-Active Selenium/Solaray Dietary
Thermal treatment of selenized yeast. Selenized yeast powder Supplement’’ indicating a natural source of bio-available
(Selenopreciset) was heated in a closed vial at 100 1C and 200 selenium also contained selenate as the major selenium species.
mg aliquots were analyzed after 1, 3 and 7 days. A commercial The proportion of selenomethionine (taken together with
bakers yeast, spiked with selenomethionine, was also heated at Selenomethionine Se-oxide) was higher than 70% in the
100 1C for 24 h, as was a selenomethonine-cysteine sample. supplements with brand names ‘‘Sundown’’, ‘‘Selenomax’’
Samples were subjected to enzymatic digestion, and selenium and ‘‘Schiff’s Antioxidant Plus’’. The low level of seleno-
speciation profiles were obtained by HPLC-ICP-MS and methionine Se-oxide and absence of S-(methylseleno)cysteine
GC-AED. in the ‘‘Sundown’’ supplement that also contained the redu-
cing agent ascorbic acid (vitamin C) is noteworthy. In the
supplements that also contained vitamin E (E Vitamin 400 IU
Results and discussion
with selenium), SeMet (23%), selenate (35%), S-(methyl-
Treatment of selenized yeast with proteolytic enzymes, fol- seleno)cysteine (7%) and selenomethionine Se-(34%) were
lowed by extraction, releases more than 90% of total selenium all present.
present as selenite, selenate and selenamino acids or their In view of the studies described later concerning the chan-
derivatives. Perfluoro acid ion pair reversed phase HPLC- ging ratios of the three organoselenium species with time and
ICP-MS affords good separation of inorganic selenium species temperature, the wide variation among the proportions in
from selenoamino acids.32–34 Except for cases wherein formu- these supplements is not unexpected. The variations among
lated supplements (yeast or yeast-free) incorporate inorganic the selenium species present in selenized yeast products stored
selenium, the major organoselenium species identified in such at room temperature for several years, and also of recently
studies was selenomethionine (SeMet).1,27,29,42 Other seleno- manufactured yeast products, suggested that speciation
amino acids, such as selenocystine, Se-lanthionine, Se- changes have occurred during storage or during yeast supple-
cystathionine, Se-methylselenocysteine, were observed in sele- ment manufacture.32 In order to test this hypothesis and to
nized yeast at levels o1% of SeMet, but at that time a number determine whether any previously unidentified organosele-
of minor selenium species remained unidentified.42 Qualitative nium species were produced when selenized yeast was heated
identification of these relied in part upon retention time at an elevated temperature (100 1C), samples were heated
matching with available or synthesized standards, but in some before characterization by HPLC-ICP-MS and GC-AED.
cases direct confirmatory HPLC mass spectral data was No attempt was made in this study to conform to any
obtained. The identity of two of these compounds in selenized pharmaceutically accepted procedures for ‘accelerated stabi-
yeast, selenomethionine Se-oxide hydrate20,32,33,36 and the lity studies’ such as would clearly be required for a full
Se–S bonded amino acid S-(methylseleno)cysteine has now characterization of selenized yeast from a pharmaceutical
been confirmed.32,33 The presence of the latter was also con- standpoint.
firmed by ethylation derivatization and capillary GC with Speciation experiments were first carried out to determine if
selenium specific atomic emission detection (GC-AED)32 and S-(methylseleno)cysteine and selenomethionine Se-oxide were
by 77Se NMR.33 present in a typical selenized yeast slurry (yeast cream) that is
It has become clear that these two compounds comprise the used at the beginning of the yeast tablet manufacturing
primary organoselenium compounds released upon proteo- process. This slurry sample (a liquid sample of live Se-yeast)
lysis other than SeMet, and that they increase in relative and the tablets manufactured from it (Selenolife yeast tablets)
proportions to it as selenized yeast ages. This was initially were subjected to standard enzymatic extraction, selenium
indicated by study of archived samples of yeast and supple- species being determined by HPLC-ICP-MS and GC-AED.

This journal is
c The Royal Society of Chemistry 2007 J. Anal. At. Spectrom., 2007, 22, 938–946 | 941
Fig. 1 GC-AED chromatograms of the derivatized enzymatic extracts of (A) selenium yeast cream slurry and (B) Selenolife yeast tablet prepared
from the yeast slurry.

A major advantage of GC-AED is the ability to obtain selenomethionine Se-oxide (observed only in the HPLC chro-
simultaneously a number of element specific chromatograms matogram as discussed earlier) retain the same overall char-
for different elements. We earlier reported comparative C, S acteristics as observed previously and no other identifiable
and Se GC-AED chromatograms for ethylchloroformate-de-
rivatized Clark yeast and established the retention character-
istics for S-(methylseleno)cysteine by referencing both sulfur
and selenium chromatograms.32 Fig. 1 shows a comparison of
selenium specific gas chromatograms for derivatives of the
yeast slurry and Selenolife yeast tablets prepared from it. A
small peak corresponding to S-(methylseleno)cysteine appears
in the chromatogram of the yeast tablet but is absent in that of
the yeast cream. Selenomethionine Se-oxide is not observed in
the chromatogram since it is derivatized to the same product
as selenomethionine in the ethylation process.
By comparison, the HPLC-ICP-MS chromatograms for the
selenium yeast cream slurry sample and the derived Selenolife
yeast tablet are shown in Fig. 2. In the yeast cream chromato-
gram, a small peak corresponding to selenomethionine selen-
oxide (1) is observed but S-(methylseleno)cysteine is absent. In
the derived tablet, S-(methylseleno)cysteine (2) appears and it
is noteworthy that selenomethionine Se-oxide hydrate is pre-
sent at a higher level than in the yeast slurry. The two
chromatograms correspond to different overall species levels
due to the different analyte quantities sampled.
The chromatographic results for the thermal treatment of Fig. 2 HPLC-ICP-MS chromatograms (Elan 5000) of the enzymatic
SelenoPreciset at 100 1C are shown in Fig. 3 (HPLC-ICP-MS) extracts of (A) selenium yeast cream slurry and (B) Selenolife yeast
and Fig. 4 (GC-AED). Both HPLC and GC chromatographic tablet prepared from the yeast slurry. (1) selenomethionine Se-oxide,
patterns for selenomethionine, S-(methylseleno)cysteine and (2) S-(methylseleno)cysteine, (3) selenomethionine.

942 | J. Anal. At. Spectrom., 2007, 22, 938–946 This journal is

c The Royal Society of Chemistry 2007
Fig. 3 HPLC-ICP-MS chromatograms (Elan DRC-e) of the enzy- Fig. 4 GC-AED temperature programmed chromatograms of the
matic extracts of SelenoPreciset yeast stored at (A) 4 1C, and heated enzymatic extracts of SelenoPreciset yeast stored at (A) 4 1C, and
(B) 100 1C for 1 day; (C) 100 1C for 3 days and (D) 100 1C for 1 week. heated (B) 100 1C for 1 day; (C) 100 1C for 3 days and (D) 100 1C for 1
(1) selenomethionine Se-oxide, (2) S-(methylseleno)cysteine, (3) week. (2) Selenomethionine (including selenomethionine Se-oxide) (3)
selenomethionine. S-(methylseleno)cysteine.

species are formed at appreciable levels. In Fig. 3, it is noted the former, notably in the ‘unidentified’ category, presumably
that a small peak (not identified) is present on the tail of peak 1 arise from derivatization processes. These observations parallel
(selenomethionine Se-oxide), which introduces uncertainty the speciation observed earlier for archived selenized yeast from
into its quantification. The successive chromatograms (A, the Clark trials,13,32 lending some validity to the application of
4 1C storage), (B–D, 1, 3 and 7 days at 100 1C) show the briefer more elevated temperatures in stability studies.
overall trend that total selenomethionine content diminished Selenized yeast samples are subjected to elevated temperatures
over the period while the S-(methylseleno)cysteine percentage during manufacture. To remove excess free selenium, yeast is
peak area (% of the total selenium eluted from column) washed 3 to 4 times at 90–95 1C, with each wash cycle lasting
increased during the heating period. The relative amount of about 30 min. Next a process called drum drying takes place at
selenomethionine Se-oxide present also increased. The total 150–175 1C for 5 s to remove moisture from the yeast. The
peak area corresponding to the three chromatographable present stability study indicates loss of selenomethionine and the
species decreased to approximately one half of the initial formation of S-(methylseleno)cysteine at such temperatures.
summed area after one week heating but at that time the In order to ascertain any changes in the sample’s selenium
proportions of S-(methylseleno)cysteine in particular and also content during the heating period, total selenium was measured
of selenomethionine Se-oxide were enhanced relative to resi- before and after heating by ICP-MS and ICP-OES. The values
dual selenomethionine. Table 3 shows percent distributions of obtained are shown in Table 5 and indicate that all initial
selenomethionine, S-(methylseleno)cysteine and selenomethio- selenium was still present after heating. There was no suggestion
nine Se-oxide based upon total eluted selenium signal. of the formation of volatile selenium compounds such as
Fig. 4 shows the corresponding GC-AED chromatogram dimethyldiselenide during the heat treatment, so it may be
and the quantitative data are given in Table 4. The ‘seleno- concluded that selenium is converted at least in part to larger
methionine’ peaks (3) correspond to the sum of selenomethio- species not amenable to direct selenoamino acid analysis.
nine and selenomethionine Se-oxide which form the same
Formation of S-(methylseleno)cysteine in selenized yeast
derivative species; the responses relative to S-(methylseleno)-
products
cysteine are thus larger than shown in the HPLC-ICP-MS
analysis. The GC-AED and HPLC-ICP-MS data show the The overall elemental composition of bakers yeast indicates
same trend with sample heating, but numerical differences from that sulfur (S) is the seventh most abundant element after

Table 3 Peak area percent of selenium compounds in enzymatic hydrosylates of SelenoPreciset yeast powder by HPLC-ICP-MSa

Selenium species 4 1C storage 100 1C for 1 day 100 1C for 3 days 100 1C for 7 days
Selenomethionine Se-oxide 16.6  1.0 27.6  1.1 36.7  0.6 39.2  0.1
S-(methylseleno)cysteine 2.6  0.1 11.0  0.1 17.6  0.2 23.4  0.2
Selenomethionine 77.3  1.2 58.7  1.3 42.1  0.5 34.2  0.3
SUM 96.5 97.3 96.4 96.8
Unidentified 3.5 2.7 3.6 3.2
a
The % selenium distributions (mean  standard error; n = 3) are expressed in terms of total selenium content eluting from the column.

This journal is
c The Royal Society of Chemistry 2007 J. Anal. At. Spectrom., 2007, 22, 938–946 | 943
Table 4 Peak height percent of selenium compounds in enzymatic hydrosylates of SelenoPreciset yeast powder by GC-AEDa

Selenium species 4 1C storage 100 1C for 1 day 100 1C for 3 days 100 1C for 7 days
S-(methylseleno)cysteine 8.3  0.7 12.1  0.5 13.6  0.2 19.7  0.2
Selenomethionine 88.5  1.2 77.1  0.4 73.5  0.4 59.2  0.6
SUM 96.8 89.2 87.1 78.9
Unidentified 3.2 10.8 12.9 21.1
a
The % selenium distributions (mean  standard error; n = 3) are expressed in terms of total selenium content eluting from the column.

Table 5 Total selenium content (mg g 1) of SelenoPreciset yeast S-(methylseleno)cysteine,33 although the yeast heating experi-
powder before and after heating at 100 1C for 7 days ment did not indicate that any free dimethyldiselenide was
ICP-MS ICP-OES formed. Experiments were performed with pure standards,
(Elan DRC-e) (Optima 4300-DV) selenomethionine and L-cysteine to test this hypothesis. Solid
standards (selenomethionine 10 mg and L-cysteine 100 mg)
Before heating 1335  6 1370  24
Heated for 7 days 100 1C 1349  56 1395  65 were mixed with a few drops of deionized water and heated for
a 24 hours at 100 1C. For HPLC-ICP-MS analysis, the solid was
Concentrations expressed as mean  standard error, n = 15.
dissolved in 10 mL of deionized water, and the chromatogram
(Fig. 5) shows the formation of S-(methylseleno)cysteine and
selenomethionine Se-oxide hydrate. The large tailing peak for
carbon (C), oxygen (O), nitrogen (N), hydrogen (H), potas- selenomethionine is due to column overloading. A similar
sium (K) and phosphorus (P). Sulfur constitutes 0.36% of experiment was performed by spiking commercial bakers yeast
yeast dry matter.43 In the course of the growth cycle of
Saccharomyces cerevisiae yeast, selenite that is potentially
toxic and poorly bioavailable, is converted to bioactive species
such as selenomethionine which have better nutritional prop-
erties. The formation of S-(methylseleno)cysteine during the
yeast growth cycle is not apparent since the slurry selenium
yeast did not indicate its presence.
However, the formation of S-(methylseleno)cysteine might
be explained by degradation of selenomethionine to form
dimethyldiselenide within the yeast structure, that could in
turn react with cysteine and/or cystine (present in yeast) to
form the Se–S bonded species. Such a reaction taking place in
the solid phase would parallel that involved in the synthesis of

Fig. 6 HPLC-ICP-MS chromatograms (Elan 5000) of a commercial

Fig. 5 HPLC-ICP-MS chromatograms (Elan 5000) of Selenomethio- bakers yeast, (A) spiked with selenomethionine and left at room
nine and L-cysteine mixture after 24 h at 100 1C, upper: full scale, temperature for 24 h; (B) zoom in of (A); and (C) spiked with
lower: zoom. (1) selenomethionine Se-oxide, (2) S-(methylseleno)cys- selenomethionine and heated at 100 1C for 24 h. (1) Selenomethionine
teine, (3) selenomethionine. Se-oxide, (2) S-(methylseleno)cysteine, (3) selenomethionine.

944 | J. Anal. At. Spectrom., 2007, 22, 938–946 This journal is

c The Royal Society of Chemistry 2007
with selenomethionine. The chromatogram for the spiked support from PerkinElmer Instruments is gratefully acknowl-
yeast heated at 100 1C for 24 h (Fig. 6) showing a peak for edged. Samples and support from Philip Taylor, MD, NIH-
S-(methylseleno)cysteine. These preliminary experiments sug- NCI, Cypress Systems and LeSaffre Yeast Corporation-Red
gest that S-(methylseleno)cysteine forms slowly from seleno- Star Nutritional Yeast.
methionine through an intermediary such as dimethyl
diselenide, which may react with L-cysteine and/or cystine in
yeast. References
1 G. N. Schrauzer, J. Nutr., 2002, 132, 1653–1656.
2 L. A. Cohen, in Hormone-Related Tumors: Novel Approaches to
Conclusions Prevention and Treatment, New York Academy Sciences, New
York, 2002, 963, pp. 148–155.
Regulations governing the QA/QC protocols followed during 3 D. Behne and A. Kyriakopoulos, Annu. Rev. Nutr., 2001, 21, 453–
nutritional supplement preparation within the United States 473.
and elsewhere are less rigid than those for drug substances, as 4 L. H. Fostera and S. Sumar, Crit. Rev. Food Sci. Nutr., 1997, 37,
211–228.
is product labeling for the consumer. A case in point is that of
5 E. Dumont, F. Vanhaecke and R. Cornelis, Anal. Bioanal. Chem.,
‘accelerated stability’ studies including thermal stability gen- 2006, 385, 1304–1323.
erally performed for pharmaceutical ‘active’ species. It may be 6 E. Manteroatienza, M. G. Sotomayor, G. Shorposner, M. A.
pointed out that ICH (International Conference on Harmo- Fletcher, H. E. Sauberlich, R. S. Beach and M. K. Baum, Nutr.
Res, .1991, 11, 1237–1250.
nization of Technical Requirements for Registration of Phar- 7 M. P. Rayman, Lancet, 2000, 356, 233–241.
maceuticals for Human Use) stipulates protocols for stability 8 K. Kose, P. Dogan, Y. Kardas and R. Saraymen, Biol. Trace Elem.
studies including, for example, comparisons of testing between Res., 1996, 53, 51–56.
storage at 25 1C and 30 1C. The guidelines for ‘accelerated’, i.e. 9 C. Ip, J. Nutr., 1998, 128, 1845–1854.
10 A. Campa, G. Shor-Posner and M. Baum, J. Am. Diet. Assoc.,
elevated temperature, studies are also specific and it is empha- 2000, 100, 418–418.
sized that our experiments in no way follow such protocols. 11 C. Ip, D. J. Lisk and H. J. Thompson, Carcinogenesis, 1996, 17,
However, our observations showing close similarities of spe- 1979–1982.
12 J. W. Finley, C. D. Davis and Y. Feng, J. Nutr., 2000, 130, 2384–
ciation and quantitation between extended (many years) sto- 2389.
rage at ambient temperatures and brief heating at elevated 13 L. C Clark, G. F. Combs, B. W. Turnbull, E. H. Slate, D. K
temperatures are at least indicative of a consistent mechanism Chalker, J. Chow, L. S. Davis, R. A. Glover, G. F. Graham, E. G.
of speciation behavior. Gross, A. Krongrad, L. J. Lesher, H. K. Park, B. B. Sanders, C. L.
Smith and P. R. Taylor, J. Am. Med. Assoc., 1996, 276, 1957–1963.
Although progress on the designation of a ‘standard refer- 14 E. A. Klein, I. M. Thompson, S. M. Lippman, P. J. Goodman, D.
ence selenized yeast (for speciation validation)’ has been a goal Albanes, P. R. Taylor and C. Coltman, J. Urol., 2001, 166, 1311–
for some time, it is not yet in place for independent validation. 1315.
15 E. A. Klein, I. M. Thompson, S. M. Lippman, P. J. Goodman, D.
This remains an arduous task, since not only the major
Albanes, P. R. Taylor and C. Coltman, Cancer Epidemiol. Bio-
selenium species, but also minor (and potentially) variable markers Prev., 2002, 11, 1253S–1254S.
minor components, must also be considered. 16 K. L. Sutton, C. A. Ponce de Leon, K. L. Ackley, R. M. C. Sutton,
Of the different brands of selenium nutritional supplements A. M. Stalcup and J. A. Caruso, Analyst, 2000, 125, 281–286.
17 R. Lobinski, J. S. Edmonds, K. T. Suzuki and P. C. Uden, Pure
analyzed, only two brands had high concentrations of seleno- Appl. Chem., 2000, 72, 447–461.
methionine and two brands appeared to contain mostly in- 18 P. C. Uden, Anal. Bioanal. Chem., 2002, 373, 422–431.
organic selenium (selenate), suggesting that label claims of 19 J. A. Caruso and M. Montes-Bayon, Ecotoxicol. Environ. Saf.,
2003, 56, 148–163.
‘‘bioavailable selenium’’ or the presence of mainly seleno-
20 E. Larsen, M. Hansen, H. Paulin, S. Moesgaard, M. Reid and M.
methionine may often be inaccurate. Selenomethionine selen- Rayman, J. AOAC Int., 2004, 87, 225–232.
oxide hydrate appears to form as soon as selenized yeast is 21 H. Infante, R. Hearn and T. Catterick, Anal. Bioanal. Chem., 2005,
dried and tableted. The wide range of the selenomethionine to 382, 957–967.
22 L. Reyes, J. Encinar, J. Marchante-Gayon, J. Alonso and A. Sanz-
selenoxide ratios observed in this and other studies certainly Medel, J. Chromatogr., A, 2006, 1110, 108–116.
depends on storage time and conditions, notably exposure to 23 C. B’Hymer and J. A. Caruso, J. Chromatogr., A, 2006, 1114, 108–
air and elevated temperatures. S-(methylseleno)cysteine also 116.
appears to form soon after selenized yeast is made and appears 24 T. Lindemann and H. Hintelmann, Anal. Chem., 2002, 74, 4602–
4610.
at increasing ratios to selenomethionine in stored samples and 25 C. B’Hymer and J. A. Caruso, Can. J. Anal. Sci. Spectrosc., 2001,
in samples subjected to elevated temperatures for briefer 46, 136–144.
periods. Analytical methods may not distinguish between 26 W. R. Wolf, H. Zainal and B. Yager, Fresenius’ J. Anal. Chem.,
2001, 370, 286–290.
selenomethionine and the selenoxide hydrate and indeed
27 K. Wrobel, S. S. Kannamkumarath and J. A. Caruso, Anal.
quantitation may usefully combine content for both species Bioanal. Chem., 2003, 375, 133–138.
as they are probably utilized in the same manner physiologi- 28 J. L. Capelo, P. Ximenez-Embun, Y. Madrid-Albarran and C.
cally. Any studies on the clinical, nutritional or toxicological Camara, Anal. Chem., 2004, 76, 233–237.
29 Z. Mester, S. Willie, L. Yang, R. Sturgeon, J. A. Caruso, M.
properties of S-(methylseleno)cysteine remain to be initiated. Fernandez, P. Fodor, R. Goldschmidt, H. Infante, R. Lobinski, P.
Maxwell, S. MacSheeny, A. Polatajko, B. Sadi, A. Sanz-Medel, C.
Scriver, J. Szpunar, R. Wahlen and W. Wolf, Anal. Bioanal. Chem.,
Acknowledgements 2006, 385, 168–180.
30 H. Infante, G. O’Connor, M. Rayman, R. Wahlen, J. Entwsile, P.
This material is based upon work supported by the National Norris, R. Hearn and T. Catterick, J. Anal. At. Spectrom., 2004,
Science Foundation under Grant No. 0094568. Financial 19, 1529–1538.

This journal is
c The Royal Society of Chemistry 2007 J. Anal. At. Spectrom., 2007, 22, 938–946 | 945
 31 H. Infante, G. O’Connor, M. Rayman, R. Hearn and K. Cook, J. 37 P. Husek, J. Chromatogr., B: Biomed. Appl., 1998, 717, 57
Anal. At. Spectrom., 2006, 21, 1256–1263. 67.
32 P. C. Uden, H. T. Boakye, C. Kahakachchi, R. Hafezi, P. Nolibos, 38 C. Haberhauer-Troyer, G. Alvarez-Liamas, E. Zitting, P. Rodri-
E. Block, S. Johnson and J. F. Tyson, J. Anal. At. Spectrom., 2004, guez-Gonzalez, E. Resenberg and A. Sanz-Medel, J. Chromatogr.,
19, 65–73. A, 2003, 1015, 1.
33 E. Block, R. S. Glass, N. E. Jacobsen, S. Johnson, C. Kahakach- 39 B. Iscioglu and E. Henden, Anal. Chim. Acta, 2004, 505, 101
chi, R. Kaminski, A. Skowronska, H. T. Boakye, J. F. Tyson and 106.
P. C. Uden, J. Agric. Food Chem., 2004, 52, 3761–3771. 40 Y. Fiamegos, C. G. Nanos and C. D. Stalikas, J. Chromatogr., B:
34 M. Kotrebai, J. F. Tyson, E. Block and P. C. Uden, J. Chroma- Biomed. Appl., 2004, 813, 89–94.
togr., A, 2000, 866, 51–63. 41 Y. Qiu, M. Su, Y. Liu, M. Chen, J. Gu, J. Zhang and W. Jia, Anal.
35 E. Block, M. Birringer, W. Q. Jiang, T. Nakahodo, H. J. Thomp- Chim. Acta, 2007, 583, 277–283.
son, P. J. Toscano, H. Uzar, X. Zhang and Z. J. Zhu, J. Agric. 42 M. Kotrebai, M. Birringer, J. F. Tyson, E. Block and P. C. Uden,
Food Chem., 2001, 49, 458–470. Analyst, 2000, 125, 71–78.
36 B. Gammelgaard, C. Cornett, J. Olsen, L. Bendahl and S. H. 43 M. Korhola, A. Vainio and K. Edelmann, Ann. Clin. Res., 1986,
Hansen, Talanta, 2003, 59, 1165–1171. 18, 65–68.

946 | J. Anal. At. Spectrom., 2007, 22, 938–946 This journal is

c The Royal Society of Chemistry 2007

View publication stats