MBAA TQ vol. 52, no. 3 • 2015 • pp.

156–164

Enzymes in Brewing
Mark Sammartino
MBAA Technical Director

The subject of enzymes in brewing is often constrained to helpful under specific circumstances. Knowing the whys and
the mash vessel. It is the objective of this article to expand this the hows can add to your brewing toolbox.
understanding to include the malting process and how it relates
to beer, as well as to bring in the commercially available en- Malting
zymes for consideration. By appreciating all of these areas, it
is easier to understand the magnitude of this complexity. To Malting, the process of converting barley into malt, is as old
consider one area without considering all others leaves gaps in as the history of beer. Malting and brewing are based upon the
understanding and thus opportunities for issues to arise. Thus, presence of carbohydrates and plant protein in the barley ker-
this review covers enzymes from malting through to commer- nels, the ability of malting to create the natural carbohydrate-
cial enzymes. Working in the field of commercial enzymes has reducing and protein-reducing enzymes, and the production of
taught me a great deal about the world of naturally occurring both simple sugars from the plant starch and an acceptable
enzymes. Appreciating the full scope of enzyme activity is protein spectrum from the plant proteins by utilizing these
important to understanding how to influence process change. enzymes. Malt is a product of the plant germination process
The approach of this article is to discuss how the enzymes within a set of controllable parameters required to achieve
relate to specific aspects of the process. It is up to the reader to consistent results for the brewer.
integrate this complexity into the other related processes in the
brewery. Malting Process
In this review, we look at enzymes from malt to commercial The malting process consists of three phases: steeping, ger-
sources. The process of malting converts the hard grains of mination, and kilning. After the plant growth process is acti-
choice to softened grains with available extract and the needed vated through steeping and germination, the green malt is kilned
chemistry to convert this extract into a useful basis for yeast to (dried and roasted) to stop the growth of the barley seedlings
make beer. The process of selecting grain, malting techniques, and to develop malt flavor and color.
and mashing processes utilizes a full understanding of the en-
zymes needed in your process, especially if you run on the Carbohydrates
edges of the process capabilities. We will begin with a discus- Common carbohydrates that we are familiar with are sugars,
sion of enzymes and their activation in malting. We should starches, and cellulose. The basic building block of all carbo-
note that the malting process utilizes these naturally occurring hydrates is a simple sugar, glucose. All carbohydrate chains
enzymes within the grain to free up the extract for our use. are chains of glucose bonded together. Starches are composed
Within the context to follow, I will be referring to the process of the fermentable α-glucose, and cellulose is composed of the
of malting barley. It should be noted that many other grains nonfermentable β-glucose. Unless specifically referenced be-
used in brewing can also be malted in a similar manner, thus yond this point, all glucose is to be considered α-glucose.
making extract more available to the brewing process. In addi- Plant starch is the stored energy of the plant that is broken
tion to freeing up the extract, the malting process gives the down by plant enzymes when the plant requires energy. In
brewer the enzyme spectrum required to make the product simple terms, the starch found in the barley kernel is of two
fermentable, thus the basis of beer. We will then move through types: amylose, containing linear chains of glucose molecules,
the enzymatic processes in mashing. Because this is a highly and amylopectin, which is a multibranched structure of glu-
integrated process of time and temperature relationships, the cose chains. Amylose and amylopectin differ in how the glu-
effects of one change can have profound impacts on other as- cose units are linked together. Both have 1-4 glucosidic bonds
pects of your process. In reviewing the enzymes in this proc- within their linear structure holding the glucose molecules
ess, we can begin to appreciate the range of these impacts. together, but amylopectin also has 1-6 glucosidic bonds at
Lastly, with today’s focus on being different or looking at the branch points, creating its complex structure. This difference
very edges of what a process might be capable of, we review is important to the understanding of how the breakdown of
the world of commercial enzymes available in brewing. Other the glucose chains leads to a difference in extract and ferment-
factors besides stretching the limits of a process may influence ability.
your interest in the use of commercial enzymes. Problems in
malting caused by grain issues such as preharvest sprout, or Proteins
variations in growing conditions affecting protein or carbohy- Proteins are the basic raw materials in tissue and cell build-
drate levels, can be effectively dealt with by using some of ing essential for life. They are part of the matrix that holds the
these products. The intent of this discussion is not to suggest cell together. Additionally, they supply the functional mole-
that you use these products but to explain how they can be cules for sustaining life within the kernel. Like carbohydrates,
proteins are complex chains of the simplest protein com-
pounds, amino acids. Unlike carbohydrates, which all are built
http://dx.doi.org/10.1094/TQ-52-3-0818-01 from a single compound (glucose), proteins are built from 25
© 2015 Master Brewers Association of the Americas different amino acids. Proteins are essential to brewing to pro-

156
Enzymes in Brewing MBAA TQ vol. 52, no. 3 • 2015 157

vide the amino acids for yeast nutrition, as well as to provide moval of heat and CO2 generated by the now respiring plant.
flavor, mouthfeel, foam, and color. Larger proteins can also Excess of either will damage the growth process and create a
lead to turbidity and sediments. whole set of other issues such as premature yeast flocculation
and poor or variable modification, along with the potential for
Enzymes kernel death. While the barley is in the beds, turning machines
Enzymes are polymers of amino acids and for all intents are are sent regularly through the growing barley to keep the root-
also proteins. They are proteins with a function. They are cata- lets from matting as well as to apply water as if to water a gar-
lysts that regulate the speed of chemical reactions involved den. What the maltster is attempting to do is to control the
within a living organism’s metabolism, without permanently growth of the emerging plant while allowing the development
changing the reactions. The enzymes of our particular interest of the enzymes needed for digestion. This digestion is focused
are the ones responsible for digesting the starch and protein on making the materials available from this stored matrix for
chains described earlier. The prominent malting and brewing plant growth. A sequence of enzymes is released by the germi-
enzymes are, but are not limited to, amylases (carbohydrate nating plant, principally proteases to begin with and some
enzymes), proteases (protein enzymes), peptidases (which β-glucanases and xylanases, all focused on digesting the ma-
break down protein pieces into amino acids), and β-glucanases trix around the starch granules stored in the kernel. Eventually,
and xylanases (cellulose enzymes). There are several addi- α- and β-amylases are released to begin the digestion of the
tional enzyme classes that have little to do with the processes starch to provide energy for the growing plant. The key to the
described here and will not be considered in this part of the maltster’s objective is to control and limit the amount of starch
discussion. We will discuss some of these enzymes later as we digestion while maximizing the amount of digestion of the
review mashing and commercial enzymes. In the review of matrix of protein, β-glucan, and xylan holding the starch gran-
malting we will not cover the actual process of malting. Our ules together. This is referred to as malt modification.
focus will remain on the enzymatic processes. Understand that
malting is a complex process of air flows, temperatures, bed Kilning
depth, and moisture control designed to facilitate the con- After the controlled growth, the green malt is transferred to
trolled growth of a barley plant. Our focus here is to look at a kiln, where heated dry air is applied to dry the grain in such a
the steps in malting to see how the enzymes become available manner as to stop growth while preserving as much of the en-
for the brewer. zyme activity as possible. In the later stages of drying, the heat
is increased and the flavor and color development is enhanced.
Steeping Various methods and times of heat application are used to facil-
As stated earlier, the malting process consists of three itate broad ranges of color and flavor used by the brewer. How
phases. The first phase, steeping, is a series of full immersions the drying cycle is approached will have a great deal to do
of large batches of barley in constantly aerated temperature- with what level of enzyme activity remains in the grain. Obvi-
controlled water, spaced by drain cycles, again with heavy ously, the more heat applied, the less remaining enzyme activ-
airflow to prevent heat and CO2 buildup. The purpose of this ity. What is less obvious is that a higher temperature in the pres-
process is to increase the internal moisture of the barley kernel ence of moisture has a greater effect in destroying the enzymes.
to above 40% while not drowning the early growing plant or The challenge is to balance flavor versus functionality.
damaging it by heat or CO2. This increased moisture signals
the internal mechanisms in the plant to begin growth, and the Self-Digestion
moisture provides a medium for the transfer of the enzymes During the growth of the barley kernel, which is occurring
inside the kernel to begin self-digestion. Thus, the growing in the late stages of steeping and throughout germination, the
begins. This process can take upward of 48 h, depending upon objective of the growing plant is to self-digest its kernel to
the quality and type of the barley, the temperatures of the vari- provide energy and nutrients for growing. This self-digestion
ous steep cycles, and the temperature of the grain and the air occurs through the growing plant’s creation of a series of en-
applied. Additionally, barley fresh from harvest and sometimes zymes designed to break down the kernel’s starches and pro-
barley that has been stored a long time can exhibit a delayed teins. The maltster’s objective is to control and harvest this
response to this hydration. This delayed response is known as process. In simple terms, the amylase activity for starch break-
dormancy (a natural process set up by the organism as a pro- down is minimized to prevent fermentable extract loss for the
tective response to keep the seed from trying to germinate under brewer, while the β-glucanase and xylanase (cellulose break-
harsh conditions). Care in hydration of the kernels needs to be down) and protease/peptidase (protein breakdown) activity is
exercised when dealing with this phenomenon. The biology of maximized to soften the kernel and expose the starch for ex-
the kernel sets the process in motion. As the plant begins to traction in mashing. Obviously, a great deal of expertise and
sprout it requires energy to grow. The energy comes from the control goes into this process.
starch stored in the seed. To get to this starch, the plant creates
a series of enzymes designed to digest the supporting materials Analytical Outcome
holding the starch, and later the starch as well. Coming out of the malting process, we have now modified
the barley to have a level of carbohydrate-digesting enzymes,
Germination expressed as α-amylase content and diastatic power, within the
The second phase of malting is germination. The steeped maltster’s analytical profile of their finished product. Addition-
barley is transferred into large growing beds and leveled to ally, we will see an increase in soluble protein and, more im-
provide equal exposure to the environmental conditions ap- portantly, an increase in the soluble-to-total protein ratio as a
plied by the maltster for control. The germination process is a direct expression of the barley modification. Lastly, we will
4–5 day process that facilities the growth of the barley seed- see a reduction in the β-glucan content and the wort viscosity
ling. Large amounts of moisture-saturated and temperature- as an expression of cellulose digestion. Obviously, if your in-
controlled air are cycled through the beds to facilitate the re- dependent focus was only to consider extract, you would be
158 MBAA TQ vol. 52, no. 3 • 2015 Enzymes in Brewing

looking to maximize each of these changes. But as we all deal with maltotriose need to be considered. A good basis to
know, what is important is balance. This balance provides for start from is to consider not more than 50% of the maltotriose
the needed economies of our processes while facilitating con- as fermentable. We should also note that there are other sugars
trol and consistency in creating the variety of products we that yeast will ferment, namely, fructose and sucrose. These
enjoy. There are some informative articles previously pub- sugars do not originate in malt; they may enter your process
lished in the TQ that reflect the importance of balance in all of through the addition of other fermentable materials such as
these processes (2–4). Additionally, they give an outstanding honey, fruits, and table sugar.
picture of the outcomes of each of these processes.
Free Amino Nitrogen
Mashing Earlier, we introduced a discussion on digested protein and
its importance to yeast growth. In barley the protein content
Process and Objectives can vary from 7% to about 15% by weight. About 30% of this
In this part of the discussion, we will consider the mashing protein will pass through into the finished beer, primarily as
profile of a single-malt mash. Obviously, there are many dif- breakdown products of the initial complex protein structures
ferent methods to mash and additional materials that can be inside the barley kernel. Protein breakdown products are amino
added. Each in itself can add complexity to our understanding. acids (single monomers of these nitrogenous compounds) and
However, if you understand the basics, much of this is trans- peptides (short chains of amino acids). As previously noted,
ferrable to other methods and materials, as long as you under- proteins and high-molecular-weight peptides are important to
stand the specific impacts of each variation. For instance, de- beer foam and body, but they can also result in haze. Smaller
coction brewing is a wonderful process to build flavor. But peptides are basically neutral with no discernible effect. How-
one needs to note that each pump-over and boil results in a ever, amino acids are vital yeast nutrients often referred to as
loss of enzymatic power. Additionally, during pump-back, if amino nitrogen or free amino nitrogen (FAN). The amount
you do not use extreme care you also will lose enzymatic needed is a direct reflection upon the amount of yeast growth
power owing to thermal degradation within the proximity of that occurs in your process. When digesting proteins to amino
the hot liquid until blended. It is not that these processes acids, we need to consider the classification of these amino
cannot be used; one just needs to understand the constraints. acids. The 25 different amino acids can be broken into four
Only 15–25% of barley malt is soluble in water. Mashing is categories (1,8). The first category is absolutely essential for
the process during which a much greater fraction of the malt yeast growth, the second is important, the third is less im-
is made soluble. During the process of mashing, a predeter- portant and used only if needed, and the fourth is not used in
mined fraction (based upon the product definition by the anaerobic processes. Therefore, in simple terms, whatever
brewer) of the solubilized compounds is converted from starch amount you start with for FAN, you need to ensure that you
to yeast-fermentable sugars. The end product of mashing is will not use it all. Expect at least 25–30% and preferably
wort. Wort contains fermentable and nonfermentable carbo- closer to 50% to pass through as unused and unwanted FAN. A
hydrates, proteins, amino acids, and other organic and inor- good starting number would be 1 ppm of wort FAN needed for
ganic compounds that are extracted from the malt. The pri- every 1 million yeast cells you grow. Therefore, a safe position
mary goal of mashing is to produce as much high-quality would be a starting FAN of 200+ ppm if you were going to
extract as possible from the grain, with some desired fraction grow 100 million cells, thus leaving you with around 100 ppm.
of this extract being converted to fermentable sugars, which As your yeast progresses through the second and into the third
will eventually lead to making alcohol in the presence of yeast categories of amino acids, you will begin to see other issues
during fermentation. develop in your flavor profiles. These issues are typically asso-
ciated with excessive sulfur products being generated as well
Amylose and Amylopectin as a propensity for sluggish or incomplete fermentations.
Earlier we discussed starches as a mix of amylose and amy-
lopectin. Barley malt contains about 60% starch. Approxi- Enzyme Overview
mately 75–80% of this starch is amylopectin, the complex Moving on to enzymes, we also previously noted that they
highly branched (1-6 glucosidic bonds) polysaccharide that are high-molecular-weight complex proteins that act as cata-
can contain over 100,000 glucose monomers, with 20–25 glu- lysts for organic reactions. Typically, a catalyst enables a reac-
cose unit strands (1-4 glucosidic bonds) between the branch tion to occur or considerably accelerates the rate at which it
points. The remaining 20–25% of the malt starch is amylose, occurs. Enzymes are extremely specific, catalyzing only one
linear strands (1-4 glucosidic bonds) of 1,000 or more glucose chemical reaction. There is a whole science around how en-
monomers in sequence. zymes work, beginning with a naming convention. In simple
terms, the substrate affected by the enzyme becomes the basis
Fermentable Sugars of Starch Digestion of the enzyme’s name, with an “ase” attached to the end. Thus,
When we discuss the digestion of these starches we break proteases affect protein, β-glucanases affect β-glucans, and so
the resulting molecules into two fractions: those that are fer- on. The science of enzymes theorizes a lock-and-key concept
mentable by yeast and those that are considered nonferment- of attachment and reaction, feedback and regulation, environ-
able. The fermentable sugars created from barley malt via en- mental effects, and so on. For our discussion, we are really
zyme digestion are glucose (single monomer), maltose (dimer only interested in the effects of time, temperature, and pH.
of glucose joined by 1-4 glucosidic bonds), and maltotriose Enzymes tend to have high specificity to the effects of temper-
(trisaccharide of glucose that is joined by 1-4 glucosidic bonds), ature and pH. Additionally, as the limits of these are reached,
and that is it! One should note that not all yeast will ferment time becomes the major driving factor in the remaining activ-
maltotriose, or at least not all of it. So when determining your ity and net performance of the enzyme. For a greater under-
fermentability from a sugar profile, your yeast’s capabilities to standing, please see P. R. Mathewson’s book Enzymes (6).
Enzymes in Brewing MBAA TQ vol. 52, no. 3 • 2015 159

pH concern about β-glucan and FAN content. However, in the real

world we know this is not always the case. It should be noted
As complex protein structures, enzymes possess clearly de-
that after kilning much of the β-glucanase and protease capa-
fined polarized (charged) areas. This polarization determines
bilities in the malt have been eliminated. Therefore, attempting
much of their shape and can be a major part of how they func-
to further reduce β-glucans or increase FAN by using extended
tion. The ion characteristics of the solution they reside in will
protein rest holds around 120°F will have only a small effect
have a major effect upon their polarization, therefore affecting
(perhaps about a 10% shift). What is critical to note is the feru-
their ability to hold a shape and function. In basic terms, the
lic acid esterase activity. Extended holds can result in an in-
pH of a solution expresses the balance of negative and positive
crease of upward of 100% in ferulic acid.
ions. In looking at mashing, we normally look at a pH range of
roughly 5.0–5.7. For the most part, the natural enzymes we are Ferulic Acid
concerned about in mashing function well within this range.
Ferulic acid by itself is not of great concern (Fig. 1). How-
So for the sake of this discussion, we no longer need to be
ever, if your wort possesses high levels of ferulic acid and your
concerned. There are some specialty mashing processes that
process allows for long hot temperature holds after the kettle
take us outside this normal pH range, resulting in opportunities
boil, you may experience a flavor impact on your beers. As
for greater discussion.
ferulic acid sits at a high temperature, it will autoconvert to
Time and Temperature 4-vinyl guaiacol (4VG). 4VG is highly volatile and possesses a
strong clove aroma and flavor. For some beers this flavor is
We will consider in greater detail what we are trying to considered normal, but for many beers it is considered a de-
achieve in the mashing process. In mashing our objective is to fect. Therefore, if you are short on FAN or long on β-glucans,
influence the enzymes created in malting to digest the barley there are commercially available solutions to address these
malt extract in such a manner as to make the wort for the beer issues. Otherwise, attention to the process becomes vital to
type we are trying to create. To accomplish this objective we maintaining a 4VG-free product.
use the variables of temperature and time to establish that in-
fluence (again, as required for the beer we are making). In Conversion Temperatures
simple terms, all enzymes have a temperature of greatest activ- Moving further along in the mashing process, we normally
ity. Unfortunately, this high rate of activity normally occurs see the use of one or two temperature holds to influence the
close to the temperature at which the enzyme becomes deac- optimal activity of the starch-converting enzymes. This step is
tivated. Therefore, time also plays a role. The closer we get to typically known as the conversion hold. Our objective is to
the optimum activity temperature for the enzyme, the shorter create a blend of fermentable and nonfermentable sugars for
the life span is for the enzyme. Thus, balance is needed, as the beer type we are making. The next four of the seven en-
well as an understanding of which enzymes are critical at each zyme groups now become of interest. We can break these
stage in mashing. down into two basic categories. The first three are starch-con-
Seven Critical Enzymes verting enzymes: limit dextrinase (optimal 131–140°F, deac-
tivated at 145°F), β-amylase (optimal 140–149°F, deactivated
With this background we can now move into the mashing at 158°F), and α-amylase (optimal 162–167°F, deactivated at
process. As stated, this process is a sequence of times and tem- 176°F). The last enzyme of interest at these temperatures is
peratures designed to influence the enzymes needed for the β-glucan solubilase (optimal 140–158°F, deactivated 163°F),
beer being made. There are many enzymes that are active which by reference to its name solubilizes β-glucans.
within a mashing process. Many of these enzymes act on sub-
strates that are either in such a large quantity that their very β-Amylase
level defines a process, or the substrates are so small they can β-Amylase is the first starch-digesting enzyme we will con-
be considered de minimis. For the sake of this discussion we sider. Typically, the ratio of fermentable to nonfermentable
will consider only the seven enzymes that are really important sugars in a mash is set by the conversion temperature. This
to control for our processes. temperature should be chosen to express the blend of the two
The First Three Enzymes
Normally, mashing begins at a certain temperature for one
of two reasons: it is designed to fit our equipment, or it is be-
lieved that there is some benefit. In any event, it is important to
fully wet the grains and begin the process of freeing up the
enzymes into solution. Three of the seven enzymes concern us
at mash in. These three enzymes have optimal activity tem-
peratures around where most brewers begin the mashing proc-
ess, 100–120°F. These enzymes are β-glucanase (optimal 113–
122°F, deactivated at 140°F), ferulic acid esterase (optimal
100–113°F, deactivated at 149°F), and proteases (optimal 113–
130°F, deactivated at 158–167°F). As you can see, some opti-
mal activities overlap in temperature, which results in multiple
activities occurring at the same time at specific temperatures.
Protein Rest Benefits
If the maltster began with great barley and has done their
job correctly, we have well-modified malt and therefore little Figure 1. Arabinoxylan and ferulic acid. Image courtesy of DuPont.
160 MBAA TQ vol. 52, no. 3 • 2015 Enzymes in Brewing

principal starch enzyme activities: α- and β-amylase. The you are now allowing for the incubation of a large amount of
β-amylase digests strands of starch and dextrins, breaking naturally occurring microflora that come with the grain. This is
them down to maltose molecules and undigestible fragments. nothing to worry about, unless you hold for extended periods
As it works, it can only break linear 1-4 glucosidic bonds and of time, as with long holds to create highly fermentable beers.
therefore cannot digest amylopectin owing to the branch Holds that exceed 3–4 hours can be problematic owing to the
points. In addition, it can only address the starch chain from microflora contributing vegetable, especially radish-type, fla-
one end, the nonreducing end. Therefore, its activity is one vor notes to your beers, so be cautious. In addition, the longer
molecule at a time, working along the chain until it reaches a you hold, the more damage your filter bed will take. Therefore,
branch point, where it must stop. with longer conversion holds consider coarser grinds and slow
agitation speeds.
α-Amylase
For β-amylase to be effective, more nonreducing ends must Our Last Two Critical Enzymes
be made available for its attack. This is where α-amylase To finish our discussion, we need to look at the last two of
comes into play. α-Amylase also only breaks 1-4 glucosidic the seven enzymes we discussed: limit dextrinase and β-glucan
bonds, but it can attack the starch molecule at any point. This solubilase. Keeping with the starch discussion, we will look at
random attack results in a large array of smaller dextrins and a the limit dextrinase first. This enzyme possesses the ability to
small amount of fermentable sugars. break the 1-6 glucosidic bonds in the branch points of amylo-
pectin. As with most enzymes of its type, it is extremely tem-
Sense of Balance perature sensitive and is deactivated rapidly at or above its
It would be great if it were that easy. But note, the optimal deactivation temperature of 145°F. Additionally, there is little
activity temperature for the β-amylase is much lower than for of the enzyme to begin with in barley and much less, after
the α-amylase. It is therefore possible to deactivate the enzyme kilning, in malt. However, it does exist and does have an im-
that makes fermentable sugar before the α-amylase can break pact, albeit an insignificant one as relates to the amylase activ-
down the starch for β-amylase attack by freeing up nonreduc- ity. I mention it here to help stimulate thoughts of “what if” as
ing ends for the enzyme to attach to. Thus, we need some bal- we look forward to the section on commercial enzymes. Dur-
ance. Keep in mind that most enzymes are active at some level ing that discussion we will refer to limit dextrinase as pullula-
up to their deactivation temperature. They are just not as active nase, as these enzymes are typically known within this indus-
as they are at their optimal temperature. So at the optimal tem- try. As you ponder scenarios, consider that there are other tools
perature for β-amylase, 145°F, α-amylase is digesting starch, available to help you achieve what you are trying to accom-
just not at the same high rate it would be if the temperature plish.
was 165°F, roughly seven degrees higher than the deactivation
temperature of β-amylase. Shifting the rate of temperature β-Glucan Opportunities
increase and the actual hold time (if used) can influence the Lastly, we look at β-glucan solubilase, a problematic enzyme
ratio of activity of the enzymes and thus affect the end ratio of when considering malt that may be borderline in modification
fermentable to nonfermentable sugars. Highly modified North (slightly high in β-glucans). Additionally, longer holds for con-
American malts possess extremely high diastatic power (starch- version or delays in your brewhouse process, holding up your
reducing capacity). In all-malt mashes, fermentability can be lautering, can also play into this enzyme. Earlier we mentioned
established by the rate of temperature increase, often negating the optimal activity temperature of this enzyme at 140–158°F.
the need for an actual conversion temperature hold. Interestingly enough, this temperature range centers closely on
the range most brewers consider for conversions. Most brewers
Time will mash off to lauter/strain at temperatures at or greater than
Given this concept of balance, we must now introduce the 163°F, close to the β-glucan solubilase deactivation temper-
variable of time into the discussion. The β-amylase works by ature. But often if brewers are stretching the limits of the
chewing up a linear strand of starch or dextrin, one maltose β-amylase activity, some will attempt to mash off in the 150s,
molecule at a time. On a relative basis, it is a fairly slow proc- leading to extended time for the β-glucan solubilase to work as
ess. The α-amylase cares little about where it attacks the well. From its name, this enzyme solubilizes β-glucan, freeing
starch, and therefore it is considered relatively fast. If brewers it from the husk material, not only allowing more into solution
are looking for more fermentability (higher alcohol, thinner, but also further degrading the integrity of the filter bed. It is a
and faster beer), they will need to place the conversion temper- quandary on how to deal with this enzyme, except to offer two
ature closer to the optimal activity of the β-amylase. Addition- solutions if you venture down the path of long holds. The first
ally, they will need to consider the relative rate of reaction, and deals with extremely well-modified malt. Keep in mind this is
the hold time for the conversion will need to be longer, again not standard material available to the public: one must request
on a relative basis. Conversely, if they are looking for less fer- and probably pay for it. To do this, maltsters must play with
mentability (lower alcohol level with greater body and sweet- the maximum ranges of processes that can stress the grain and
ness), they would choose a higher temperature to favor the create or flirt with premature yeast flocculation potentials in
α-amylase activity, while carefully considering the deactiva- the malt. So if you take this path, understand that there are no
tion temperature of the β-amylase and noting that the reaction solutions to premature yeast flocculation yet, other than to
will occur quickly. avoid it and apply a general-use commercial β-glucanase in
your process (see more later).
Words of Caution
As you look at the opportunity before you, I offer a couple Expanding Your Knowledge
words of caution and advice. When dealing with mash temper- For a greater understanding of enzymes in malting and brew-
atures close to or below 147°F, one needs to consider the mi- ing, read the two articles reproduced in this issue of the TQ:
crocharacteristics of the process. As you drop near to 145°F “Development of Enzymes During Malting and their Func-
Enzymes in Brewing MBAA TQ vol. 52, no. 3 • 2015 161

tion in Mashing” by Neville Prentice and “Biochemistry of Attenuation Control

Malting” by B. R. Sebree (7,9). I have utilized these two ref- The largest family of enzymes available for commercial use
erences for much of my career and find them to be very in- is categorized typically under attenuation control. You will also
formative. For additional considerations, also see an excellent see that some of these will be categorized elsewhere as well.
TQ article by L. Narziss (5) and a JIB article by Vriesekoop These too can be broken into families: amylases, glucoamy-
et al. (10). lases and amyloglucosidases, and limit dextrinases (pullula-
nases) (Fig. 2).
Commercial Sources
Attenuation Control: Maltose-Producing Amylase
We have discussed the basics of creating the enzymes in
Earlier, we looked at the natural occurrence of α- and β-amy-
malt for the brewer, and then mashing with those enzymes,
lases in malt. These two enzymes are available in several
targeting specific objectives within those processes. As we
forms commercially. As we walk through these, consider your
progress through this final section we will begin with many of
processes, because there may be several reasons to consider
the same enzymes discussed earlier, but these are now coming
the use of these enzymes. These enzymes can increase the rate
from nonmalt sources. Typically, these are derived in fermen-
of reaction, increase the yield, and help you attain your end
tations similar to brewing culture operations, but the organisms
alcohol targets, especially if you are using either poorly modi-
producing the enzymes are either bacteria or fungi (that are not
fied malts or cereal grains lacking in enzyme power. To begin,
brewer’s yeast). As I progress through each of these enzyme
β-amylases are hard to come by. Typically, this enzyme is only
groups I will not reflect on specific benefits of using one com-
commercially available as a malt extract. It is not widely used
pany’s product versus another. It should be noted that often
within the brewing industry because of its cost and the avail-
they are extremely similar, but sometimes they are quite differ-
ability of other products. There is currently only one other
ent, primarily because of the various host organisms chosen to
alternative to β-amylase, and this is a widely used product:
make the enzymes. In nearly all cases the enzymes marketed
fungal α-amylase. Although this product is very temperature
by these companies possess smaller side activities of addi-
sensitive, it works similarly to malt β-amylase in producing
tional enzymes. These activities can be beneficial or not. Be-
maltose as it digests starch. There has been industry discussion
cause this is a natural process and the organism is influenced
around this area, and I believe some progress has been made.
to produce the specific enzyme of interest by influencing its
This is a changing industry. Staying aware of changes is rec-
environment, the influence applied cannot totally prevent the
ommended if you have a need or interest. If you need to tune a
organism from making additional enzymes. This is why you
conversion, these products can help maintain a maltose-based
may see side activities. Some companies will claim them, oth-
wort, thus avoiding glucose-suppressing issues with high real
ers will not. In most cases the activity levels are not deter-
degree of fermentation (RDF) creating products. If you are
mined or regulated within their processes. When looking at
unaware of glucose suppression, it is a serious issue when look-
these products it is wise to ask about side activities as well as
ing to work with high-RDF products. I will review this in some
conducting bench-top mashes for evaluation of outcome. Also,
detail later in this discussion.
within this discussion I will not reference dosing rates because
these too will vary from product to product. For the purpose of Attenuation Control: α-Amylases
keeping this discussion to a reasonable length I will not con-
Moving on into the α-amylases, there is a vast range of
sider the range of products to use if you are looking at very
these. Some that are available will have similar sensitivities to
high additions of or are brewing with specialized grains such
temperature and pH as the malt α-amylase, and you can look
as sorghum or greater than 50% unmalted grains. We will,
at these as enhancements. However, there are other amylases
however, touch upon the use of barley, rye, and wheat, because
for mid and high to very high temperatures that were devel-
these additions can pose significant challenges to any brewer.
oped primarily for the syrup industry. These products also
The choices you make when looking for help via the use of
come as blends or individual products. For brewing, the use of
commercial enzymes can be greatly influenced by knowledge
about the specifics of these grains.
Commercial Enzyme Categories
Most commercial enzyme companies break their products
into several basic areas. The first two major groups deal with
attenuation control and with optimization of mashing and beer
filtration. The next major group often is associated with cost-
effective cereal cooking. However, many of these enzymes can
also be used in improving your attenuation control. Sometimes
enzyme classifications may roll over from one area to the next.
The next groups we will discuss get more specific. These
groups fall into categories that are often called cost-effective
adjunct and malt opportunities, optimal or improved fermenta-
tion control, specialty grains, and finally one very specific
classification that can address improved haze control. In al-
most all circumstances the enzymes are added in the main
mash or cooker mash unless otherwise indicated. In most cases
it is important to deactivate the enzymes in your kettle boil, as
is done with natural malt enzymes, thus stopping any changes Figure 2. Starch-reducing enzymes, showing what portions of the
driven by their activity in your products. molecule they work on ( = glucose). Image courtesy of DuPont.
162 MBAA TQ vol. 52, no. 3 • 2015 Enzymes in Brewing

blends makes good sense in that you can reduce viscosity ear- Optimization of Mashing and Beer Filtration:
lier at a lower temperature and then at the high temperature the β-Glucanases
enzyme will provide extract and so on. This sort of discussion
falls closer to the use of adjuncts and the liquifaction and ge- If you look back at the discussions around grain modifica-
latinization of the starches in the cooking process. However, tion, you will begin to understand the critical effect high levels
for attenuation control, and more closely related to malt proc- of β-glucans and arabinoxylans have on processing your prod-
esses, these amylases can be added as a supplement to proc- uct (Fig. 3). Just getting the product through your straining
esses that need additional amylase power to get a reasonable device can be problematic, at best, if you have borderline or
conversion within a reasonable time. poorly modified grains. Additionally, if you choose, as many
do, to add specialty grains that have low levels or are free of
Attenuation Control: Glucoamylases the needed enzymes, or are totally lacking in modification
The most widely used products for attenuation control within from malting, you are probably experiencing problems. These
brewing are in the family of enzymes known as glucoamy- problems can also be translated later into your process and
lases. You may on occasion hear these referred to as amylo- beer flavor. For those that filter their products, the presence of
glucosidases, but for the purpose of this discussion you can elevated levels of β-glucans and of xylans (as little as 10%
consider these to be the same. Glucoamylases attack starch equivalent by molecular weight) can result in frustrating filter
molecules from their nonreducing end, clipping off single glu- runs. As pointed out earlier, I believe in a consistent process
cose molecules one at a time. They are capable, for this discus- flow. I want my beer to possess the flavor attributes I choose
sion, of only breaking the 1-4 glucosidic bonds and therefore and not those caused by difficult or delayed runoffs and filtra-
cannot proceed past the 1-6 branch points in amylopectin. When tion. Husky, grainy, and oxidized notes are just a few of the
used with well-modified malt (an ample supply of α-amylase issues created by this struggle, a struggle that you can make go
to break up the starch molecules, especially past the 1-6 branch away with very little effort. In the market today there are a
points) these enzymes can easily achieve a very high degree of large number of available β-glucanases, xylanases, and combi-
fermentability. Under nonspecial circumstances, RDFs over nation products. In addition, you may run into products labeled
80% are not difficult to attain. Values approaching 85% are as pentosanases and cellulases. These are just different exam-
fairly easy with some special considerations, and values as ples of what can be used. The pentosanases are technically
high as 87% have been achieved in all-malt mashes. I mentioned xylanases, which I refer to specifically later, and the cellulases
glucose suppression (or repression, as referred to by some) are typically blends of β-glucanases and xylanases, which I
earlier, and typically such issues can come into play with the also refer to later. What your problems are, or what your ob-
use of glucoamylases. With the use of glucoamylases our ob- jectives are, will play into your decision on which to use. The
jective is to convert most of the starch and dextrins to glucose. most popular product to use is just a β-glucanase. Most brew-
If you do this and get most, perhaps 70%, of it converted over ers who use these do so to remove the variation they are get-
to glucose, you will normally not see an issue. The glucose ting with their malt. If you are not paying for highly controlled
suppression issue comes with using small amounts of gluco- malt processing, you may be getting variability in your grains.
amylase as an enhancement to an RDF that is naturally at- By adding a small amount of typical β-glucanase, much of this
tained. A naturally based RDF will produce mainly maltose, variation can go away. Additionally, you may see a yield in-
normally greater than 60% as such. If you are using well-mod- crease that easily offsets the cost of the enzyme.
ified malts, it is not abnormal to naturally achieve RDFs just Optimization of Mashing and Beer Filtration:
under 80%. Wanting to get just a little more is where the prob- Xylanases
lem lies. Once you convert some of the sugars to glucose, you
begin to enter a no man’s land of sugar ratios for yeast fermen- Getting a little more specific, we will now look at xylanases,
tation. Most yeast will start to have issues when you approach more exactly arabinoxylanases. As you will recall, arabinoxy-
25–30% glucose. The problem comes when the yeast finish the lans are copolymers of two pentose sugars, arabinose and xy-
glucose and then must either continue to ferment or shut down. lose. This molecule is an integral part of the structural makeup
If the fermentation shuts down, it can get interesting. You can of the grain, closely associated with the β-glucans in this struc-
have unfinished products, elevated diacetyl, and so on. When ture. In most modification discussions and papers, little credit
considering all of these processes and products, knowledge is given to arabinoxylans, when in fact these are the evil broth-
and understanding are paramount to your success. ers of the β-glucans. The issues with arabinoxylans come with
the modification of the grain as well as the use of specialty
Attenuation Control: Limit Dextrinases (Pullulanases) unmalted grains that are high in arabinoxylans, such as wheat
One of the more interesting groups of enzymes available is and rye. Variability in modification of the grains during malt-
limit dextrinases (pullulanases). For the sake of consistency, I ing is normally the cause of arabinoxylan issues arising from
will refer to them here as pullulanases. These enzymes possess malting. The addition of a xylanase can be beneficial to your
the ability to break the 1-6 branch points in amylopectin. One process if you are seeing issues derived from the presence of
might think this enzyme would have a major impact on RDF. xylans, but you need to approach any hydrolysis of arabinoxy-
However, in reality, because of the normal activity of the α-amy-
lases, there is not. We can, however, use these either alone or
in conjunction with glucoamylases to speed up a conversion or
very slightly increase RDF. Terminal RDFs of 88% are pos-
sible under special conditions. Note that when used alone pul-
lulanases can have an interesting impact. They can raise your
RDF several percent while keeping the sugar profile similar to
a natural maltose-based profile, thus avoiding the glucose sup-
pression issue while getting a bump in fermentability. Figure 3. β-Glucan molecule. Image courtesy of DuPont.
Enzymes in Brewing MBAA TQ vol. 52, no. 3 • 2015 163

lans with great caution. We discussed earlier the close associa- the unused protein down to amino acids for your yeast. Prote-
tion of ferulic acid with this molecule. Notably, its digestion ases can be problematic in that they can reduce the foam char-
with enzymes can yield higher than normal levels of free feru- acteristics of your beers. However, if you are careful to use
lic acid in your wort and then, depending upon your process, only what you need to achieve the results required for your
issues with taste from the conversion product, 4VG. When you fermentations, you can normally avoid issues. The problems
consider the breakdown of xylans with commercial enzymes, with proteases come with overdosing. They have traditionally
consider that the ferulic acid is more closely tied to the water- been given a black eye because of the use of proteases as chill
soluble portion of the xylans. If you select the wrong enzyme, stabilizers. In this use, the enzymes were left to work unre-
hydrolyzing major fractions of non-water-soluble xylans can stricted for long periods of time, and they were typically over-
be problematic. Lastly, I mentioned wheat and rye specifically dosed for effect. Cautious use in mashing can have the needed
here because of their high xylan content. This is also true of impact to stabilize your process, should you be short on availa-
unmalted barley, but wheat and rye possess a protein that sup- ble FAN.
presses the effectiveness of most commercial xylanases. There
are a few commercial enzymes that can be effectively used Optimal or Improved Fermentation Control:
under these conditions. Understanding that this can be an issue α-Acetolactate Decarboxylase
will help you find the right solution because you can ask the As part of a normal fermentation cycle, fermenting beers
right questions. Using whatever is on the shelf will get you generates diacetyl. The diacetyl comes from a side reaction of
into trouble. yeast converting available amino acids for its own use. From
this reaction, the resultant α-acetolactic acid is expelled from
Optimization of Mashing and Beer Filtration:
the yeast as a by-product. In solution, this compound under-
Combination goes spontaneous oxidative decarboxylation, taking the com-
Today, most commercial β-glucanase/xylanase products come pound to diacetyl. Under normal conditions, yeast will take the
with some mixed activity. Because issues typically arise from diacetyl back into the cell and, via an energy-reclaiming proc-
variability in modification, these combination enzymes are most esses, an enzymatic transition using a reductase will convert
useful. Additionally, when working with unmodified grains, the diacetyl to acetoin. Again, the yeast cell will expel the
these enzymes will open up the grain matrix to allow full ex- acetoin, but the flavor impacts of acetoin are much less than
posure of the grains to the brewing processes. In all cases, diacetyl. There are techniques that can be applied from FAN
caution needs to be taken to avoid a 4VG issue. Your process levels to air content and so on that can effectively manage the
and hot delays will determine if a problem can or will occur. creation and later uptake of diacetyl. However, if you are expe-
riencing control issues that cannot be dealt with in any other
Cost-Effective Cereal Cooking: Amylases Again way, there is an enzyme known as α-acetolactate decarbox-
If you are cooking cereal grains such as corn and rice, the ylase that is added after the brewhouse. This enzyme will grab
use of exogenous amylases may be beneficial. The objective in onto the expelled α-acetolactic acid and convert it directly to
cooking is to gelatinize and then liquefy these grains to expose acetoin, while still in solution, outside the yeast cell. What this
the starch for enzymatic attack for conversion to fermentable does is allow you some freedom in fermenting capacity as well
sugars. Looks fairly easy—add some malt and give it a boil to as some control of diacetyl, if necessary, as you stretch the
blow up the starch granules. However, if you are working with limits of your processes through the addition of nonmodified
poorly modified malt or a large component of unmalted grains, grains. I must caution you that this enzyme is not a solution for
getting enough enzymatic power into the cooker can be a prob- high diacetyl in a product. It will not have any effect on al-
lem. This problem can be translated into taste impacts from ready formed diacetyl.
boiled grain as well as a loss of enzymatic power for your con-
version, because you boiled the grain portion placed in the Specialty Grains: Various Enzymes
cooker and killed all of the enzymes. If you back away from When we look at enzyme solutions for use with specialty
the malt addition to your cooker, you can substitute the use of grains, we need to ask some questions around how much, what
commercially derived amylases. The benefits are obvious to materials, and what portion of the grain bill remaining is good
the rest of your process, but the hidden benefit is yield. By quality well-modified malt. In simple terms, the industry
using a mid- and high-temperature blended amylase you can groups percentages of specialty grains into ranges. For exam-
liquefy this grain even without boiling (energy savings). You ple, we look at 0–10, 10–15, 15–25, 25–40, 40–60, and >60%
will need to get the mash to above 90°C, but a boil is not re- (to include 100%). When using well-modified malt, the 0–10%
quired. Additionally, the traditional enzymes would be fully range offers few issues. Sometimes you may need a little help
destroyed soon after you need them, because the gelatinization with the β-glucan and xylans, as discussed earlier, but no other
takes place very close to their deactivation temperature. With issue should arise. As we move up the scale, you are looking at
the blended amylases, the high-temperature amylase takes over a need to add more and more aggressive enzyme complexes to
and will continue to function though most of the boil, freeing gain access to the nonmodified grains and to keep the proc-
up additional starch (increased extract). esses running free of delays. The higher you go, the less FAN
you have available, because you are diluting the modified malt
Cost-Effective Adjunct or Malt: Proteases and portion of your grain bill. Again, if you have well-modified
Peptidases malt you can get to upward of 50% substitution without the
Typically, if you are going to have a problem with FAN lev- need of a protease. As you increase your addition of specialty
els in your wort you are either stretching the limits of your grains, the need for β-glucanases and xylanases can go higher
processes or you are using non-protein-bearing materials that and higher. As you increase these ratios, you move away from
are diluting the materials that bring you the FAN. In this case, simple blended β-glucanase and xylanase products to products
you may need to add a protease or peptidase to digest some of that have more of the same but with varying temperature sensi-
164 MBAA TQ vol. 52, no. 3 • 2015 Enzymes in Brewing

tivities. In that manner, the enzymes provide a broad attack on mentioned here. It should be noted again that what is men-
the β-glucan and xylans. Therefore, in looking at the 10–15% tioned here is not an all-encompassing review of what en-
addition range, the more basic solutions mentioned earlier are zymes influence the process; these are just the main ones.
appropriate. Sometimes the more simple solutions can be It should also be clear from a review that malting is not
applied to the 15–25% range as well; it may be just a matter of simply a process of growing barley. It is a controlled process
dosage rate. However, once you go above 25%, the problems of influencing natural enzymes along the way. These enzymes
can be challenging. At this point you might see a need for an provide the brewer with the tools to make beer. It is also easy
amylase to be added as well, but this is a matter for close re- to see that what happens in malting translates well to brewing
view. Commercial enzyme suppliers normally have complex and beer. The brewing process takes consideration of all that
blends of β-glucanase, xylanase, pentosanases, amylases, pro- happens in malting and more. The small effects of other en-
teases, and peptidases. Sometimes you need all of these and zymes can have a devastating effect on products. Awareness is
sometimes you do not. My recommendation to you is to ap- the driver to making a great beer.
proach these issues with your eyes wide open and with some When all else fails or there is a need to stretch the limits of
expertise in your pocket. Remember, if you are using wheat or the process beyond what can be done naturally, there are tools
rye, the game changes a little. Lastly, as you use these un- at the brewer’s disposal: commercial enzymes. As noted within
malted grains in high percentages, often the filterability of the context here, even these enzymes need to be approached
your straining device suffers. Sometimes it is an effect caused with your eyes open. As with almost everything done in brew-
by an increase in β-glucans and xylans, and sometimes it a ing, there are consequences to each move made in a process. It
loss of husk materials. As you venture down this road, con- is the balance that makes the best beers. This balance is gained
sider the use of filter aids in your mash as well. Addition of through greater understanding of as many of the causes and
rice hulls, for example, can greatly improve your run-off and effects as possible. From this awareness we can relate to proc-
wort characteristics. ess change and make the best decisions for products.
Improved Haze Control: Proline-Specific Endoprotease
REFERENCES
A recent introduction to the family of commercial enzymes
is an enzyme that can be added postfermentation for the pur- 1. Baekgaard, L. (2012). Brewing with barley: Comparing protease ac-
pose of chill haze stabilization. This highly specific protease tivities with the resulting proteins and peptides in beer using activity-
based protein profiling and LC-MS/MS. Proceedings of the World
attacks a specific sequence of amino acids that tends to be
Brewing Congress. www.mbaa.com/meetings/archive/2012/Proceedings/
proline-rich. Traditionally, proline-rich protein fractions have pages/5.aspx
been tied to chill haze issues. These proline-rich fragments 2. Bamforth, C. (1999). A critical assessment of malt analysis from the
form the protein–polyphenol complexes associated with chill brewer’s perspective. Tech. Q. Master Brew. Assoc. Am. 36:301-306.
haze. By breaking down these proline-rich fragments, brewers 3. Hertrich, J. D. (2013). Topics in brewing: Malting barley. Tech. Q.
have been able to effectively improve the chill haze stability of Master Brew. Assoc. Am. 50:29-41.
beers. The enzyme specifically catalyzes a carboxyl-terminal 4. Hertrich, J. D. (2013). Topics in brewing: Malting. Tech. Q. Master
hydrolysis of proline, thus rendering it unavailable to assist Brew. Assoc. Am. 50:131-141.
with building these protein–polyphenol haze complexes. As an 5. Narziss, L. (1976). The influence of mashing procedures on the activ-
aside, there has been some discussion regarding this enzyme’s ity and effect of some enzymes—A survey. Tech. Q. Master Brew. As-
ability to also reduce, at least analytically, a portion of gluten soc. Am. 13:11-21.
6. Mathewson, P. R. (1998). Enzymes. Eagan Press: St. Paul, MN.
in beer, thus rendering a product analytically low-gluten or
7. N. Prentice (1976). Development of enzymes during malting and their
gluten-free. The actual effect of this enzyme to change the function in mashing. Tech. Q. Master Brew. Assoc. Am. 13:91-101.
product so it does not impact celiacs is under evaluation. 8. Pugh, T. A., Maurer, J. M., and Pringle, A. T. (1997). The impact of
wort nitrogen limitation on yeast fermentation performance and diace-
Summary tyl. Tech. Q. Master Brew. Assoc. Am. 34:185-189.
9. Sebree, B. R. (1997). Biochemistry of malting. Tech. Q. Master Brew.
By now it should be obvious that the enzyme framework in Assoc. Am. 34:148-151.
brewing far exceeds the basics of α- and β-amylase. Although 10. Vriesekoop, F., Rathband, A., MacKinlay, J., and Bryce, J. H. (2010).
the most important enzymes related to the process of brewing, The evolution of dextrins during the mashing and fermentation of all-
they are shadowed by the magnitude of the remaining enzymes malt whisky production. J. Inst. Brew. 116:230-238.