Liu 1997

2
Chemistry and Nutritional Value of

Soybean Components
The soybean is one of the most economical and valuable agricultural commodi-
ties because of its unique chemical composition. Among cereal and other legume
species, it has the highest protein content (around 40%); other legumes have a
protein content between 20% and 30%, whereas cereals have a protein content
in the range of 8-15%. The soybean also contains about 20% oil, the second
highest content among all food legumes. (The highest oil content is found in
peanut, which is about 48% on dry matter basis. The third highest oil content is
chickpea, which is about 5%. The remaining food legume species have oil contents
in the range of 1-3.6%) (Salunkhe et al. 1983). Other valuable components found
in soybeans include phospholipids, vitamins, and minerals. Furthermore, soybeans
contain many minor substances, some of which, such as trypsin inhibitors, phy-
tates, and oligosaccharides, are known to be biologically active. Others, such as
isoflavones, are just being recognized for their powerful ability to prevent human
cancers and other diseases (Messina et al. 1994, Chapter 10 of this book). In
this chapter the chemical components of soybeans are discussed with respect to
their occurrences, properties, nutritional value, physiological roles, and assay
methodology.
I. Proximate Composition
On the average, oil and protein together constitute about 60% of dry soybeans.
The remaining dry matter is composed of mainly carbohydrates (about 35%) and
ash (about 5%). Since the water content of stored mature beans is usually about
13% to ensure storage stability, on a wet basis, soybeans contain about 35%
protein, 17% oil, 31 % carbohydrate, and 4.4% ash.
In general, cultivated soybeans comprise approximately 8% hull, 90% cotyle-
25
K. Liu, Soybeans
© Chapman & Hall 1997
26 / Soybeans: Chemistry, Technology, and Utilization
Table 2.1 . Proximate Composition of Soybeans and Their Structural Parts

Chemical Composition
(% dry matter)
Percentage in
Whole Seeds Protein Lipid Carbohydrate Ash
Hull 8 9 86 4.3
Hypocotyl axis 2 41 11 43 4.4
Cotyledons 90 43 23 29 5.0
Whole seeds 100 40 20 35 5.0
Source: Data adapted from Wolf and Cowan (1975).
dons, and 2% hypocotyl axis (Table 2.1). Cotyledons contain the highest percent-
age of both protein and oil, whereas the hull has the lowest values of these
components. In fact, the oil content in hulls is so low that it can be regarded as
a trace amount. The hypocotyl axis has a protein content similar to cotyledons
but its lipid content is about half that in cotyledons. Since the cotyledon is the
major component in the whole seed, its composition is very close to that of the
whole seed regardless of great compositional differences among structural parts.
The actual composition of the whole soybean and its structural parts depends
on many factors, including varieties, growing season, geographic location, and
environmental stress. Liu et al. (l995a) reported that among the 10 selected
soybean genotypes grown in Arkansas, on a dry matter basis, protein varied from
39.5% to 50.2%, oil 16.3% to 21.6%, and protein plus oil 59.7% to 67.5%.
Among the lines in the U.S. germplasm collection, however, the range is even
greater, with protein varying from about 30% to over 50% and oil from about
12% to almost 30% (Orf 1988).
Hurburgh (1994) compiled soybean protein and oil regional data from eight
years of U.S. surveys and found that soybeans grown in the Western corn belt
were consistently one percentage point lower in protein than soybeans grown in
the remainder of the United States. There was year-to-year variability in protein
patterns among regions also. Oil percentages were more variable than protein
percentages among different years.
Drought and temperature also affect the chemical composition of soybeans.
Dornbos and Mullen (1992) reported that severe drought increased protein content
by 4.4 percentage points, whereas oil content decreased by 2.9 percentage points.
As drought stress increased, as measured by accumulating stress degree days,
protein content increased linearly and oil content decreased linearly at each
air temperature.
II. Lipids
During seed development, soybeans store their lipids, mainly in the form of
triglycerides, in an organelle known as oil bodies. In some literature, oil bodies
Chemistry and Nutritional Value of Soybean Components I 27
are also referred to as lipid bodies, spherosomes, oleosomes, or lipid-containing

vesicles. Figure 2.1 shows an electron micrograph of a soybean cotyledon cell
in which the structural elements are identified. The cell has walls, membranes,
and cytoplasm. However, unlike regular plant cells, many organelles such as
the nucleus, mitochondria, and endoplasmic reticulum disappeared as the seed
matured. Instead, the cell contains large protein bodies surrounded by many
small oil bodies. Like other oilseed crops, oil bodies in soybeans are relatively
homogeneous in size, ranging between 0.2 and 0.5 !l in diameter. They are small
in size compared with those in peanuts, which range between 1.0 and 2.0 !l in
diameter (Jacks et al. 1967).
During processing, components extracted from soybeans by organic solvents
such as hexane are classified as crude oil. Major components of crude oil are
triglycerides (or triacylglycerols). Minor components include phospholipids, un-
saponifiable material, free fatty acids, and trace metals. Unsaponifiable material
consists of tocopherols, phytosterols, and hydrocarbons. The concentrations of
these minor compounds are reduced after typical oil processing (Table 2.2).
A. Triglyeerides
Refined soybean oil contains more than 99% triglycerides (Table 2.2). Triglycer-
ides are neutral lipids, each consisting of three fatty acids and one glycerol that
links the three acids. The functional properties, oxidative stability, as well as the
nutritional value of edible oils in general and soybean oil in particular are all
determined by their fatty acid composition, geometric configuration, and posi-
tional distribution.
Figure 2.1. Electron micrograph of soybean cotyledon cells. PB = protein bodies, S =

spherosomes, and CW = celt wall. From Saio and Watanabe (1968).
Table 2.2. Typical Composition of Crude and Refined Soybean Oil

Components Unit Crude oil Refined oil
Triglycerides % 95-97 >99
Phosphatides % 1.5-2.5 0.003-0.045
Free fatty acids % 0.3- 0.7 <0.05
Unsaponifiable matter % 1.6 0.3
Plant sterols % 0.33 0.13
Tocopherols % 0.15- 0.21 0.11-0.18
Hydrocarbons % 0.014 0.01
Trace metals
[ron ppm 1- 3 0.1-0.3
Copper ppm 0.03- 0 .05 0.02-0.06
Source: Data adapted from Pryde (1980).
I. Fatty Acid Composition
Each natural fat or oil, regardless of its origin, has a unique fatty acid composi-
tion (Table 2.3). Like many other oils of plant origin, most fatty acids in soybean
oil are unsaturated. The highest percentage of fatty acid in soybean oil is linoleic
acid, followed in decreasing order by oleic, palmitic, linolenic, and stearic acid.
Soybean oil also contains some minor fatty acids, including arachidic, behenic,
palmitoleic, and myristic acid.
There is a large genetic variation in fatty acid composition of soybean oil,
mainly resulting from plant breeding. The range of fatty acid composition among
soybean germplasm has been reported to be: CI6:0, 8-l7%; CI8:0, 3-30%;
C18: 1,25-60%, CI8:2, 25-60%; and CI8:3, 2-15 % (Hammond and Glatz 1989).
A few genotypes having large variation in fatty acid compo sting from normal
cultivars have been successfully released as commercial varieties with a compara-
tive yield. Among them are soybeans with C18:3 as low as 3%, soybeans with
total saturate as low as 7%, and soybeans with C18:0 as high as 20%. Oils from
these lines are targeted for specific applications for their nutritional or functional
advantages over common soy oil. Chapter 11 provides additional discussion on
the subject.
Interestingly, Liu et al. (l995a) reported that among the 10 normal genotypes
tested, there are significant correlations between soybean seed size and individual
unsaturated fatty acids: positive with oleic acid (r = 0.808), and negative with
linoleic (r = -0.796) and linolenic acid (r = -0.603). There are also correlations
among unsaturated fatty acids: negative between oleic and linoleic (r = -0.806)
or linolenic acid (r = -0.815), and positive between linoleic and linolenic acid
(r = 0.407). A subsequent study reported in the same paper with soybeans grown
in 1993 generally confirmed these findings.
Due to differences in fatty acid composition, fats and oils exhibit different
physical properties as well as oxidative stability during storage and food applica-
Table 2.3. Typical Fatty Acid Compositions of Selected Edible Fats and Oils
Relative Percent
Name of fat Lauric Myristic Palmitic Palmitoleic Stearic Oleic Linoleic Linolenic Arachidic Gadoleic Behenic Erucic
and oil C12:0 C14:0 C16:0 C 16:1 C18:0 C18:1 C18:2 C18:3 C20:0 C20:1 C22:0 C22:1
Soybean oil 0.1 11.0 0.1 4.0 23.4 53.2 7.8 0.3 0.1
Canola oil 3.9 0.2 1.9 64.1 18.7 9.2 0.6 1.0 0.2
Com oil 12.2 0.1 2.2 27.5 57.0 0.9 0.1
Sunflower oil 0.5 0.2 6.g 0.1 4.7 18.6 68.2 0.5 0.4
N Safflower oil 0.1 6.5 2.4 13 .1 77.7 0.2
\0
Rapeseed oil 0.1 2.8 0.2 1.3 23.8 14.6 7.3 0.7 12.1 0.4 34.8
Rice bran oi l 0.4 0.5 16.4 0.3 2.1 43.8 34.0 1.1 0.5 0.4 0.2
Peanut oil 0.1 11.6 0.2 3.1 46.5 31.4 1.5 1.4 3.0
Olive oil 13.7 1.2 2.5 71.1 10.0 0.6 0.9
Palm oil 0.3 1.1 45.1 0.1 4.7 38.8 9.4 0.3 0.2
Cottonseed oil 0.9 24.7 0.7 2.3 17.6 53.3 0.3 0.1
Lard 0.1 1.5 24.S 3.1 12.3 45.1 9.9 0.1 0.2 1.3
Beef tallow 0.1 3.3 25.S 3.4 21.6 38.7 2.2 0.6 0.1
Minimum 0.0 0.0 2.8 0.0 1.3 13.1 9.4 0.0 0.1 0.0 0.1 0.0
Maximum 0.5 33 45.1 3.4 21.6 71.1 77.7 9.2 1.5 12. 1 3.0 34.8
Source: Unpublished data, adapted from Van Den Bergh Food Ingredients Group, Lisle, IL.
tion/processing, In terms of physical properties, the greater the chain length of

a fatty acid, the higher the melting point. A chain length of ten or twelve carbons
marks the transition from liquid saturated fatty acids to those solid at room
temperature. An introduction of a double bond into a long chain fatty acid lowers
its melting point significantly. Oils containing a high percentage of saturated
fatty acids have a high melting point, giving a semisolid or solid appearance,
whereas those containing a high percentage of saturated fatty acids have a low
melting point, thus giving a liquid appearance.
The presence of a double bond in an unsaturated fatty acid also makes it
susceptible to oxidation, leading to the development of an off flavor. The more
double bonds present, the less stable the fatty acid. Linolenic acid contains three
double bonds, thus is the least stable. Because soybean oil contains relatively
high proportions of both linoleic and linolenic acids, the chemical stability of
soybean oil has been a problem. To increase the melting point as well as the
oxidative stability of soy oil, hydrogenation becomes necessary. The process
chemically adds two hydrogen atoms to a double bond of an unsaturated fatty
acid in the oil. As a result, lower unsaturated fatty acids are converted to saturated
acids, and higher unsaturated fatty acid to lower unsaturated acids in the oil.
In oil industries, the degree of unsaturation of a fat is normally expressed in
terms of the iodine value (IV) of the fat. The IV is the number of grams of iodine
that will react with the double bonds in 100 grams of fat. The average IV of
unhydrogenated soy oil is in the range of 125-l35. A typical salad and cooking
oil made from partially hydrogenated soy oil may have an IV of 110-120. A
typical semisolid household shortening made from partially hydrogenated soy-
bean oil may have an IV of 90-95. The melting characteristics of fats and oils
is measured empirically based on density variations between solid and liquid and
is expressed as solid fat index (SFI). However, because it is labor intensive and
prone to errors, the SFI method is being replaced by a new method known as
nuclear magnetic resonance (NMR), which measures the solid content of edible
oils. The result is expressed as solid fat content (SFC). Since different food
applications require fats and oils with different physical properties, it is important
to consider these parameters when choosing an oil for a specific food application.
Further discussion on soybean oil stability and food application is covered in
Chapter 7.
2. Geometric Configuration
In a fatty acid containing a double bond, there are two geometric configurations:
cis and trans (Fig. 2.2). In the cis configuration, two hydrogens are located in
the same direction, whereas in the trans configuration, they are in the opposite
direction. As a result, the linearity of the fatty acid hydrocarbon chain is affected.
The geometric configuration of fats and oils has a significant impact on their
physical and chemical properties, as the trans configuration generally melts at a
higher temperature than the cis configuration.
Chemistrv and Nutritional Value of Soyhean Components / 31
cis Isomer trans Isomer

Figure 2.2. Geometric configurations in an unsaturated fatty acid.
All naturally occurring fatty acids of plant origin are in the cis form. Trans
fatty acids are generally formed when oils and fats are hydrogenated or heated
at a high temperature . All unsaturated fatty acids, such as oleic, linolenic, and
linoleic acids, can exist in one or more geometric isomers.
3. Positional Distribution
Distribution of fatty acids at the glycerol molecule of a triglyceride is another

quality factor. The position of the carbon atoms in the glycerol moiety of a
triglyceride can be stereospecifically numbered (SN) as 1, 2, and 3 (Fig. 2.3).
The attachment of fatty acids in soybean oil on these positions is not based on
a random order. Rather, according to Evans et al. (1969), it is based on the
following rule: (I) Saturated fatty acids (mainly palmitic and stearic acids) and
those with chain length greater than 18 carbons are first distributed equally and
randomly on the positions 1 and 3; (2) Oleic and linolenic acids are treated alike ,
or as a unit, and distributed randomly and equally in all three positions; and (3)
all remaining positions are filled by linoleic acid.
The orderly fatty acid distribution in soybean oil as well as other fats and oils
can be altered to a random pattern by an industrial process known as interester~fi
cation. A general chemical interesterification involves heating oil to drive away
water residue. mixing the oil with a catalyst. normally sodium methylate or
sodium ethyl ate. breaking the emulsion after completion of the reaction, and
separating and drying the oil layer. Although it does not change fatty acid
o1\
CH 2 0H Position 1 CH 20CR,
HQ--CH
I Position 2
)( I
R2 COHC
I
CH 0H
I )(
CH 0CR
2 Position 3 2 3
Glycerol (sn-glycerol) Triglyceride (triacylglycerol)
Figure 2.3. Molecular structure of glycerol and triacylglycerol showi~g their SN posi-
tions.
composition, interesterification generally increases crystallization tendencies

(melting point) of fats and oils, Details are provided in Chapter 6,
Reports differ regarding the effect of fatty acid distribution within triglyceride
molecules on oxidative stability. Some reported that when specific fatty acids
occupy the 1 and 3 positions of a triglyceride, oxidative stability is greater than
when those same fatty acids occupy the 1 and 2 positions (Neff et al. 1992).
Others found that the positional distribution has no effect on oxidative stability
(Park et al. 1983)
B. Phospholipids
Crude soybean oil contains 1-3% phospholipids. Among the total phospholipids
in soybeans, there are about 35% phosphatidyl choline, about 25 % phosphatidyl
ethanolamine, about 15% phosphatidyl inositol, 5-10% phosphatidic acid, and
the rest is a composite of all the minor phospholipid compounds. Figure 2.4
shows the molecular structure and formation of three major types of phospholipids
found in soybeans. The parent compound is phosphatidic acid, which is not
present in the free form in active cells except as an intermediate in the biosynthesis
of other phosphoglycerides. Others are esters of phosphatidic acid.
Both triglycerides and phospholipids are saponifiable but phospholipids are
polar lipids. Removal of polar lipids from crude oil is carried out by centrifugation
following hydration at an elevated temperature, the process commonly known
as degumming. Phospholipids are good emulsifying agents, soluble in alcohol
and insoluble in acetone. In living tissues, they are the major components of
cell membranes.
It should be emphasized that phosphatidyl choline's common name is lecithin.
However, in broad usage, the term "lecithin" generally refers to the entire phos-
pholipid fraction separated from soybean crude oil by degumming. Lecithin
processing and utilization are covered in Chapters 6 and 7, respectively.
- ~O
o H O-CH 2CH 2 -NH, ---t~~ Phosphatidyl ethanolamine
II
R-C-O- CH2
I
R'-C-O- CH 0
II
o
I
CH - O-P-OH
2
II
I
+ -~o
HO-CH 2CH 2 - W(CH 3h ---I~~ Phosphatidyl choline
o (Lecithin)
~OH -~o
Phosphatidic acid H O / H ' '' ---t~~ Phosphatidyl inositol
OH
Figure 2.4. Molecular structure and fonnation of phospholipids commonly found in soy-
beans.
Chemistry and Nutritional Value of'Soybean Components / .U
C. Nutritional Value of Soybean Oil
In the gastrointestinal tract, dietary fats and oils are hydrolyzed into monoglycer-
ides, free fatty acids, and glycerol before absorption in the intestine. Following
reesterification within the intestinal cells, they are incorporated in the chylomi-
crons and secreted into the blood via the lymphatic system. From then on, they
carry out numerous nutritional and physiological functions. First, they are a
concentrated source of energy. Approximately 30-40% of our daily energy intake
is derived from fat, most of which are triglycerides containing fatty acids with
16-18 carbon atoms. Second, they are a unique source of essential fatty acids
(EFA). Third, they regulate the lipid blood level and are carriers for fat-soluble
vitamins. And fourth, they are normal constituents of cellular structure, especially
cell membranes.
Soybean oil is a leading edible oil widely used in various food products,
including salad and cooking oil, shortening, margarine, mayonnaise, and salad
dressing. Like other edible oils, besides culinary roles, soy oil provides us with
calories, essential fatty acids, and fat-soluble vitamins. Since unhydrogenated
soy oil contains about 53% linoleic acid and 8% linolenic acid, whereas partially
hydrogenated soy oil still contains about 23% linoleic and 3% linolenic acid,
soybean oil is an excellent source of essential fatty acids. It is also a healthy oil,
comparing favorably with canola oil and other highly unsaturated oils.
In recent years there has been a dramatic evolution in our knowledge of dietary
fats and oils and their health implications. Some discoveries have changed our
long-held beliefs, whereas others have contributed to controversies. In previous
years we treated fat only as a macronutrient. Currently our concern is with not
only saturated or unsaturated fat but also individual fatty acids. Furthermore,
stereo configuration of an individual fatty acid, such as trans fatty acids, has
been another concern. This chapter provides a limited discussion on the subject;
however, Chow (1992) discusses fatty acids in greater detail.
1. Essential Fattv Acids
Two long-chain fatty acids are now considered essential: linoleic (L'l9,12-
octadecadienoic acid, or C 18:2n-6) and linolenic (L'l9,12,5-octadeatrienoic acid,
or C 18:3n-3). although the essentiality of C 18:3n-3 in high animals and humans
has been debated for many years. They are essential because mammals, including
humans, cannot synthesize them. More specifically, our body cannot introduce
double bonds between the terminal methyl group and the first double bond situated
in the carbon chain of the respective fatty acid. The only alternative our body
has is to introduce acetyl groups to elongate the carbon chain and to form new
double bonds between the last double bond and the carboxylic group of the fatty
acid. In other words, syntheses of arachidonic (C20:4n-6) and decosahexaenoic
acid (C22:6n-3), the two important precursors for prostaglandins and other eicosa-
noids, depend on dietary supplies of linoleic and linolenic acids.
Depressed growth, scaly dermatoses, increased permeability of skin, fatty liver,

kidney damage, increased susceptibility to infection, and impaired reproduction
are typical symptoms reported when higher animals and humans lack a dietary
supply of essential fatty acids (EFA) (Chapkin 1992). Ingestion of approximately
1-2% of daily calories as CI8:2n-6 is widely accepted as the amount needed to
meet the EFA requirement of rodent species and humans. Recent dietary estimates
to prevent n-3 depletion of liver and brain phospholipids in humans are 14 g/
day of C1S:2n-6, 3g/day of CI8:3n-3 and ISg/day of total polyunsaturated fatty
acids (Simopoulos \989).
2. Health Implications of Individual Fatty Acids
Increasing evidence indicates that the types and levels of fats and oils consumed
significantly influence the weB-being of the general population. Dietary lipids
have been found to playa significant role in the pathogenesis of cardiovascular
diseases, cancer, and other disorders. This role depends on the length of the
carbon skeleton and on the number and the geometry of the double bonds.
The estimated effect of each individual dietary long-chain fatty acid on serum
cholesterol levels is summarized in Table 2.4.
Our original understanding of the specific fatty acids that alter levels of serum
cholesterols came from the pioneering work of Ahrens et al. (1957), Keys et al.
(1957), and Hegsted et al. (1965). Using human subjects, these investigators found
that the type of dietary fats can produce changes in serum cholesterol levels. They
also found that saturated fatty acids raised total cholesterol levels almost twice as
much as polyunsaturated fatty acids lowered them. In general, the risk of coronary
Table 2.4. Estimated Effects of Different

Dietary Fatty Acids on Serum Cholesterol Levels
Effect on Serum
Fatty Acid Cholesterol"
Saturated
Caprylic (8:0) ±o
Capric (10:0) ±o
Lauric (12.0) i
Myristic (14:0) it
Palmitic (16.0) i
Stearic (18.0) ±o
Monounsaturated
Oleic (18: I n-9; cis) 1
Elaidic (18: 1 n-9; trans) i
Polyunsaturated
Linoleic (18:2 n-6)
a-Linolenic (18:3 n-3)
Source: From Vessby (1994).

"When substituted for dietary carbohydrates.
Chemistry and Nutritional Value or Soybean Components / 35
heart disease (CHD) rises continuously as serum total and low-density lipoprotein
(LDL) cholesterol concentrations increase and falls with increasing levels of high-
density lipoprotein (HDL) cholesterol (Martin et al. 1986).
In the same study, Keys et al. (1957) reported that monounsaturated fatty
acids, mainly oleic acid, had no effect on serum cholesterol levels. This belief
was held for a long time even though Hegsted et al. (1965) showed that dietary
natural monounsaturated fatty acids lowered serum cholesterol levels to the same
approximated degree as polyunsaturated fatty acids in some diet trials. Twenty
years later, Mattson and Grundy (1985) confirmed the finding of Hegsted et al
(1965). This was followed by quite a few new studies (Mensink and Katan 1989,
Wardlaw and Snook 1990). It is now generally agreed that monounsaturated
fatty acid is as effective in reducing serum total and LDL cholesterol levels as
polyunsaturated fatty acids.
It has long been known that the major cholesterol-elevating fatty acids are the
saturated fatty acids with 12, 14, and 16 carbon atoms. For stearic acid with 18
carbon atoms, Bonanome and Grundy (1988) showed that it did not seem to
elevate the serum cholesterol concentrations compared with carbohydrates or
oleic acid. However, when compared with linoleic, stearic acid causes somewhat
lower HDL-cholesterol and higher LDL-cholesterollevels (Zock and Katan 1992).
3. Health Implications of trans Fatty Acids
Trans isomers are formed mainly during hydrogenation of polyunsaturated

fatty acids. One example is the trans isomer of oleic acid known as elaidic acid.
In the United States, many commercial margarines contain more trans fatty acids
than saturated ones. Tub margarines average about 17% trans isomers, and the
harder stick margarines somewhat more, around 25%. In contrast, butterfat con-
tains only 2-7% trans fatty acids, depending the cow's diet (Anonymous 1991).
The trans fatty acid level in the U.S. diet is estimated at 12.5-15.2 g/person per
day (Enig et al. 1990).
There has been a continuing controversy regarding the health effect of trans
isomers in dietary fats and oils. For a long time, trans fatty acids were considered
to be neutral with regard to their effects on serum cholesterol concentrations, in
spite of the fact that as early as 1961, Anderson et al. conducted one of the first
studies designed to look at the effect of hydrogenated fats on serum lipids and
found that trans isomers elevated serum cholesterol levels close to that of the
average of saturated fatty acids with chain lengths between C l l and CIK. In 1985,
the U.S. Food and Drug Administration (FDA) commissioned a full-scale review
of the possible health effects of trans fatty acids. Among the major scientific
studies examined, two concluded that the compounds raised serum cholesterol
levels; four found no effect; and two were inconclusive. The FDA thus concluded
that the substance did not pose a significant hazard, but the agency recommended
further research (Senti 1985).
Then Mensink and Katan (1990) looked at the effect of dietary trans fatty
acids on serum lipids with a large group of men and women. Their study is the
first to demonstrate that trans fatty acids not only raise serum LDL cholesterol
levels but also lower serum HDL cholesterol levels when compared with oleic
acid, and that their hypercholesterolemic effect was about half that of a mixture
of saturated fatty acids (CI2:0, CI4:0, and CI6:0). However, the amount of trans
fatty acids given in the study was quite high (11 % of daily energy intake), and
doubts have been voiced as to whether these findings could be extrapolated to
lower intake levels. Subsequent study in the same laboratory showed a linear
dose-response relationship between consumption of trans fatty acids and hyper-
cholesterolemic effect (Zock and Katan 1992).
New knowledge about the health implication of trans fatty acids has created
some concern among both the scientific community and the general public about
the safety of hydrogenated oils. Extreme positions have ranged from calling for
the elimination of trans fatty acids from the diet by avoiding specific foods, to
removing them from processed foods and/or including their amounts on food
labels, to denying any adverse effects or the need to modify food processing and
production, food labels, health claims, or intakes.
In response, the American Society for Clinical Nutrition (ASCN) and the
American Institute of Nutrition (AIN) have recently formed a task force on trans
fatty acids. Their position (ASCN/ AIN 1996) is: Consumption of trans fatty
acids in the United States has been relatively constant, and new food technologies
are yielding decreases in the trans fatty acid content of commercially prepared
foods. When intake of trans fatty acids (as hydrogenated fat) is compared with
that of saturated fat, total and LDL cholesterol levels in serum are lower, but
both trans fats and saturated fats increase total and LDL concentrations when
compared with cis fatty acids or native unhydrogenated fats. Since epidemiologic
data are conflicting with respect to cardiovascular disease outcomes, no conclusion
can be made that the intake of trans fatty acids is a risk factor for CHD. Nor
can it be expected that substituting trans- for cis-containing fats will reduce the
risk of CHD. Because the major contributors in the diet are fried or baked foods
and margarines in which partially hydrogenated vegetable oils may replace fat
sources richer in saturated fatty acids and cholesterol, the debate about trans
fatty acids should not detract from dietary recommendations to limit the intake
of saturated fat and total fat.
The North American branch of the International Life Sciences Institute (ILSI)
also formed its Expert Panel on Trans Fatty Acids and Coronary Heart Disease.
The critical review by the Panel, summarized in Kris-Etherton and Nicolosi
(1995), offered a similar view to that of the ASCN/AIN task force.
III. Proteins
The component present in the greatest amount in soybeans is protein, averaging

about 40% of total dry matter. Thus, it has been suggested that soybeans should
be called protein seeds rather than oilseeds. At present, unlike soy oil, which is
Chemistry and Nutritional Value of Soybean Components / 37
used mainly for human consumption, soy protein is used largely as feedstuff.
Only a small portion is for direct human consumption as either traditional soyfoods
or protein ingredients. Therefore, soy protein has been severely underutilized,
particularly in the Western world where use of whole soybeans has not been
accepted by the majority of the population and demands for animal meat require
a steady supply of soy meal. The situation is very unfortunate. Monographs on
chemistry and utilization of soy protein have been available (Smith and Circle
1972, Applewhite 1989). There are also a few review articles on the structure
and characteristics of soy protein in particular (Nielsen 1985a, 1985b, Murphy
1985) and of food proteins in general (Pernollet and Mosse 1983, Hettiarachchy
and Ziegler, 1994).
A. Protein Classification and Nomenclature
Unlike soy lipids, the fraction and classification of soy protein are rather complex.
Reports on such subjects do not always agree. Furthermore different nomenclature
systems regarding certain proteins or protein fractions are used in the literature.
These issues are partly due to the complexity of seed proteins themselves and
partly due to different methods used for extraction and isolation. Another reason
that has led to confusion in interpreting data from different laboratories might
be the lack of a purified form of protein fraction under study. Nevertheless, to
better understand soybean protein, we need to know how seed proteins in general
and soybean protein in particular are classified.
Based on biological function in plants, seed proteins are of two types: metabolic
proteins and storage proteins. Metabolic proteins include enzymatic and structural,
and are concerned in normal cellular activities, including the synthesis of the
second type. Storage proteins, together with reserves of oils, are synthesized
during soybean seed development. Following seed germination they provide a
source of nitrogen and carbon skeletons for the developing seedling. The majority
of soybean protein is storage protein.
Based on solubility patterns, legume seed proteins are di vided into albumins and
globulins. Albumins are soluble in water, whereas globulins are soluble in a salt
solution. Under this classification system, most soy protein is globulin. Globulins
in numerous legume species are further divided into two distinct types: legumin
and vicilin. Compared with vicilins, legumins have larger molecular size, less solu-
bility in salt solutions, and higher thermal stability. They also constitute a major
part of the total legume seed globulins. In many grain cereals, there exist a large
amount of ethanol-soluble fraction known as prolamins, while in rice a high volume
of a fraction can only be extracted with dilute acid/alkali solutions. Such fraction
is known as glutelins (Pernollet and Mosse 1983, Nielsen 1985a).
In addition, certain soy proteins have their trivial names. For example, the two
types oflegume globulins , legumins and vicilins, are commonly known as glycinin
and conglycinin in soybeans, respectively. These common names are apparently
derived from the genus name of soybean plant, Glycine. Others, particularly those
with enzymatic function, are based on the biological function of proteins them-
selves. Examples include hemagglutinin, trypsin inhibitors, and lipoxygenases.
Although classification based on solubility in a solvent series provided a
convenient means to compare the proteins from various seeds, there was no
assurance that the same polypeptides would not be extracted into more than one
solubility class due to their association with other proteins. A more precise means
of identifying proteins has been based on approximate sedimentation coefficients
using ultracentrifugation to separate seed proteins (Naismith 1955, Wolf and
Briggs 1959, Thanh and Shibasaki I 976a, Howard et al. 1983). Under appropriate
buffer conditions, soy protein exhibits four fractions after ultracentrifugation.
These fractions are designated as 2, 7, 11, and ISS (Fig. 2.5, at an ionic strength
of 0.5). Here S stands for Svedburg unit. It is computed as the rate of sedimenta-
tion per unit field of centrifugal strength based on the following equation:
where c is the distance from the center of the centrifuge, t is time, and w is
angular velocity. The value for S ranges between I and 200, with a unit of 10- 13 sec.
In general, analysis of the four fractions of soybean proteins based on ultracen-
trifugation has shown that lIS and 15S fractions are pure proteins. More specifi-
cally, the lIS fraction is the soybean glycinin and accounts for at least V3 of
extractable protein, whereas the ISS fraction is thought to be a polymer of
glyeinin and accounts for about 10% of extractable protein. In contrast, the 2S
and 7S fractions are heterogeneous. The 2S fraction accounts for about 20% of
the extractable protein and includes the Kunitz and Bowman-Birk trypsin inhibi-
tors and cytochrome C. The 7S fraction accounts for an additional third of the
extractable protein and consists of eonglycinin, a-amylase, lipoxygenase, and
hemagglutinin (Nielsen 1985a).
The II S fraction and glycinin refer to the same soybean globulin, whereas
the 7S globulin and ~-conglycinin refer to another soy protein. However, strictly
J.I =0.5 J.I = 0.1
7S Globulin .. 9S IDimer 01 7S )
Figure 2.5. Effect of ionic strength on the ultracentrifuge pattern of water-extractable

proteins from soybeans at pH 7.6. From Wolf (1969).
Chemistry alld Nutritional Value of Soybean Components / 39
speaking, ~-conglycinin is only a component of the 7S fraction. It should also

be emphasized that the terms, 2S, 7S, lIS, and 15S are only for the purpose of
classification and identification. By no means do they imply that sedimentation
constants are always these exact whole numbers among different studies. In fact,
these constants as well as separation patterns of different fractions depend largely
on conditions of buffer composition, pH, and other factors. As shown in Figure
2.5, a portion of the 7S fraction observed at pH 7.6, 0 .5 ionic strength, dimerizes
at 0.1 ionic strength to form a new peak named 9S. Because of the shortcoming
of this system, the Oilseeds Division of American Association of Cereal Chemists
once formed a Soybean Protein Nomenclature Committee in 1967 (Wolf 1969).
Unfortunately, no consensus on a comprehensive system was made before the
committee disbanded.
B. Isolation of Major Storage Proteins
Localized in organelles known as protein bodies, storage proteins constitute the

major part of total seed proteins. Legume storage proteins are large multimeric
molecules of at least two main types : vicilin and legumin, each of which consists
of a family of closely related molecules. They contain more amide and arginine
and less sulphur amino acid residues than the average metabolic protein, although
they have an acid isoelectric point. In soybeans, the two main types of storage
proteins are ~-conglycinin and glycinin and account for 65-80% of total seed
proteins. Therefore, discussion in this section is limited to these two proteins.
1. Protein Bodies
As shown in Figure 2.1, bound by a single membrane, protein bodies are more
or less spherical. The size of soybean protein bodies range from 2- 20 11m but
many fall into the narrow range of about 5-8 11m. A comprehensive review on
protein bodies in plant seeds can be found in Pernollet ( 1978).
Saio and Watanabe (1966) homogenized soybeans in cottonseed oil and then
separated the protein bodies by differential centrifugation in cottonseed oil-carbon
tetrachloride mixture. In contrast, Tombs (1967) isolated protein bodies from
defatted soybean flour (350 mesh) by first extracting with 20% (w/v) sucrose
solution containing 50 mM citrate (pH 5), followed by sucrose density gradient
centrifugation. In the same study, Tombs (1967) fractionated protein bodies into
light and heavy fractions. The former contained 97.5% protein and the latter
78.5%. RNA, phytic acid , and lipids were also found in the protein bodies.
Regardless of the isolation method, caution should be made to avoid breakage
of protein bodies during their isolation, because soybean protein bodies are
extremely fragile and readily lyse when cotyledon cells are ruptured in water.
On an individual protein basis, proteins in the protein bodies are mainly glycinin
and conglycinin. Other proteins found in protein bodies include lectin, trypsin inhib-
itors, and a large number of other unidentified polypeptides (Lei and Reeck 1987).
2. Isolation Procedures
Extraction by dilute salt solutions and separation of soybean proteins using

ultracentrifugation was first reported by Naismith (1955). Shortly after that, Wolf
and colleagues made an extensive study to extract, fractionate, and characterize
soy proteins (Wolf and Briggs 1959, Wolf et al. 1961). More extensive research
was done by many later investigators using such separation techniques as gel
electrophoresis (Saio et al. 1986), column chromatography (Thanh et al. 1975a)
and their combination (Thanh et al. 1975b).
The most popular method is the procedure described by Thanh and Shibasaki
(1976a) for a simultaneous preparation of both 7S and lIS globulins (Fig. 2.6).
Briefly, defatted soybean meal is extracted with O.03M (or 0.06M) Tris [tris(hy-
droxymethyl)aminomethane] buffer (pH 8.0) containing O.OlM mercaptoethanol
at room temperature for 1 hr. The extract is adjusted to pH 6.4 with 2N Hel.
The lIS globulin is collected by centrifugation and dialyzing the extract at this
pH for 3 hr in the cold. A crude 7S globulin is separated from the supernatant
Defatted soy meal

I Extract w"h 0.03 M Tric-HCI (pH, 8)
~ containing 0.01 M mercaptoethanol
Whole buffer extract
ft ~a."",,2N Hel
Precipitate Supernatant
(Crude 115 fraction) (Crude 75 fraction + whey protein)
Adjust pH to 4.8
Precipitate
/ ~
Supernatant
Dissolve In O.03M Trls-HCI (Whey protein)
buffer, adjust pH to 6.2
Precipitate Supernatant
(Polymerized form) (75 fraction)
Figure 2.6. Simultaneous isolation of the soybean 7S and lIS globulins. Adapted from
Thanh and Shibasaki (1976a).
by adjusting to pH 4.8. The prepared 7S globulin is washed with pH 6.2 Tris

buffer and then dispersed in the standard buffer (protein concentration, 2-3%)
by adding 2N NaOH drop wise to a pH 7.6 before adjusted back to pH 6.2. The
solution is kept at 3-5°C overnight. A trace of precipitate is finally removed by
centrifugation. Later workers found that the crude glycinin (11 S) is about 90-95%
pure (Moreira et al. 1979).
Known as fractional isoelectric precipitation, the aforementioned method is
based on the different solubilities of 7S and 11 S globulin in dilute Tris buffers
(0.03-0.06M) in the pH region of 6.1-6.6 (Fig. 2.7). The 7S globulin precipitates
at pH values between 4.0 and 5.6, whereas the liS globulin precipitates between
4.4 and 6.8. Therefore, in the pH 6.1-6.6 region, the 7S globulin dissolves while
most of the 11 S globulin precipitates. The optimum pH for their resolution
is 6.2-6.4.
One advantage of Thanh and Shibasaki's method is that it affords large-scale
isolation of the two major proteins without significant cross-contamination. Two
---
/1 \
0.4
. .....\
E
c
0
0
co
0.3
// \ \.
\
~
w 118
() 0.2 " 78
z
«
/
CD
a: "
0
\
(J)
CD
« "
0.1
/' . •
0
//"/1
",.
\
'\
'-.
~ .~
•
3 4 5 6 7 8
pH
Figure 2.7. pH-dependent precipitation curves of the 7S and lIS globulins in 0.06 M
Tris-HCl buffer, as measured by turbidity of protein solutions with a concentration of
0.02%. From Thanh and Shibasaki (l976a).
similar procedures suitable for commercial processing were later developed and
patented. One involves extraction of protein from defatted soy flakes at pH 8
with an aqueous solution of 0.03-0.06 M sodium chloride and 0.5-0.8 mM
sodium bisulfite. The classified extract was adjusted to pH 5.5 with Hel and the
precipitated protein separated by centrifugation. The supernatant was adjusted
to pH 4.5 and the additional precipitated protein was separated from whey protein
and sugars. The proteins that precipitated above pH 5.5 consist of about 90%
liS, whereas the remaining proteins that precipitate at 4.5 consist of about 70%
7S and 30% llS fractions (Howard et al. 1983).
The second procedure, described by Lehnhardt et al. (1983), invol ves extraction
of defatted soy flakes at pH 8.0, clarification of the extract, precipitation of the
protein at pH 4.3 with HCI, and separation of the protein curd from whey protein
and sugars. The curd was resuspended in an aqueous medium of 0.1 M sodium
chloride and 7.5 mM sodium bisulfite. The suspension was brought to pH 5.3
with NaOH. At this pH, the 7S globulin dissolved and the remaining insoluble
protein is comprised of mostly lIS.
The proteins that contaminate crude glycinin or f3-conglycinin may be removed
by a variety of techniques including fractional ammonium sulfate precipitation,
chromatography with hydroxylapatite, gel filtration, conconavalin A affinity chro-
matography, diethylaminoethyl (DEAE)-anion exchange chromatography, and
centrifugation in sucrose gradients (Nielsen 1985a). Because of the different
methods used, reports by different researchers on the characterization of soy
protein fractions are sometimes controversial. In addition, the association-
dissociation phenomenon observed under different experimental conditions (Fig.
2.5, Catsimpoolas et al. 1969, Thanh and Shibasaki 1979) contributes to the
complexity of soy protein isolation and characterization.
C. Characterization of Major Storage Proteins
1. f3-Conglycinin (7S Globulin)
In the literature, reference is made to a-, 13-, and 'T-conglycinins. These terms
were coined by Catsimpoolas and Ekenstam (1969) to refer to three immunologi-
cally distinct fractions separated from a crude conglycinin preparation. However,
a subsequent report by Catsimpoolas (1969) showed that a-conglycinin contains
enzymatic activities characteristic of 2S fractions. 13- and 'T-Conglycinins contain
no enzymatic activities and are differentiated from one another on the basis of
their capacity to undergo reversible polymerization at neutral pH in response to
lowering ionic strength from 0.5 to 0.1 (Catsimpoolas 1969); f3-conglycinin
exhibits a sedimentation coefficient of about 7S at high ionic strength and one
of 9-lOS at the lower ion concentration, whereas 'T-conglycinin fails to undergo
this change. In addition, based on the method offractional isoelectric precipitation,
Thanh and Shibasaki (l976a) reported that the former was precipitated at pH 4.8
whereas the latter remained soluble. Because of its highest proportion in the 7S
Chell1istrr and Nutritional Value of Sovheal1 Components / 43
fraction, l3-conglycinin has been studied extensively. The other two components,
a- and "t-conglycinins, have received scant attention since then.
It is now generally agreed that l3-conglycinin is a trimer with a molecular
weight of about 180 kilo daltons (kDa). It has three prevalent types of subunits,
designated as a', a, and 13. The individual subunits can be resolved from one
another by ion-exchange chromatography after denaturation of the native complex
with urea (Thanh and Shibasaki 1976a) or by sodium dodecyl sulfate (SDS)-
polyacrylamide gel electrophoresis (PAGE) (Fig. 2.8). Based on the electropho-
retic mobility of the second method, the molecular weights (MW) of a', a, 13
subunits were estimated at 57. 57, and 42 kDa, respectively (Thanh and Shiba-
saki 1976b).
Another subunit, named 13', is present only in some soybean varieties (Coates
et al. 1985, Morita et al. 1996). The primary structure of the 13' subunit is still
unknown, but the amino acid composition suggested that this subunit is rich in
sulfur-containing amino acids (Coates et al. 1985). The characteristics is very
interesting from both points of the nutritional value of soy proteins and of the
physico-chemical properties of the food products. The second point is related to
polymerization of the 13' subunit through disulfide bonding.
+
a'
u
Figure 2.8. Electrophoretic separation of soybean seed extracts in SDS-gels with (right-
gel) and without (left gel) urea. Bands labeled a', a, and ~ are subunits of ~-conglycinin.
Other identified bands are acidic (An) and basic (Bn) polypeptides of glycinin. From Fontes
et al. (1984).
All three major subunits are rich in aspartate/asparagine, glutamate/glutamine,

leucine and arginine. The two subunits, a and a', are very similar in amino acid
composition. Both are devoid of cystein and have low levels of methionine. In
contrast, [3 subunit contains no methionine (Nielsen 1985a). In addition, all of
the [3-conglycinin subunits are glycoproteins and contain 4-5% carbohydrate.
Thus the 7S globulin is considered to be glycosylated (Pernollet and Mosse 1983).
[3-Conglycinin is heterogeneous with regard to subunit composition. Among
the 10 theoretically possible multiple forms, six (B j to B 6) have been demonstrated
to occur (Thanh and Shibasaki 1976b). The subunit compositions of the six
components are as follows: B j is composed of la' and 2[3; B2 , 1a and 2[3; B3 ,
1a, 1a' and 1[3; B4 , 2a and 1[3; B s, 2a and 1a'; and B 6 , 30.. Furthermore, [3-
conglycinin undergoes a complicated association-dissociation phenomenon in
response to changes in ionic strength and pH. The phenomenon was initially
reported by Naismith (1955) and then confirmed by many other workers. Accord-
ing to Thanh and Shibasaki (1979), the six forms of trimers are able to reversibly
dimerize at low ionic strength or in the pH region of 4.8-11.0. The resulting 9S
form is a superdimer of two trimers facing each other and becomes a hexamer.
At extreme pH, 2.0 or 12.0, dissociation into subunits (a, a', and (3) are also
reversible so that the six molecular species B j -B 6 can be reconstituted by mixing
the three subunits in urea solution and subsequently dialyzing the solution against
a phosphate buffer. Subsequent studies by other workers led to the identification
of the Bo form of [3-conglycinin. It contains three [3 subunits (Sykes and Gayler
1981, Yamauchi et al. 1981).
Most recently, Morita et al. (1996) succeeded in isolation and purification of
4 major molecular components of [3-conglycinin, 3a, 2al[3, 1~2[3, and 3[3, from
seeds of an a' subunit-deficient soybean strain. By using such cultivars with a
simple subunit composition and chromatography at intermediate ionic strengths
over a period sufficient to reach an effective dissociation equilibrium between
monomer and dimer, they also succeeded in crystallization of single [3-conglycinin
components. Consequently, they were able to compare 3a and 3[3 components
regarding their secondary structure and found that the 3a component showed a
higher a-helix content than 3[3. Moreover, they also found two novel components
from the same seeds: 2[31[3' and 2[31des(V1-RI26)a. The former contained a
minor subunit [3' pointed out by Coates et al. (1985), and the latter contained a
subcomponent which would be derived artificially from limited hydrolysis of the
a-subunit by a novel serine protease found previously in some soybean strains
(Morita et al. 1994). The proteolysis would occur in the course of extraction
and purification.
2. G/ycinin (llS Fraction)
Glycinin is the purified form of the 11 S globulin. It is the largest single fraction
of total seed protein (25-35%) and accounts over 40% of the total seed globulin
Chemistry and Nutritional Value of'Soybean Components / 45
URE! \~~~A
(A-SS-B)6
6(A-SS-8) • 6(A) + 6(8)

13ME
60K 40K 20K
Figure 2.9. Mechanisms for disassembly of soybean glycinin into subunits and further
into polypeptides. 13ME: 13-mercaptoethanol. From Nielsen (l985b).
(Murphy and Resurrection 1984). In contrast to ~-conglycinin, only a small

portion of glycinin is glycosylated (Lei and Reeck (1987).
The currently accepted model of glycinin is a hexamer with a MW of about
360 kDa. Its monomeric subunits have the generalized structure A-S-S-B, where
A represents an acidic polypeptide of 34-44 kDa; B is a basic polypeptide of
about 20 kDa; and S-S is a single disulfide bond that links the two polypeptides.
The subunits and polypeptides can be disassembled by the use of urea plus a
disulfide reductant such as ~-mercaptoethanol (Fig. 2.9). The reduced and dena-
tured acidic polypeptide components can be resolved by chromatography with
DEAE-Sephadex, whereas the basic components can be separated by chromatog-
raphy with CM-Sephadex (Moreira et a1. 1979). Some of peptide components
of ~-conglycinin, together with other seed proteins or subunits, can also be
separated by SDS-PAGE (Fig. 2.8).
Five major subunits have been characterized (Moreira et a1. 1979, Staswick
et a1. 1981), namely A 1aB 2, AlbB lb , A 2B la , A 3B4, and AsA4B3' Based on physical
properties, these subunits can be separated into two distinct groups. Group I
includes the first three subunits whereas Group II includes the remaining two
subunits. The Group II subunits have more uniform apparent MW and contain
more methionine than members of Group II (Table 2.5). This last feature is
important for breeders to increase the methionine content in seeds.
Table 2.5. Five Major Soybean Glycinin Subunits and Their Paired Acidic and
Basic Polypeptides
Subunit Mol. wt. No. of
Group Subunit Structure (KD) Methionine
Gl AhB, 58 5-6
G2 AlbBlb 58 5-6
I G3 A,B 1, 58 7-8
II G4 AlB 4 62 3
II G5 A,A4B 3 69 3
Source: Data adapted from Nielsen (l985a).

Among the five major subunits, the GS subunit (AsA4B3) is an exception from
the general structure, Its acidic component consists of two polypeptides, AsA4,
and contains a proteolytic cleavage site about 100 amino acids from the NHT
terminal end of the precursor polypeptide. Consequently, the A4-component sepa-
rates from disulfide linked AsB3 component upon denaturation. In addition, the
nomenclature and subunit assignments given in Table 2.S were developed based
on peptide components in glycinin subunits from cultivar CX63S-I-I-I. However,
the Raiden cultivar does not appear to contain GS subunit but instead contains
an acidic polypeptide designated A6 (Staswick and Nielsen 1983). Since then,
little is known about its amino acid sequence and the identification of its basic
polypeptide partner.
Nevertheless, there is a genetic polymorphism of lIS globulin in soybeans.
The lIS globulin of ordinary varieties can be classified into two types, As- and
A4-types (Kitamura et al. 1980) (also known as A4 and A3 types, respectively,
by Staswick and Nielsen 1983), depending on the presence or absence of the
AsA4 acidic polypeptide component and its paired basic polypeptide. The absence
of the subunit is controlled by a single recessive allele (Harada et al. 1983).
Almost all the U.S. varieties examined to date contain all five lIS subunits while
about 20% of the Japanese varieties lacked the AsA4B3 subunit (Kitamura 1995).
Comparison of the amino acid sequence of five major subunits revealed that the
acidic and basic polypeptides each exhibited considerable NH,-terminal sequence
homology (Moreira et al. 1979). There is also a high degree of homology in the
interior portion of the acidic polypeptides. Regarding sulfhydryl (-SH) content
of glycinin, Wolf (1993) reported that natural glycinin contained 0.62-2.2 -SH
groups/mol with a mean value of 1.4 ± O.S. When reduced with ~-mercaptoethanol
(ME) in the absence of denaturants, the -SH content increased and leveled off
at 21 -SH groups/mol in the range of ME concentration of 0.04-0.1 M. Apparently,
there was a large discrepancy in the number of -SH groups measured vs. the
number of predicted from the current model of glycinin structure.
The secondary structure of glycinin was predicted to be 25% a-helices, 2S%
~-sheet, 42% turns, and 8% unordered forms on the basis of its amino acid
sequence by modeling (Argos et al. 1985). Recently, Abbott el al. (1996) reported
Fourier transform infrared (FTIR) spectra of glycinin as 24O/C a-helical, 30% ~
sheet, 31 % turns, and 12% unordered. Their data also indicate that glycinin has
the same secondary structure in solution and in hydrated solids. Nothing is
known about the tertiary structure of glycinin. However, glycinin has a complex
quaternary structure consisting of two layers of trimers. Each trimer has three
acidic and three basic polypeptides paired and held together by disulfide and
hydrogen bonds, with acidic and basic peptides alternating (Badley et al. 1975).
These bonds can be disrupted by urea, strong acid, strong base, heat, or sodium
dodecylsulfate in combination with a disulfide reducing agent. As a result, the
quaternary structure is altered.
Chemistry and Nutritional Value of Sovbean Components / 47
3. Differences between 75 and 115 Globulins
Due to differences in composition and structure, the two major soybean globu-
lins, ~-conglycinin (7S) and glycinin (lIS), exhibit differences in both nutritional
quality and functional properties. In general, the 11 S globulin contains 3-4 times
more methionine and cysteine per unit protein than 7S protein (Kitamura 1995).
Because soybean protein is generally deficient in these sulfur-containing amino
acids, the II S protein becomes more valuable from a nutritional point of view.
The two globulins also show considerable differences in key functional proper-
ties, including gel-making ability, thermal stability, and emulsifying capacity
(Yamauchi et al. 1991). In general, the lIS protein has a better gel formation
ability than the 7S globulin. On the other hand, the 7S protein has a greater
emulsifying capacity and emulsion stability than lIS globulin. Furthermore, the
presence or absence of the A5A4B3 subunit in glycinin has been shown to exert
significant effects on the gelling properties of soymilk and tofu gel hardness
(Murasawa et al. 1991); it is easier to make nigari tofu with a smoother and
more uniform gel using the soymilk lacking the subunit.
Both II Sand 7S proteins form gels when induced by heat and/or a coagulant
as in tofu making. In the heat-induced gel formation, Utsumi and Kinsella (1985)
reported that when gels were made by heating at 80°C for 30 min the 7S gels
were harder than I I S gels. However, according to German et al. (1982), the
denaturation temperature of 7S protein is lower than that of II S. In other words,
the 11 S fraction requires higher heating temperature to form a gel than 7S
globulin. Thus, the heating temperature of 80°C in the study by Utsumi and
Kinsella (1985) may be too low for 11 S to form a gel with consistency, resulting
in stronger 7S and weaker 11 S gels. Also, Nakamura et al. (1986) found that
there was a change with heating time in hardness of gels made from 7S and 11 S
proteins. When solutions were heated at 100°C for a short time «5 min), 7S
gels were harder than I I S, but a long heating time gave opposite results.
In the presence of calcium sulfate. II S protein coagulates faster and forms
larger aggregates than the 7S fraction. More important is that the I I S gel is
harder than the 7S gel. It also has a higher water-holding capacity and higher
tensile values, expands more on heating, and is more sensitive to the softening
effect of phytic acid, as compared with the 7S gel (Saio et al. 1969, Hashizume
et al. 1975). When calcium sulfate was replaced with glucono-8-lactone (GDL).
similar differences between lIS and 7S proteins and their gel properties were
also found (Fig. 2.10).
Many studies with proteins have now identified relationships between the
structure of a protein and its functional properties. These include:
I. Heat instability of the constituent subunits of a protein, such as glycinin,

is related to the heat-induced gel-forming ability.
H,,?o11S
7r---=-------------------------------~
~ 6
75
O~;~~~~--~-L--~~--~~~--~~
o 60 120 180 240
time / min
Figure 2,10, Heating time dependence of the breaking stress for 7S and lIS soy protein.
Concentrations of protein and glucono-8-lactone are 4.0% and 0.4%, respectively. Heating
temperature: 60°C; sample size: 16 mm diameter x 10 mm; compression rate, 1.0 mmls; test
temperature, 27°C. Bars attached to symbols represent the standard deviation. From
Kohyama and Nishinari (1993).
2. Hydrophobicity is an important factor in the emulsifying properties.

3. The surface properties of a protein depend on the confonnational stabil-
ity-the more unstable, the higher the emulsifying properties.
Furthermore, in the formation of heat-induced gel, disulfide exchange plays an
important role. The number and topology of free sulfhydryl residue are closely
related to the heat-induced gel-forming ability and the properties of gel, but not
to its emulsifying properties (Utsumi et al. 1993). Our understanding of such
relationships has made it possible to improve the functional properties and nutri-
tional values of soybean proteins by protein engineering (Kim et al. 1990, Utsumi
et al. 1993, Nonaka et al. 1994) and genetic improvement (Kitamura 1995).
D. Trypsin Inhibitors
Protease inhibitors are substances which, when added to a mixture of a protease
(such as trypsin or chymotrypsin) and a substrate, bind to the enzyme and produce
a decrease in the rate of substrate cleavage. Protease inhibitors of a protein nature
are ubiquitous. All but one of these inhibitors act on serine proteases, which use
the hydroxyl group of a serine residue in their catalytic sites during cleavage action
toward protein substrates (Read and Jame 1986). Trypsin and chymotrypsin, the
two major proteolytic enzymes of the pancreas, all belong to the serine proteases.
1. Types of Inhibitors in Soybeans
The protein proteinase inhibitors that have been isolated from soybeans are
of two types: the Kunitz trypsin inhibitor (TI) and the Bowman-Birk (BB) inhibi-
Chemistry and Nutritional Value oj Soybean Components / 49
tor. The Kunitz inhibitor was first isolated and crystallized by Kunitz (1945).
The isolation involves extracting soybeans with water and precipitating the inhibi-
tor with alcohol. It has an MW between 20 and 25 kDa, with a specificity directed
primarily toward trypsin. The inhibitor was shown to combine tightly with trypsin
in a stoichiometric fashion, i.e., 1 mole of the inhibitor inactivates 1 mole of
trypsin. The complete amino acid sequence of the inhibitor was established by
Koide et al. (1973). It consists of 181 amino acid residues and two disulfide
bonds, with a reactive site at residues Arg 63 and Ile 64 •
The soybean BB inhibitor was first described by Bowman (1944) as an acetone-
insoluble factor in contrast to the alcohol-insoluble Kunitz inhibitor. Isolation
involves extracting beans with 60% alcohol solution and precipitating the inhibitor
with acetone. Later, Birk (1961) resumed investigation of the acetone-insoluble
factor and succeeded in purifying and characterizing the inhibitor. Cumbersome
descriptive terms have been used in the literature to refer to the BB inhibitor:
acetone-insoluble factor, purified AA inhibitor, and trypsin and chymotrypsin in-
hibitor.
The complete amino acid sequence of BB inhibitor was determined by Odani
and Ikenaka (1973). It is a single polypeptide chain of71 amino acids including
seven disulfide bonds. Its MW is about 8 kDa. The BB inhibitor is capable of
inhibiting both trypsin and chymotrypsin at independent reactive sites, the trypsin
reactive site being located at residues Lysl6 and Ser17 and the chymotrypsin
reactive site being at the residues Leu44 and Ser5. The conformation (secondary
structure) of BB inhibitor has been reported by a number of investigators. Steiner
and Frattali (1969) concluded from a circular dichroism study that BB inhibitor
had Iowa-helical structure. Recently Wu and Sessa (1994) found that BB inhibitor
has 61 % f)-sheet, 38% unordered form, I % f)-turn, and 0% a-helical form. Their
data also suggested that BB inhibitor has a stable conformation even after disulfide
bonds are broken by heating.
In addition to protein protease inhibitors, soybeans contain trace amounts of
free fatty acids and their acyl CoA esters, which have been reported to inhibit
trypsin (Liu et al. 1990). These nonprotein types of trypsin inhibitors have been
shown to be responsible for increased trypsin inhibition in fermented soybeans
(Wang et al. 1975, Liu and Markakis 1991). They were also shown to be responsi-
ble for trypsin inhibitory activity of an 85% aqueous ethanol extract of soybeans,
referred to as a calcium-sensitive inhibitor (Liu and Markakis 1991). However,
unlike trypsin inhibitors of protein nature, these nonprotein trypsin inhibitors
lack specificity and are very susceptible to cation suppression (Liu et al. 1990),
therefore, their physiological significance is assumed to be negligible.
2. Health Implications
The significance of soybean trypsin inhibitors lies in their nutritional implica-

tions for both humans and animals. As early as 1917, Osborne and Mendel
observed that soybean meal had to be heated in order to support the growth of
rats, An assumption is that trypsin inhibitors present in soybeans are responsible
for growth depression by reducing the digestibility of proteins. However, the
issue was found to be not that simple when it was reported that soybean diets
containing predigested protein or free amino acids still retarded the growth of
rats (Desikachar and De 1947, Liener et al. 1949). This observation indicates
that inhibition of proteolysis by trypsin inhibitors was not the sole factor responsi-
ble for growth depression. Furthermore, rats fed a raw soybean extract from
which trypsin inhibitors had been removed showed improved growth performance
when compared with control rats fed diets containing raw soybeans from which
inhibitors had not been removed. However, only about 40% of the growth inhibi-
tion produced by raw soybeans could be attributed to the presence of trypsin
inhibitors (Kakade et al. 1973).
Another significant finding was that raw soybeans as well as trypsin inhibitors
themselves could cause hypertrophy of the pancreas of chicks (Chernick et al.
1948). Since the pancreas is responsible for the production of most enzymes
required for the digestion of food, dietary components that affect pancreatic
function could markedly influence the availability of nutrients from the diet. The
mechanism whereby the trypsin inhibitor induces pancreatic enlargement is still
not fully understood. Green and Lyman (1972) showed that pancreatic enzyme
secretion in rats is controlled by a negative feedback mechanism. The amount of
pancreatic secretion is determined by the level of free trypsin and/or chymotrypsin
present in the intestine. As the level of trypsin goes below a threshold level, the
pancreas is induced to produce more enzymes. The mediating agent between the
enzymes and the pancreas is the hormone cholecystokinin (CCK), which is
released from the jejunal endocrine cells when the level of trypsin or chymotrypsin
in the intestine becomes depleted. The dietary TI evokes increased pancreatic
enzyme secretion by forming inactive trypsin-TI complex. This leads to endoge-
nous loss of essential amino acids being secreted by a hyperactive pancreas. The
loss of methionine and cysteine in this way could be particularly acute since
soybean protein is low in these amino acids. A later study undertaken by Liener
et al. (1988) confirmed the existence of feedback control of pancreatic enzyme
secretion in humans.
A recent study with 6 human subjects showed that unlike soybean BB inhibitor
installation of soybean Kunitz inhibitor caused an increase in trypsin and the
pancreatic secretory trypsin inhibitor without changes in plasma CCK. This
observation indicates that: (1) pancreatic exocrine secretion oftrypsin and chymo-
trypsin in humans is regulated by different mechanisms, and (2) although the
inhibitors have similar in vitro inhibition patterns, their in vivo effects are different
Reseland et al. (1996).
Much controversy has arisen in recent years regarding the physiological roles
of protease inhibitors as medical research demonstrates that protease inhibitors
have the ability to serve as cancer chemopreventive agents in both in vitro and
Chemistry and Nutritional Vulue or Sovhean Components / 51
in vivo systems (Kennedy 1993). At least one inhibitor in soybeans, the BB

inhibitor. has been shown to have clear anticarcinogenic activity in both in vitro
and in vivo carcinogenesis assay systems (Kennedy 1994). Unlike most of the
other potential classes of cancer chemopreventive agents that have been studied,
protease inhibitors have the ability to affect many different kinds of carcinogenic
processes in an irreversible manner. Also unlike most other agents, they are
effective at extremely low levels. Therefore, even though the mechanism of
action of protease inhibitors in the prevention of cancer is not yet clear, it is
clear that the protease inhibitors are powerful anticarcinogenic agents.
3. Elimination
Although the nutritional effect of protein protease inhibitors on humans is the

subject of much controversy, it is generally agreed that growth inhibition and its
accompanying pancreatic hypertrophy, hyperplasia, and adenoma in rats or other
animals fed with soybeans and other legumes containing active trypsin inhibitors
are partly or fully due to the antitryptic activity of the inhibitors (Liener and
Kakade 1980, Gumbmann et al. 1986, Liener 1994). As shown in rats (Fig. 2.11),
the degree of TI destruction parallels improvement in the nutritive value of soy
proteins measured by protein efficiency ratio (PER) assay, although destruction
of other anti nutrients by heat may also be responsible for nutritional improvement.
Extensive efforts have been made to inactivate trypsin inhibitors from soybeans
and other legume seeds. Many approaches have been based largely on heat
treatment, including live steam (Liu and Markakis 1987, Anderson 1992), boiling
in water (Liu and Markakis 1987), dry roasting (McNiven et al. 1992), dielectric
heating (Borchers et al. 1972), microwave radiation (Pour-EI et al. 1981), microni-
C. 100
\ /0
E /e _ _ _ _ - 3.1 a:::
"- w
:J
f- 80 (PER) ° ~
27 .2
>~x'"
~ "0
:~ 60 a:::
U 2.3 u
>-
ct c
2
--
CD
'0
40
:ii 1.9 w
:cc V· 0 c
'(j)
c 20 - ............ 0
'w 0...
a.
>-
~o- 1.5 a...
~ 1
0 2 4 6 10 20
Minutes at 100°C
Figure 2.11. Effect of steaming on trypsin inhibitor activity and protein efficiency ratio
(PER) of soy meal. From Rackis (1974).
52 I Soybeans: Chemistry, Technology, and Utilization
zation (Hutton and Foxcroft 1975), and extrusion cooking (Peters and Czakor
1989).
In addition to heating temperature and time, the moisture condition prior to
and during heat treatment has a significant effect on the effectiveness of TI
destruction by heat. For example, cooking whole soybeans reduces trypsin inhibi-
tor activity to about 15% of that in raw beans. However, for complete removal
of trypsin inhibitors, soaking prior to cooking is necessary, even though soaking
has no effects on TI activity (Liu and Markakis 1987).
Based on the activity loss of the purified inhibitors, the Kunitz inhibitor was
thought to be more heat labile than the BB inhibitor (Birk 1961). However,
DiPietro and Liener (1989) demonstrated that an in situ BB inhibitor is inactivated
at a faster rate than the Kunitz inhibitor upon heating.
Heat treatment reduces not only TI activities, but also solubility of the whole
seed protein (Anderson 1992). More importantly, excessive heat treatment can
cause loss of essential amino acids in soy protein (Rios-Iriarte and Barnes 1966,
Skrede and Krogdahl 1985). Therefore, in applying heat to soy products, it is
essential to use an optimum condition (temperature, time, moisture, and pressure)
to maximize destruction of TI and at the same time to minimize reduction of
soy protein solubility as well as loss of essential amino acids. However, this is
easier to say than do. In fact, the amount of heat required to eliminate growth
inhibitors in raw soybeans is sufficient to destroy cystine to make this amino
acid the first limiting (Rios-Iriarte and Barnes 1966). In actual situations, heat
treatments do not completely inactivate all inhibitor activity (Table 2.6). The
possible adverse effects of residual inhibitors in soy products are largely unknown.
Adjunct treatment with various chemicals, including various thiol-containing
compounds (such as cysteine, N-acetyl-cysteine, and glutathione) and sodium
sulfite has been found to facilitate inactivation at lower temperatures (Liener
1994). Friedman and Gumbmann (1986) reported that the treatment of raw soy
flour at 75°C with O.03M sodium sulfite for 1 hr completely inactivated trypsin
Table 2.6. Trypsin Inhibitor Activity (Average Value ± Standard Deviation) of Raw
and Cooked Soybeans and Some Commercial Soy Protein Products
Trypsin Inhibitory Activity Percentage of
Product TVVmg Raw Soybeans
Raw soybeans 171.0±3.4 100.0

Boiled soybeans (20 min) 24.3±1.1 14.2
Soy protein concentrate 48.9±1.8 28.6
Soy protein isolate I 23.9±1.1 14.0
Soy protein isolate II 32.1±0.6 18.8
Source: Data adapted from Liu and Markakis (l989b).

Note: TVI = trypsin units inhibited, where I TV is defined as 0.01 Absorbance at 410 nm under
the assay condition defined in the text (4 mL assay volume, pH 8.1, at 37°C for 10 min reaction,
with porcine trypsin).
inhibitors, leaving no sulfite residue in the soy proteins. Their rat feeding tests
showed that sulfite treatment is better than heat treatment alone in terms of
nutritional improvement. However, any treatment of foods with chemicals should
be viewed with caution with respect to regulatory issues.
Another alternative approach in lowering trypsin inhibitors from soybeans is
plant breeding. Already a few genetic varieties of soybeans that lack the Kunitz
trypsin inhibitor have been reported (Orf and Hymowitz 1979, McNiven et al.
1992). These lines have lower trypsin inhibitor activity in comparison with normal
soybean genotypes. Consequently, compared with normal lines, soybean isolines
devoid of the Kunitz inhibitor not only require shorter heating times to inactivate
TI activity (McNiven et al. 1992), but also support better growth of rats (Friedman
et al. 1991).
4. Assay Methodology
Because TI inactivation by heat treatments parallels the improvement in nutri-

tive value, the TI activity becomes an important parameter in the quality of
legume-based food and feedstuffs. The determination ofTI levels in these products
has been carried out by a variety of methods, but many are colorimetric. An
original procedure involved incubation of serial levels (including 0 level) of a
dilute alkali extract of soy samples with trypsin, initiation of the reaction by
adding a synthetic substrate, benzoyl-DL-arginine-p-nitroanilide hydrochloride
(BAPA), and spectrophotometric determination of p-nitroaniline released from
the substrate as a result of trypsin action (Kakade et al. 1969). However, questions
concerning the reliability of the original procedure led to a collaborative study
organized by the American Association of Cereal Chemists (AACC) and the
American Oil Chemists' Society. A modified procedure was adopted for an
official method by AACC (AACC 1983). Although reported separately, Smith
et al. (1980) and Hamerstrand et al. (198 I) modified the AACC method in a
similar way by using a single inhibitor level instead of serial inhibitor levels.
This modification bypasses the cumbersome data interpretation required in the
official method, which is done by either extrapolating to zero or averaging over
a range of inhibition levels. The reason for their modification is based on two
observations: the pattern of enzyme activity vs. inhibitor concentration is diverse,
and extrapolation for data interpretation uses data that are not in the region in
which zero-order kinetics is followed.
In all these methods, sufficient preincubation of the inhibitor with the enzyme
is considered necessary for obtaining equilibrium data. However, in the assay of
purified soybean Kunitz inhibitor or BB inhibitor, Liu and Markakis (l989a)
demonstrated that preincubation of the inhibitor with enzyme followed by addition
of the substrate (S-last test) tended to underestimate the inhibition value in
comparison with preincubation of the inhibitor with the substrate followed by
enzyme addition (E-last test). When a dilute soy extract was assayed for trypsin
54 I Soybeans: Chemistry, Technology, and Utilization
inhibition, the same order effect was found (Fig. 2.12). Further investigation
showed that this "reactant sequence effect" on trypsin inhibition assay is time-
and pH-dependent (Liu and Markakis 1989a, 1989b), and the phenomenon is
also observed with chymotrypsin inhibitor assay (Liu and Markakis 1990). They
attributed this sequence effect to limited hydrolysis of the inhibitor by the very
enzyme they inhibit, according to the reactive site model proposed by Ozawa
and Laskowski, Jr. (1966). The model holds that trypsin inhibitors have a trypsin
susceptible bond in their reactive site and that there is a cleavage by trypsin of the
bond in some inhibitor molecules during their interaction. The cleaved inhibitor is
known as the modified inhibitor and has two peptide chains strongly held together
by a disulfide loop. Although both virgin and modified inhibitors are active
toward trypsin inhibition, the modified one reacts more slowly than the virgin
inhibitor. The preincubation of the inhibitor with trypsin in the S-last test favors
hydrolysis of the inhibitor by trypsin, leading to production of more modified
inhibitor than in the E-Iast test. This explains why the E-Iast test gives higher
inhibition value for the same amount of the inhibitor.
Consequently, an improved colorimetric method for determining antitryptic
0.50 .......- - - - - - - - - - - - - - - - .
0.40
0.30
o
:;
<{
0.20
0.10
c The E-Iast test

o.oo~~O~~~~~~~r_~--~--~~--~
0.0 0.2 0.4 0.6 O.B 1.0
ML soy extract in 4 mL assay mixture
Figure 2.12. Effect of the sequence of mixing the reactants on the assay of antitryptic
activity in a dilute aqueous soy extract. In the S-last test, 0.5 ml of porcine trypsin solution
prepared with 20 mM acetate buffer, pH 3.5, was premixed with 1.0 mL of the diluted
soy extract prepared with the acetate buffer. After 3 min, 2.0 mL of BAP A solution was
added and the reaction was allowed to proceed for 10 min. In the E-Iast test, the enzyme
was added 3 min after mixing the substrate with the sample solution. From Liu and
Markakis (1989b).
Chl'lIlistry and Nurri/ional Value of Soybean Componenls / 55
activity in soybean products is proposed (Liu and Markakis ] 989b). Compared

with the AACC method (AACC 1983), the proposed modification is not limited
to a use of E-last test procedure. It also includes: (1) Water rather than dilute
alkali is used for extracting the inhibitors, (2) the aqueous extract is destablized
with a Tris buffer and filtered before, rather than after, the reaction, (3) porcine
rather than bovine trypsin is used, (4) a single inhibitor level is used instead of
serial inhibitor levels, and (5) the volume of the reaction mixture is reduced from
10 to 4 rnL. The proposed procedure is more sensitive and apparently less tedious
than previous procedures.
E. Lectin
Lectins, also known as hemagglutinins, are proteins and possess a remarkable

ability to agglutinate erythrocytes and other types of cells. They are found predom-
inantly in plant seeds, particularly those of the legumes, but they are also present
in other parts of plants such as roots, leaves, and bark (Sharon and Lis 1972).
Lectins are characterized by a relatively high content of 4-hydroxyproline. Their
ability to agglutinate cells results from their ability to bind specifically to saccha-
rides on the surface of cells (membranes) and act as bridges between cells.
Because of this feature, lectins have provided a new tool for cell biologists to
investigate the architecture of cell surfaces. Evidence has accumulated to suggest
that lectins have a function in the regulation of cell wall extensibility (Cooper
et a!. 1987). More recently, lectins in legume roots have been shown to play an
important role in recognition of Rhizobium during formation of nodules for
nitrogen fixation (Smith and Gallon 1993).
Seed lectins are primarily localized in the protein bodies of the cotyledon cells.
Soy lectin sedimentates with the 7S fraction during ultracentrifugation . First
purified and studied by Pallansch and Liener ( ] 953), the hemaggl utinin in soybean
seeds has a MW of approximately 120 kDa and is comprised of four identical
subunits. each of which has a MW of 30 kDa. In addition to reacting with
carbohydrates, the soybean hemagglutinin is a glycoprotein containing 5 glucos-
amine and 37 mannose residues per mole (Lotan et al. 1974).
For a long time lectins have attracted the attention of food scientists and
nutritionists because some of these proteins, such as ricin from the castor bean.
are toxic to animals. The ability of soybean lectin to inhibit the growth of rats
was first demonstrated by Liener (1953) who showed that it accounted for about
25 % of the growth inhibition produced by raw soybeans. Such growth inhibition
cannot be explained by trypsin inhibitor activity alone, and was confirmed by
severa] later investigators, including Donatucci (1983). However, a study by
Turner and Liener (1975) showed that rats fed soybean extracts from which the
soy lectin had been removed selectively by affinity chromatography grew just
as poorly as those receiving the original crude soybean extract. As demonstrated
also by animal studies, in addition to growth inhibition, the soybean lectin is
linked to an enlargement of the pancreas, a lowering of blood insulin levels, an

inhibition of the disaccharidase and proteases in the intestines, degenerative
changes in the liver and kidneys, and an interference with absorption of nonheme
iron and lipid from the diet (Liener 1994).
Like the trypsin inhibitors, soy lectin is readily destroyed by moist heat treat-
ment; its inactivation closely parallels the destruction of the trypsin inhibitors in
soybeans. However, the soy lectin is quite resistant to inactivation by dry heat
treatment (de Muelenaere 1964). This feature may explain why low residual but
significant levels of lectin activity were detected in a number of soybean products
intended for human consumption (Calderon de la Barca et al. 1991). There are
genetic variants for soy lectin levels. Pull et al. (1978) reported seeds of five
cultivars lacked lectin after screening 102 varietal samples. However, using a
much more sensitive assay methodology, Tsien et al. (1983) detected measurable
levels of lectin activity in these same seeds and found that they are 1000-10,000-
fold less than those in seeds of a common cultivar. In vivo assay with rats
indicated that a soybean cultivar with lower lectin activity had higher PER than
a normal variety (Donatucci 1983).
Lectin activity is commonly measured by a serial dilution technique in which
the end-point is determined either visually or spectrophotometrically (Liener
1955, Hwang et al. 1974) or by other means (Kohle and Kauss 1980). The activity
is expressed as the least amount of lectin (the highest dilution) necessary to
agglutinate the red blood cells of a given animal. The cells are sometimes treated
with a protease in order to enhance the sensitivity. In addition to measuring the
capacity to agglutinate red blood cells, lectins may also be tested for their sugar
specificity. This is normally done by incorporating various concentrations of a
given sugar into the test system and measuring to what extent agglutination is
inhibited. With few exceptions, lectins interact with the nonreducing terminal
glycosyl groups of polysaccharides and glycoproteins.
Because the aforementioned common method uses animal and human red
blood cells, one must choose the blood or blood group for which the lectin is
specific. Improper selection of red blood cells may result in low sensitivity or
even negative results. For soybean lectin, trypsin- or papain-treated human or
rabbit erythrocytes have proved to be the most sensitive blood system. To over-
come the limitation of common methods, Kaul et al. (1991) proposed a new
procedure using polystyrene latex beads bound by various glycoproteins. The
technique not only obviates the need for fresh blood, but also provides a conve-
nient method for the determination of sugar specificity of a given lectin. In the case
of soy lectin, the covalent coupling of the latex beads to N-acetyl-galactosarnine or
lactosamine proved to be the most sensitive agglutinating system.
F Lipoxygenases
Lipoxygenase (LOX), EC 1.13.11.12 linoleate:oxidoreductase, is an iron-contain-
ing dioxygenase that catalyzes the oxidation of certain polyunsaturated fatty
acids, producing conjugated unsaturated fatty acid hydroperoxides. The enzyme

also has an ability to form free radicals, which can then attack other constituents.
LOXs have been found in plants, animals, and fungi. In plants, they are present
in various organs and the highest activities are found in legume seeds (Whi-
taker 1991).
LOXs in soybeans are of particular interest because they have been implicated
as the principal cause of the undesirable flavors, commonly known as "greeny"
or "beany," associated with soybean products. Numerous articles have been
published on the identification and characterization of soybean LOXs and their
roles in producing flavor compounds in soy products, among which are Axelrod
et al. (1981), MacLeod and Ames (1988), Hildebrand et al. (1988), Robinson et
al. (1995), and Wilson (1996).
I. Occurrences
Soybean seeds are the richest known source of LOXs. Four LOX isozymes
have been isolated and identified as L-I, L-2, L-3a, and L-3b. The last two are
so similar in behavior and composition that they are often considered a single
type (L-3). All isozymes are monomeric proteins with a molecular weight in the
range of 100,000, and contain one atom of tightly bound nonheme iron per
molecule. L-l , the best-characterized enzyme among the isozymes, differs from
the others in being heat stable, having a pH optimum of approximately 9, and
preferring anionic substrates (linoleic and linolenic acids). This last feature has
occasionally led to L-I being misleadingly described as an "acid LOX." L-2 and
L-3 are less heat stable, prefer esterified substrates, and have pH optima close
to neutrality. With linoleate as a substrate, the isozymes also differ in their
product regiospecificity; L-l shows a preference for the 13 position as the site
for hydroperoxidation , whereas L-2 and -3 use either position 9 or 13 (Christopher
and Axelrod 1971).
It has been reported that L-3 is the most abundant isoenzyme in mature soybeans
on a protein basis. L-l is almost as abundant as L-3. L-2 is least abundant but
has the highest specific activity. Therefore, on the basis of enzymic activity,
similar amounts of L-2 are present in soybeans (Hildebrand et al. 1988). Further-
more, weather conditions have been found to playa considerable role in influenc-
ing the activities of the LOX isoenzymes in soybeans; the differences in activity
between the generations of a cultivar were larger than those between the different
cultivars from the same year (Marczy et al. 1995).
2. Oxidative Reaction and Off-flavor Formation
LOX catalyzes hydroperoxidation of linoleic acid and other polyunsaturated

lipids that contain a cis,cis-lA-pentadiene moiety, in the presence of molecular
J
Linoleic acid
Lipoxygenase
~
OOH
~ ~
"
H "OOH
13-HPOD
Figure 2.13. Lipoxygenase catalized oxygenation of linoleic acid to produce 13-hydro-
peroxy-cis-9,trans-ll-octadecadienoic acid (13-HPOD) .
oxygen (Fig. 2.13). The primary products are hydroperoxides. Three steps are
assumed for the reaction: (1) activation of the native enzyme, (2) removal of a
proton from the activated methylene group, and (3) insertion of the oxygen into
the substrate molecule with formation of the hydroperoxide (Robinson et al. 1995).
The initial products of lipoxygenase activity may be degraded into a variety
of C-6 and C-9 products through the action of hydroperoxide lyases and/or
isomerases (Fig. 2.14). These volatile carbonyl compounds are aldehydes, ke-
tones, and alcohols. Many of them have an objectionable odor or undesirable
flavor, which is responsible for off flavors associated with various soy products.
Among the volatile compounds, hex anal is primarily responsible for the "greeny"
flavor of soy products due to its extremely low flavor threshold (less than I ppm)
(Fujimaki et al. 1965).
3. Other Features
LOX also catalyzes cooxidation of such pigments as carotenoids and chloro-

phyll by a free radical mechanism that requires the presence of a polyunsaturated
fatty acid. This ability leads to the use of enzyme-active full-fat soy flour in
bleaching wheat flour. The enzyme-active flour also helps release bound lipids,
improve dough rheology, and increase the loaf volume of bread (Frazier 1979).
However, because of some adverse effects, such as the formation of off-flavors,
it is used only up to 0.5% in wheat flour. LOXs, particularly L-2 and L-3 isozymes,
have recently been shown to participate in the degradation of SH- groups in
soy milk during soybean grinding (Obata et al. 1996). Thus, their activities affect
gel formation properties of soy protein during tofu making.
Chemistry and Nutrilional Value (if Soybean Components / 59
OOH.'·
~/:~COOH
13-Hydroperoxy-cis-9,trans-ll-octadecadienoic acid
Hydroperoxide isomerase
Hexanal 12-0xo-cis-9-dodecenoic acid
OH
COOH
13-Hydroxy-12-oxo-cis-9-octadecenoic acid (a-ketol)

o
or II
COOH
OH
9-Hydroxy-12-oxo-trans-1O-octadecenoic acid (y-ketoll
/
"
~'~COOH
:' OOH
9-Hydroperoxy-trans-l 0, 13-cis-octadecadienoic acid
11 Hydmpew,id, Iy.~
cis-3-Nonenal 9-0xo-nonanoic acid
Figure 2.14. Possible degradation products arising from the action of hydroperoxide
lyases and isomerases. From Robinson et al. (1995).
4, Elimination
Because the most important cause of off flavors in soy products is the action
of LOXs on linoleic and linolenic acids, many attempts to improve the flavors
of soy products have centered around the methods of inhibiting or inactivating
these enzymes. Most of the methods developed commercially make use of the
heat sensitivity of LOXs. For examples, Wilkens et al. (1967) made a soymilk
with a good flavor by grinding soybeans in hot water to inactivate Lax. Mustakas
et al. (1969) showed that heat-treating whole beans, followed by grinding, gave
full-fat flour with good flavors even after 2 years of storage. Similarly, Nelson
et al. (1976) found that for more effective elimination of beany flavor, a soymilk
could be made by blanching the whole beans prior to grinding.
Unfortunately, using heat sufficient to inactivate LOXs often leads to some
in solubilization of soy proteins, loss of protein functionality, and introduction of
a cooked or toasted flavor. For this reason, several techniques involving milder
heat treatment have been developed. They are effective mainly aided by adjusting
the moisture or pH, or using aqueous alcohol, or their combinations. Brown et
al. (1982) reported a procedure in which the moisture content of soybeans was
adjusted to 16-18% before the beans were subject to a lO-sec steam heating.
They claimed that the method brought about 99% reduction of LOX activity
while keeping the protein solubility above 70%. Eldridge et al. (1977) found that
soaking or wet-milling whole beans in 29-90% aqueous ethanol for 24 hr at
room temperature could decrease soy LOX activity by > 99%, with optimum
flavor scores at 40-60% concentrations. Later, Borhan and Snyder (1979) showed
that the use of low concentrations of ethanol in water was more effective in
inactivating LOXs if the soaking temperature was increased. For example, LOXs
could be denatured by soaking in 10-15% aqueous ethanol at 45°C for 24
hr. Low alcohol concentration is used because it minimizes denaturation of
soybean protein.
In terms of pH control, LOXs are most stable at pH 6.0. Therefore, any higher
or lower pH would help inactivate the enzyme. Baker and Mustakas (1973)
reported that LOX is very sensitive to moist heat at an acidic medium, but moist
heat treatment at an alkaline pH gives a higher proportion of soluble soy protein.
In the method of Borhan and Snyder (1979) just described, additional use of pH
adjustment could reduce the soaking time required to inactivate the enzyme. In
addition, LOXs have their optimal pH for activity, ranging from neutral to an
alkaline region. An acidic pH would help suppress LOX activity. Che Man et
al. (1989) reported that LOX inactivation was irreversible when treated at pH
3.0 and below, irrespective of the acids used. Since no heat treatment was
involved, more than 70% proteins remained soluble.
LOX activity is greatly enhanced when soybeans are damaged or crushed. List
et al. (1977) noted that the quality of both crude and hydrogenated oils was
adversely affected by field and storage damage of the soybeans, presumably
Chemistry and Nutritional Value oj Soybean Components / 61
due to higher lipoxygenase activity in damaged beans. Therefore, most heat

inactivation methods are effective only when applied to intact beans. And efforts
should be made to ensure minimal damage of soybean seeds during handing.
Since many conventional methods discussed so far for controlling LOXs have
disadvantages and limitations, the genetic elimination of LOX from the seeds
has been attempted by several groups of investigators and has shown some
promise. Mutants lacking one or two of the individual LOX isozymes (Hildebrand
and Hymowitz 1981, Kitamura et al. 1983, Kitamura 1984) and even all three
(Hajika et al. 1991) have been reported. The agronomic performance of mutant
plants in the field is apparently unaffected by the absence of these individual
isoforms. More importantly, some of these mutants have been shown to have
reduced off flavors in their seed products (Moreira et al. 1993, Nishiba et al.
1995). In one recent report, Kobayashi et al. (1995) analyzed the volatile com-
pounds extracted by simultaneous distillation and extraction from soybean ho-
mogenate (soymilk) of the normal and the LOX-lacking soybean varieties by
gas chromatography (GC) and GC-MS (mass spectrophotometry). Their results
indicated that almost all the peaks of the volatile compounds obtained from
soymilk of mutants lacking L-2 and L-3 or lacking all three LOXs were markedly
lower than those of a normal variety (Fig. 2.15). These data suggest that the
concentration and composition of the flavor compounds of the LOX-lacking
soybeans are quite different from those of the normal soybeans.
Most recently, Wilson (1996) summarized studies on the comparison of LOX-
null and LOX-containing soybeans for such soyfoods as soymilk and tofu and
concluded that the LOX-null lines have the functional properties of normal
soybeans but with less beany flavor. Therefore, traditional soyfoods made from
LOX-null lines may become more acceptable in Western diets. It may also be
possible to replace the more costly soy concentrates and isolates with LOX-null
soybean flour for food ingredients.
5. Assay Methods
A number of methods have been developed for determining the oxidation rate
of a substrate catalyzed by LOX. The most commonly used are the manometric
technique based on the measurement of oxygen uptake by the substrate and the
spectrophotometric assay in which the increase in extinction brought about by
the formation of the conjugated diene from lin oleate is followed. The third method
frequently used involves the determination of the peroxide group by a variety
of techniques. The manometric procedure has a wide range of applicability,
because it may be employed with crude preparations as well as with purified
extracts. The spectrophotometric method has the advantage of simplicity and
rapidity, but its use requires optically clear solutions for measurements in the
ultraviolet (UV) region. Peroxide tests are less accurate and, therefore, less
suitable for quantitative application.
19
a)
26
12
11
7
N2~~
ZS I.S. n
.1 j
1
24 68: 1~
141
3 IS rJ
11~1 31
32 33341~
0 10 20 30 40 SO 60 7o
b)
26
19
P
11
10
J 7
10 12
20
pi 17
18
30
2iPZS rJ28
40
l
50
32 3S
60 7o
26 c)
I i 1 2 7
10 12
11
19
Is12122nZS rJ T so
3S
0 10 20 30 40 60 7o
Figure 2.15. Gas chromatogram of volatile substances extracted by simultaneous distilla-

tion and extraction from soymilk made from a normal soybean variety (a), a mutant
lacking L-2 and L-3 (b), and another mutant lacking L-I, L-2, and L-3 (c). Elution time
is min. I.S. (internal standard) was methyl decanoate. From Kobayashi et al. (1995).
Chemistrv and Nutritional Value of Sovhean Components / 63
Ben-Aziz et a1. (1970) described an improved colorimetric method in which

Tween 20 is added to the linoleate solution. As a result, it remains soluble even
at acid pH values and permits spectrophotometric determination of conjugated
dienes in the reaction mixture at any desired pH. The technique helped Axelrod
et al. (1981) to develop another colorimetric procedure that permits measurement
of each individual LOX isozyme based on differences in their optimal reaction pHs
and optimal absorption at 234 nm or 280 nm. Yet, more accurate determinations of
the three isozymes have been carried out by an SDS-PAGE method (Kitamura
1984) or immunological techniques (Yabuuchi et al. 1982).
Recently Suda et al. (1995) reported a simple and rapid method for selectively
screening individual LOX isozymes in soybean seeds. The method is based on
the difference in bleaching activities of these isozymes in contact with methylene
blue or ~-carotene. It consists of three procedures designed to test each individual
isozyme. Each procedure starts with mixing soybean extract or wetting soybean
flour with one of the two dye-substrate solutions, followed by color measurement
either spectrophotometrically or visually within 10 or 5 min, respectively.
c. Nutritional Quality of Soy Protein
The soybean is a major source of vegetable protein for human and animal nutrition
in many countries today. Like proteins of other sources, soy protein provides
calories, essential amino acids, and nitrogen. Yet, the nutritional quality of pro-
teins is a function of many factors, including amino acid composition, digestibility,
and amino acid requirements of the organism fed the protein. Proteins of high
quality are those fully digested, with an amino acid composition closely matching
the amino acid pattern required for the animal or humans consuming the protein.
This section provides a general discussion of these factors affecting nutritional
quality of soybean protein. Additional information may be found in Torun et al.
(1981) and Young (1991).
f . AmiI/O Acid Requirements for Humans and Animals
The amino acid requirements for man are not fully understood. They are
estimated by two expert groups, one organized by the Food and Agriculture
Organization and the World Health Organization of the United Nations (FAO/
WHO) and the other organized by the Food and Nutrition Board of the National
Academy of Sciences (FNB/NAS). Table 2.7 lists the amino acid patterns required
for preschool children and adults as estimated by FAO/WHO (1985), along with
those for rats and broiler chickens. In general, human infants and preschool
childrcn require a source or mixture of food proteins that has a relatively high
concentration of essential amino acids (EAA). With growth and development,
the needs for EAA decline; thus adults need a lower concentration of EAA per
unit of protein to maintain nutritional adequacy than do young children. However,
64 / Soybeans: Chemistry, Te chnology, and Utilization
Table 2.7. Estimated Amino Acid Patterns (mg/g Dietary Protein) required for
Preschool Children, Adults, Rats, and Broiler Chickens
Preschool" Adultsa Broiler
Amino Acid (2-3 Years) (~ 18 Years) Rats b Chickens'
Arginine 50 42
Hi stidine 19 16 25 19
Isoleucine 28 13 46 27
Leucine 66 19 62 58
Lysine 58 16 75 48
Methionine and 25 17 50 32
cyst(e)ine
Phenylalanine 63 19 67 61
and tyrosine
Threonine 34 9 42 28
Tryptophan II 5 12 8
Valine 35 13 50 38
Total 339 127 479 361
aData from FAOIWHO (1985).

bData from NRC-NAS (1972).
' Data from Woodham and Deans (1975).
there is evidence that the current requirements for most EAA in adults are far
too low and thus a new set of tentative estimates for the amino acid needs for
adults has been developed (Young et al. 1989).
When comparing the amino acid patterns required for humans with those for
animals (such as rats and broiler chickens), differences exist (Table 2.7). First,
most animals require higher levels of EAA than humans. Second, some animals,
such as the rat, have a higher requirement for lysine and methionine than humans.
And third, arginine is considered an EAA for most animals but nonessential for
humans. Therefore, the same protein may exhibit different quality when fed to
humans and animals.
2. Amino Acid Composition of Soy Protein
Like proteins of most other leguminous plants, soy protein is low in sulfur-
containing amino acids, with methionine being the most significant limiting amino
acid, followed by cyst(e)ine and threonine (Eggum and Beames 1983). However,
soy protein contains sufficient lysine, which is deficient in most cereal proteins.
This makes it particularly valuable to be combined with cereal proteins as they
are complementary for lysine and methionine.
Zarkadas et al. (1993) measured the amino acid profiles of two soybean cultivars
they grew. Their results (Table 2.8) are in general agreement with those reported
for defatted soy meal (Cavins et al. 1972) and for defatted soy flour and grits
(Kellor 1974). Maple Arrow is a widely-grown normal cultivar, whereas OT89-
Table 2.8. Comparison of the Amino Acid Composition (mg/g Protein) between a
High Protein Soybean Genotype (OT89-16) and a Normal Variety (Maple Arrow)
Amino Acid Maple Arrow OT89-16 F Value"
Essential
Cyst(e)ine 25.00±0.67 23.25±1.03 1.75
Histidine 34.38±5.65 32.28±4.93 0.28
Isoleucine 51.58±O.50 50.05±1.37 0.96
Leucine 81.69±O.73 78.83±2.67 1.42
Lysine 68.37±1.06 62.06±2.09 5.39
Methionine 1O.70±0.3l 9.64±O.50 3.10
Phenylalanine 56.29±0.63 52.62±1.92 7.08
Threonine 4 1.94± 1.79 42.05±1.22 0.00
Tryptophan 12.73±0.41 12.20±0.47 0.55
Tyrosine 41.55±0.64 38.83±1.42 5.64
Valine 54.27±0.32 51.08±O.23 9S.67**
Nonessential
Alanine 40.23±1.l1 3S.04±2.07 0.63
Arginine 77.16±2.35 SO.21±4.63 0.34
Aspartic acid 6S.S6±3.39 SI.77±4.52 2.76
Glutamic acid 190.16±342 200.07±3.34 4.14
Glycine 36.72±0.15 35.97±1.23 0.33
4-Hydroxyproline 1.40±0.02 1.12±O.OS 13.S5*
Proline 52.91±2.3S 51.76±l.lS 0.16
Serine 54.05±1.66 5S.13±2.12 2.43
Source: Data adapted from Zarkadas et al. (1993).

aF values from analysis of variance between the two cultivars; *p<.05; **p<.01.
16 is a newly bred high-protein genotype. Although significant differences exist

for a few amino acids (valine and 4-hydroxyproline), the amino acid profiles of
the two soybean cultivars are very similar. Following are the general findings of
Zarkadas et al. (1993).
1. Glutamic acid is the most abundant amino acid.

2. The acidic amino acids (Glu and Asp) constitute approximately one-
fourth of the total amino acids present, compared to the basic amino
acids (Lys, Arg, and His), which constitute one-fifth.
3. The amino acids with hydrophobic side chains (Gly, Ala, Val, Leu, and
lie) account for a further 19.0-20.0% of the total protein compared to
the mean values of 9.1-9.8% for total aromatic amino acids (Phe, Tyr,
and Try).
4. Mean values for proline account for a further 5.2-5.3%.
5. Proteins for the two soybean cultivars contain all of the essential amino
acids required for human or animal nutrition, namely isoleucine, leucine,
lysine, methionine and cyst(e)ine, phenylalanine and tyrosine, threonine,
tryptophan, valine, and histidine, although they are limited in two major
amino acids: methionine, followed by tryptophan,
3, Protein Digestibility
Protein digestibility is defined as a percentage of protein absorbed after ingest-

ing a certain amount of protein by humans or animals. It is closely related to
amino acid bioavailability. Protein digestibility is a major index of protein quality,
because, although a given amino acid may be present in the protein, it may not
be necessarily available to the organism for nourishment. In other words, proteins
cannot be utilized unless they are digested. A number of factors have been
identified that affect digestibility, including the presence of biologically active
components, heat treatment, and the chemical form of soy protein itself. These
factors affect protein digestibility by either making a certain portion of proteins
fail to hydrolyze into individual amino acids or rendering hydrolyzed amino
acids less available to the organism.
Purified soy proteins may be totally available to the animal organisms, however,
they do not exist in nature, and in most cases, are mixed with other components
naturally present in soybeans. Among these substances, trypsin inhibitors are
thought to be the major factor. As discussed earlier, proteinase inhibitors can
reduce protein digestibility by decreasing or inhibiting the action ofthe pancreatic
enzymes trypsin and chymotrypsin. The effect is evidenced by a parallel relation-
ship between an increase in protein quality as measured by PER in rats following
heat treatment and destruction of trypsin inhibitors (Fig. 2.11).
However, the nutritional effect of protease inhibitors in humans has been a
subject of much controversy (Liener 1994). It remains unclear whether there is
a definite relationship between the protein quality of plant foods and their protease
inhibitor levels. First, it has been suggested that native, undenatured soy protein
is in itself refractory to enzymatic attack unless denatured by heal. The suggestion
is based on an experiment in which the crude soybean extract from which the
inhibitor had been removed was subjected to digestion with trypsin in vitro.
Proteins in heated soy extract had a higher digestibility than those from unheated
soy extract from which inhibitors had been removed (Green et al. 1973). In
addition, Fukushima (1968) reported that most soybean proteins are globular
molecules that are resistant to proteolytic attack unless their internal structure is
disrupted. Thus, it appears that the effect of heat treatment on improving protein
quality is due to not only inactivating trypsin inhibitors, but also denaturing soy
proteins and/or disrupting protein's internal structure.
Second, the proteinase inhibitors of soybeans were found to be, at least in
part, inactivated through pepsin digestion or incubation with human gastric juices.
About 80% of the original trypsin inhibitory activity of Kunitz inhibitor prepara-
tion and about 23% of the original inhibitory activity of Bowman-Birk inhibitor
preparation disappeared during pectin digestion, whereas purified Kunitz inhibi-
Table 2.9. Protein Digestibility of Various Soyfoods in Humans

Soyfood Protein Digestibility (%)
Roasted soy meal 78

Fennented whole soybean (natto) 90
Deep-fried soy curd (tofu) 91
Boiled whole soybean 92
Freeze-dried soy curd (tofu) 93
Soy film (yuba) 100
Source: Data adapted from JSNFS (1984).
tors was rapidly inactivated by incubation with human gastric juices (Krogdahl
and Holm 1981).
Third, there is a remarkable variation in protein digestibility in humans among
various traditional soyfoods (Table 2.9). The highest protein digestibility is with
yuba whereas the lowest is with roasted soybean meal. To find out which factors
are responsible for such variation, Ikeda et al. (1995) recently investigated these
Oriental soyfoods. They found that there was no significant correlation between
protein digestibility of the soybean foods examined by an in vitro assay and their
levels of inhibitory activity against trypsin and a-chymotrypsin. Their kinetic
analysis revealed a difference in susceptibility of proteins to proteolytic action
between soy film and roasted soy meal. Since protein in yuba has been proposed
to have an unfolding, stretched structure by hydrophobic binding with lipid, they
concluded that the chemical form of proteins in soyfoods may play an important
role in their protein digestibility.
Other biologically active substances present in soybeans, such as phenolics
and phytin, are also known to reduce protein quality. Ritter et al. (1987) reported
that phenolics-reduced and phytate-phenolics-reduced soy protein isolates were
slightly more digestible than controls. Their kinetic studies indicated that differ-
ences in in vitro digestibility of soy protein isolates was probably due to an
accumulation of end products rather than steric hindrance at enzyme-substrate
reaction sites. In addition. new studies with calves showed that the antigenicity
of certain soy proteins, particularly when they are under-denatured, appears also
responsible for reduced digestibility of soy proteins (Sissions and Tolman 199 I.
Lalles et a1. 1996).
4. Methods for Assaying Protein Quality
When expressing nutritional quality of a food protein in a numeric term, a

major consideration is the way to measure it. Numerous methods have been
developed to evaluate the nutritional quality of food proteins (Young and Scrim-
shaw 1978, Bodwell and Marable 1981, Kim and Barbeau 1991). Generally, they
can be divided into three major categories: chemical methods, in vivo assays using
either animals or human subjects, and in vitro assays using various proteinases.
A few methods and their uses in assaying soy protein quality are selectively
discussed here.
a. Amino Acid Score
One simple way to estimate the quality of dietary protein is to compare the
amino acid pattern of a test protein expressed as percent of total protein with
the amino acid pattern of the requirement for an animal or humans. The value,
known as the amino acid score, is the percentage of the most limiting amino
acid of the test protein expressed as a percentage of the organism's requirement.
Because the amino acid pattern requirements of humans differ from those of rats,
very different estimates of protein quality are obtained when amino acid scores
are based on the rat pattern rather than on the human pattern. In most cases, the
amino acid scores based on the rat pattern are lower than those based on the
human pattern. For example, isolated soy protein has a score of 100% for the
human pattern. However, when the rat pattern is used, the score becomes 58%,
with sulfur amino acids being most limited. Other shortcomings of the method
include: (1) It does not count the effect of protein digestibility, which is an index
of amino acid bioavailability, and (2) the emphasis on limiting amino acids
obscures the very positive aspect of soy protein, that is, it has a high lysine content.
b. Protein Efficiency Ratio
The protein efficiency ratio method was recognized officially in both the United
States and Canada. In a PER assay, young rats are fed a 10% protein diet for 4
weeks and both growth rate and feed consumption are measured. The PER is
computed by dividing the growth rate by the amount of protein consumed. It is
customary to feed casein as a control protein in PER tests, to which a PER of
2.5 is arbitrarily assigned. The PER of test proteins is adjusted proportionately.
Soy protein has an average PER of 2.3 (Torun et al. 1981). However, since PER
values are determined by rat bioassays, they tend to underestimate the protein
quality of soybeans for humans because the rat has higher relative requirements
for sulfur-containing amino acids.
c. Protein Digestibility Corrected Amino Acid Score
To overcome some shortcomings associated with amino acid scores as well
as the PER assay, a new method has been adopted by the FDA and accepted as
the method of choice by the World Health Organization (FAOIWHO 1990). It
is called the Protein Digestibility Corrected Amino Acid Score (PDCAAS). As
the name indicates, it is in fact based on human amino acid needs and the
digestibility of the protein and is expressed as follows:
PDCAAS = . Ami~o acid. pattern of a protein. x

Ammo aCId requIrements for an orgamsm
digestibility of the protein
Table 2.10. Protein Digestibility Corrected

Amino Acid Score (PDCAAS) of Common
Food Proteins
Protein Product PDCAAS
Peanut meal 0.52
Pea protein concentrate 0.73
Beef 0.92
Isolated soy protein 0.92
Soy protein concentrate 0.99
Egg white 1.00
Source: Data adapted from FAOIWHO (1990).
Therefore, PDCAAS is a measure of how limiting the limiting amino acid is in

a protein after an adjustment for digestibility. Based on the new method, soy
protein in a purified form is equivalent in quality to animal proteins (Table 2.10).
Similar results were also reported by Zarkadas et al. (1993) who compared the
essential amino acid composition of two soybean genotypes, two soybean prod-
ucts, and two high quality animal proteins with the suggested EAA requirements
for humans and broiler chickens.
d. Bioassays with Human Subjects
According to Young and Scrimshaw (1978), clinical evaluations of protein

quality are based on the same principles applied to the corresponding animal
assays, but require some modification for human subjects. Most studies of the
nutritional value of soy protein in human beings have been based on measures
of growth, nitrogen balance, and dietary nitrogen utilization in infants, children,
adolescents, and adults. Young (1991) reviewed many of these studies and con-
cluded that well-processed soy-protein isolates and soy-protein concentrates can
serve as the major, or even sole, source of protein intake and that their nutritional
value is essentially equivalent to that of food proteins of animal origin. Further-
more, under conditions of an anticipated normal usage of soy protein, methionine
supplementation is not only unnecessary, but also may even be undesirable for
young children and adults. However, for newborns, the data suggest that modest
supplementation of soy-based formulas with methionine may be beneficial.
e. In vitro Assays
Biological assays using either rats or humans are expensive and time consum-
ing. Therefore, there has been a renewed interest in using in vitro assays to
measure protein quality and/or digestibility. One approach, described by Satterlee
and coworkers (1977), attempts to determine PER values from empirical equations
based on the amino acid composition data and in vitro estimates of protein
digestibility by a multiple enzyme system. The calculated values are designed

as the C-PER (computed protein efficiency ratio). Other in vitro methods for
measuring protein digestibility include membrane or cell dialysis (Gauthier et
al. 1982), the pH drop method (Hsu et al. 1977), and gel electrophoresis (Kim
and Barbeau 1991). In the membrane or cell dialysis method, protein digestibility
was calculated from the percentage of nitrogen in the dialysate (released peptides
and amino acids that dialyze through a membrane), whereas the multienzyme-
pH drop method is based on a correlation between the drop in pH of a protein
suspension after 10 min enzymatic digestion and apparent in vivo digestibility
in rats. Enzymes used include pepsin, pancreatic, papain, trypsin, chymotrypsin,
and peptidase.
5. Hypocholesterolemic Effects of Soy Protein
Much evidence from studies on experimental animals and human subjects

indicates that substituting soy protein for animal protein in the diet reduces the
concentration of total and low-density lipoprotein (LDL) cholesterol in plasma
or serum. Therefore consumption of soy protein helps to reduce risks of cardiovas-
cular disease. The evidence was summarized by Sirtori et al. (1993) and Carroll
and Kurowska (1995), and further confirmed recently by Anderson et al. (1995)
who performed metanalysis of the effects of soy protein intake on serum lipids
and concluded that 25g, 50g, and 75g of soy protein per day would produce
decreases of 8.9 mg/dl, 17.4 mg/dl and 26.6 mg/dl, respectively, in serum choles-
terol. Chapter 10 provides a detailed discussion of the subject.
6. Allergenicity of Soy Protein
Food allergy or hypersensitivity is an abnormal immunologic response to

certain components in food, and occurs only to certain individuals. Briefly, an
allergenic food component, known as allergen, induces the immune system of
certain population to produce immunoglobulin E (IgE), one of the five antibodies
that our bodies use to fight infection and resist diseases. The IgE, now bound
by mast cells of many tissues or basophil cells in the blood, recognizes the food
allergen in the blood and binds to it, becoming cross-linked with another IgE
molecule as a consequence. This cross-linking of IgE molecules on the surface
of the cell starts a cascade of biochemical events, which ultimately result in
release of histamine and other cellular substances from the cell and leads to allergic
symptoms. The common symptoms include itching, rash, hives, abdominal pain,
nausea, vomiting, diarrhea, wheezing, and, in very rare severe cases, drop in
blood pressure and death. These reactions usually occur within several minutes
to one hour after the food is taken. For additional information on the chemistry
and biology of food allergens, refer to Hefle (1996).
Although the incidence of soybean allergy in adults is relatively uncommon,
epidemiological studies with young children indicated that it is among the most
Chelllislrr and Nutritional Vallie o( SoviJeal1 Components / 71
common forms of food allergy in that segment of the population (Sampson and
McCaskill 1985). Only peanut, cow's milk and egg allergies occur more fre-
quently. With the growing consumption of soybean products all over the world,
the incidence of soy allergy is expected to increase. The allergenicity of soybeans
occurs to not only humans but animals as well. Traditionally, the presence of
certain natural antinutritional factors in soybeans, such as protease inhibitors,
lectins, and oligosaccharides, has been attributed to reduced efficiency of utiliza-
tion of soy protein compared with milk protein. However, evidence has now
indicated that the antigenicity of (particular) soy proteins contributes to the overall
problem of mal digestion and gastrointestinal disturbances of animals fed with
soy products, since antigenic proteins have the ability to activate the immune
system, which directly relates to suboptimal digestion of these proteins, enhanced
maintenance requirements, and loss of endogenous proteins (Sissions and Tolman
1991). The increased incidence of soy allergy in humans and a link between
allergenicity and reduced efficiency of utilization of soy proteins in animals have
led to increased numbers of publications on the subject in recent years (Burks
et al. 1991, Ogawa et al. 1991, 1993, Samoto et al. 1994, 1996, Astwood and
Fuchs 1996, Lalles et al. 1996, Lalles and Peltre 1996).
The allergenicity of soybeans resides in the protein fraction. Soy products
such as soy oil or lecithin would not normally contain allergens, unless protein
contamination had occurred. Various IgE-binding soy proteins have been reported
in a number of clinical studies with humans. These proteins include Kunitz
trypsin inhibitor (Moroz and Yang 1980) and three globulin fractions (2S, 7S,
and II S) (Shibasaki et a1. 1980). While Burks et al. (1991) reported IgE to both
the 7S and II S fractions in eight patients with atopic dermatitis, Ogawa et a!.
(1991) identified a major allergen in soybeans, known as Cly 111 Bd 30 K. It was
later characterized as the 34 kDa oil body-associated protein, or P34. Since this
protein shows considerable sequence similarity to the thiol proteinases of the
papain family. including the allergenic thiol proteinase Dcr p I from house dust
mites. it was recommended that Clr m Bd 30 K or p34 should be named Ch'
11/ I. More recently. Samoto et al. (1996) found that some 34 kDa soy proteins
in soymilk bind preferentially to the a'- and a-subunits of ~-conglycinin while
others form dimers through a disulfide bond.
From animal studies. particularly with preruminant calves, most proteins from
soybeans, including the storage globulins, protease inhibitors and Jectins, have
been shown to participate in the immediate and semi-delayed immune reactions.
However, among soy proteins, ~-conglycinin is a strong allergen for a delayed
type hypersensitivity in the calves (Lalles et a1. 1996). In addition, a low molecular
weight soy protein, identified as P22-25, has also been shown to strongly induce
antibody responses in calves (Hessing et al. 1994), although the nature of the
protein is unknown.
Regardless of elusive identification of specific soy allergens, soybean products,
when sufficiently heat-processed, are generally considered to be hypoallergenic
because the immunological reactivity of most protein components of soybeans

is destroyed by heat treatment (Burks et aL 1991). However, like most other
food allergens, some soy protein components with strong allergenic activity may
be resistant to heating and other processing. They may also be resistant to
digestion and therefore resistant to proteolytic enzymes, acids, and bile salts, and
consequently remain intact through the stomach and intestines and induce adverse
reactions in certain human individuals or animals.
One preventive measure for the food-allergic individual is to avoid the offend-
ing food. However, this requires diligent reading of labels and ingredient listings
and certain knowledge in reading them. Another method is the use of enzymes
to reduce the allergenicity of food components (Burks et aL 1991). One particular
example is the creation of hypoallergenic rice by treating rice with proteolytic
enzyme (Watanabe 1993). Chemical methods are also effective in removing
allergens from food. Samoto et aL (1994) reported that a combination of treatment
with 1M Na2S04, acidification to pH 4.5, and centrifugation removed more than
90% of the 34 kDa protein from defatted soymilk.
Recently, new technologies like recombinant DNA techniques are emerging
to prevent food allergy (Astwood and Fuchs 1996). The use of recombinant DNA
techniques to identify and characterize food allergens has resulted in improved
methods of allergen detection, more accurate allergy diagnoses, and new strategies
for rational and safe immunotherapy. The recombinant DNA techniques are also
being used to reduce or eliminate allergens from the food supply, possibly
resulting in production of hypo allergenic food crops. With respect to soybeans,
although their allergy is relatively uncommon, development of soybean cultivars
lacking the 34 kDa allergen or containing less the a'- or a-subunit of ~-conglyci
nin has been suggested (Samoto et aL 1996).
IV. Carbohydrates
The term carbohydrate, also known as saccharide, refers to a class of compounds

with the general chemical formula C n(H20)m and their derivatives. It includes
simple sugars (mono- and disaccharides), oligo saccharides, and polysaccharides.
Polysaccharides are also known as complex carbohydrates, commonly including
starch and some cell wall structural compounds (cellulose, hemicellulose, and
pectin). Therefore, in one form or another, carbohydrates make up the bulk
of the organic components of the living world. Based on solubility in water,
carbohydrates can also be grouped into water soluble and water insoluble. Because
water solubility decreases as the number of sugar units increases, in general,
compounds consisting of one to several sugar units are water soluble whereas
polysaccharides are water insoluble.
The chemistry of soybean carbohydrates was briefly reviewed by Aspinall
(1988). On average, moisture-free soybeans contain about 35% carbohydrates.
Therefore, they are the second largest component in soybeans. However, the eco-
nomical value of soy carbohydrates is considered much less important than soy
protein and oil. As a result, relatively fewer efforts have been made to study soy
carbohydrates and their potential utilization. The principal use of soybean carbohy-
drate has been in animal feeds where it contributes calories to the diet. Such feed
is used primarily for ruminants, because they can digest the compound better than
monogastric animals. Because research has shown the health benefits of dietary
oligosaccharides (Tomomatsu 1994) as well as a link between the consumption of
dietary fiber and reduced risk of colon cancer and other diseases (Burkitt and Trow-
ell 1975), the value of soy carbohydrates is under further exploration.
Soluble Carbohydrates
Mature soybeans contain trace amounts of monosaccharides, such as glucose and

arabinose, and measurable amounts of di- and oligosaccharides, with sucrose in
the range of 2.5-8.2%; raffinose, 0.1-0.9%; and stachyose, 1.4-4.1 % (Hymowitz
et al. 1972). Basically, the oligosaccharides in soybeans are nonreducing sugars,
containing fructose, glucose, and galactose as two or more units, linked by ~
fructosidic and a-galactosidic linkages (Fig. 2.16).
Among the soluble carbohydrates, raffinose and stachyose receive more atten-
tion, mainly because their presence has been linked to flatulence and abdominal
discomfort associated with human consumption of soybeans and soy products.
In fact, flatulence is one of the major factors limiting soybean utilization as food
(Streggerda et al. 1966). It has been noted most frequently with soy products
from which the carbohydrate has not been removed or degraded. Humans are
not endowed with the enzyme called a-galactosidase necessary for hydrolyzing
the a-galactosidic linkage present in these oligosaccharides. When consumed by
humans, the oligo saccharides cannot be digested in the human duodenal and
small intestinal mucosa. The intact sugars go directly into the lower intestine
where they are metabolized by microorganisms that contain the enzyme. This
results in production of such gases as carbon dioxide, hydrogen, nitrogen, meth-
Stachyose
Figure 2.16. Structure of oligo saccharides found in soybeans.

ane, etc., depending on the individual diet and microflora spectrum. Consequently,
the host experiences flatulence and undesirable side effects (Cristofaro et al.
1974, Liener 1994).
In addition to flatulence in humans, reports have shown that oligo saccharides
may have a detrimental effect on the nutritive value of soy meal to animals.
Coon et al. (1988) noted that soybean meal has a lower metabolizable energy
value for poultry than would be expected based on the proximate analysis.
Removal of the oligosaccharides from the soy meal increased polysaccharide
digestibility by about 50% and increased metabolizable energy values by 25% in
leghorn roosters. However, recent feeding study with dogs showed no significant
differences in nutrient digestibilities between conventional and low oligosaccha-
ride soybean meal (Zuo et al. 1996). The different effect of oJigosaccharides on
nutrient digestibilities between roosters and dogs can apparently be attributed to
the difference in their abilities to digest dietary oligosaccharides.
Unlike trypsin inhibitors and lectins, oligosaccharides are heat stable and heat
treatment alone is ineffective in eliminating them . Several alternative approaches
have therefore been sought. A very effective method has been aqueous ethanol
extraction (Rackis et al. 1970). The method is used mainly in production of high
quality soy protein concentrates. However, it results in a by-product known as
soy molasses, which poses a disposal problem. One use of this heavy brown
syruplike product is as cattle feed additive. To increase the economic value of
soy molasses, a study has been conducted to convert the soluble carbohydrate
portions in the molasses to lactic acid by lactic acid fermentation (Montelongo
et al. 1993).
Another effective means of eliminating oligosaccharides has been enzymatic
hydrolysis by oligosaccharide-degrading enzymes. This procedure uses either
endogenous enzymes produced during germination or exogenous enzymes pro-
duced by fermentation. Reports have shown that germination leads to an almost
complete disappearance of oligosaccharides (Abdullah et al. 1984) and tempeh
fermentation by the mold R. oligo!>porus produces a similar flatus-removing effect
(Calloway et al. 1971). Other examples of using exogenous enzymes include
treating soymilk with an industrial preparation of microbial a-galactosidase (Cruz
et al. 1982) and fennenting soymilk with cultures of lactic acid-producing bacteria
possessing the enzyme (Mittal and Steinkraus 1975). Because soluble carbohy-
drates reside mainly in the whey and are degraded during fennentation, soy protein
isolates, tofu, tempeh, and soy sprouts are virtually devoid of flatus activity.
An ultimate solution to the flatus problem would be the genetic removal of
oligosaccharides by plant breeding. It is known that there is considerable variation
in the raffinose and stachyose content among varieties of soybeans (Hymowitz
et al. 1972). More recently, the use of mutation breeding or genetic engineering
has created soybean lines with low oligosaccharides and they are available for
mass production (Kinney 1996). The meal from these lines has shown an improved
Chemistrv and Nutritional Value of Soybean Components / 75
metabolizable energy content when fed to animals. Soy products made from
these lines would produce less flatulence in humans.
Although the presence of oligo saccharides in soybeans and soy products is
generally considered undesirable in terms of their flatus activity, recent studies
have shown some beneficial effects of dietary oligo saccharides in humans (Masai
et al. 1987, Takasoye et al. 1991, Tomomatsu 1994). These include:
I. Increasing population of indigenous bifidobacteria in the colon which,

by their antagonistic effect, suppress the activity of putrefactive bacteria,
2. Reducing toxic metabolites and detrimental enzymes,
3. Preventing pathogenic and autogenous diarrhea by the same mechanisms
as described in the reduction of detrimental bacteria,
4. Preventing constipation due to production of high levels of short-chain
fatty acids by bifidobacteria,
5. Protecting liver function due to reduction of toxic metabolites,
6. Reducing blood pressure,
7. Having anticancer effects,
8. Producing nutrients such as vitamins, also due to increased activity
of bifidobacteria.
Consequently, oligosaccharides have been developed into one of the most popular
functional foods components, particularly in Japan, where in 1990 alone, about
40 million pounds of nine different types of oligosaccharides were produced
(Tomomatsu 1994).
B. Insoluble Carbohydrates
The insoluble carbohydrates in soybeans include cellulose, hemicellulose, pectin,

and trace amounts of starch. They are structural components found mainly in
cell walls. The seed coat makes up about 8% of the whole soybean by dry
weight and contains about 86% complex carbohydrates (Table 2.1). Therefore,
it contributes a noticeable amount of insoluble carbohydrates to the whole bean.
Kikuchi et al. (1971) studied the chemical and physical properties of soybean
cell wall polysaccharides and their changes during cooking. They found that soy
cell walls contain about 30% pectins, 50% hemicellulose, and 20% cellulose.
Therefore, most soybean carbohydrates (oligosaccharides and complex polysac-
charides) fall into a general category known as dietary fiber. According to a
general definition, dietary fiber consists of the endogenous components of plant
material in the diet that are resistant to digestion by humans (Olson et al. 1987).
The effect of dietary fiber in human diets on nutritional response has received
increasing attention by both scientists and the general public during the last two
decades. Medical research has indicated a clear relationship between several

common diseases in affluent societies and lack of fiber in the diet (Burkitt 1971,
Burkitt and Trowell 1975). Certain physiological responses have been associated
with the consumption of dietary fiber. These responses include an increase in
fecal bulk, a reduction in plasma cholesterol, a blunting of the postprandial
increase in plasma glucose, and a decrease of nutrient bioavailability (Vahouny
and Kritchevsky 1986).
During processing of soybean protein isolates, insoluble polysaccharides are
removed by solubilizing the protein, followed by centrifuging or filtering. Like
soy hulls, the insoluble fraction is used mainly as a feed ingredient. However,
as demands for more dietary fiber in processed foods increase, the carbohydrate
in soybeans could be used as a valuable source of food fiber.
V. Minor Components
In addition to lipids, proteins, and carbohydrates, soybeans also contain various

minor components, including minerals, vitamins, phytin, and phenolics.
A. Minerals
Dry soybeans have an ash content of approximately 5%. Since the major forms
of minerals in ash are sulfates, phosphates, and carbonates, the oxygen content
of the ash accounts for much of its weight. Among the major mineral components
in soybeans, potassium is found in the highest concentration, followed by phos-
phorus, magnesium, sulfur, calcium, chloride, and sodium. The contents of these
minerals range from 0.2 to 2.1 % on average values. The minor minerals present
in soybeans and soy products include silicon, iron, zinc, manganese, copper,
molybdenum, fluorine, chromium, selenium, cobalt, cadmium, lead, arsenic, mer-
cury, and iodine. The contents of these minor minerals range from 0.01 to 140
ppm. Like other components, minerals in soybeans are also influenced by variety,
growing location, and seasons (O'Dell 1979, Perkins 1995).
During processing, the majority of mineral constituents follow the protein or
meal portion of soybeans rather than the oil. Some portion of calcium, magnesium,
and phosphorus can be extracted with the phospholipids and become part of the
oil. Others, such as iron and copper, when present in the oil, are regarded as
important contaminants since they are strong peroxidants. These minerals may
arise from the original beans or from metal contact during processing.
B. Vitamins
Soybeans contain both water-soluble and oil-soluble vitamins. The water-soluble

vitamins present in soybeans mainly include thiamin, riboflavin, niacin, panto-
thenic acid, and folic acid. They are not substantially lost during oil extraction and
subsequently toasting of flakes. Fernando and Murphy (1993) reported consistent

recoveries of 84% for thiamin and 95% for riboflavin in the whole soy flour
made from three soybean varieties. The contents in these samples ranged from
6.26 to 6.85 Ilg/g and 0.92 to 1.19 Ilg/g for thiamin and riboflavin, respectively.
However, they also reported that during processing of soybeans involving water,
such as tofu making, losses of these vitamins are remarkable. The ranges of
retention for both thiamin and riboflavin in tofu were found to be 7.6-15.7% and
11.7-21.1 % respectively. The amount of ascorbic acid (vitamin C) is essentially
negligible in mature soybeans, although it is present in measurable amounts in
both immature and germinated seeds (Bates and Matthews 1975).
The oil-soluble vitamins present in soybeans are vitamins A and E, with
essentially no vitamins D and K. Vitamin A exists mainly as the provitamin [3-
carotene. Like ascorbic acid, its content is negligible in mature seeds but measur-
able in immature and germinated seeds (Bates and Matthews 1975). Vitamin E
is also known as tocopherol. It has four isomers, namely, a-, [3-, 't- and ()-
tocopherols (Fig. 2.17).
According to Guzman and Murphy (1986), the tocopherol content varies sig-
nificantly from one soybean variety to another. The amounts of a- , 't-, and ()-
tocopherols in the soybean range from 10.9 to 28.4, 150 to 191, and 24.6 to 72.5
Ilg/g (dry basis), respectively. Processing of soybeans into tofu results in 30-47%
loss of vitamin E, but the tofu is a greater source of tocopherols than the whole
beans on a dry basis. Pryde (1980) reported that crude soy oil contains 9-12
mg/g of a-tocopherol, 74-102 mg/g of Hocopherol, and 24-30 mg/g of ()-
tocopherol. The amount of [3-tocopherol in soybeans is insignificant, being less
than 3% of the total.
During solvent extraction of soybeans, vitamin E goes with oil. In fact, it is
x
HO
a- tocopherol x=y=CH 3
13 - tocopherol x=CH 3 , y=H
y - tocopherol x=H, y=CH 3
b- tocopherol x=y=H
Figure 2.17. Structural differences of the four tocopherol isomers found in soybeans.
considered an important constituent of soy oil partly because it is the most

effective natural antioxidant and partly because it is good for human nutrition.
All tocopherol isomers tend to decrease during oil refinery, with 1:-tocopherol
losing the most. They are lost mainly in the deodorization step. In addition, Chu
and Lin (1993) found that both high moisture content and long storage of soybeans
led to the lower content of the total tocopherol present in the crude soy oil, with
moisture content being a more important factor than storage time. They also
observed that the presence of oxidized soybean oil in the oil extractor had a
significant effect in decreasing the total vitamin E content in the extracted oil.
C. Phytate
Phytate is the calcium-magnesium-potassium salt of inositol hexaphosphoric acid

commonly known as phytic acid. Phytate and phytic acid are also referred to as
phytin in some literature. A proposed structure of phytic acid is presented in
Figure 2.18. Reviews on phytin in food systems and their nutritional implications
can be found in Erdman (1979), Cheryan (1980), Maga (1982), and Liener (1994).
J. Occurrence
As in most seeds, phytate is the principal source of phosphorus in soybeans.

Among the 15 soybean varieties measured, the phytate content ranged from 1.00
to 1.47% on a dry matter basis (Lolas et al. 1976). This value represented
51.4-57.1 % of the total phosphorus in seeds. However, the actual content depends
on not only variety, but also growing conditions and assay methodology. The
phytate content in several commercial soy protein products was also reported,
with soy meal having a level of 1.42%, flakes, 1.52%, and isolates, also 1.52%
(Maga 1982).
In many cereals and oil seeds, phytate is known to be located in the protein
H9 G 8. OH
o=p-o o-p=o
6 6
Figure 2.18. Structure of phytic acid.

Chemistry and Nutritional Value of Sov/Jean Components / 79
bodies, mainly within their globoid inclusions, However, confusion exists over
the location of phytate in protein bodies of soybeans, Tombs (1967) reported
that phytate in soybeans is associated with protein bodies but there appears to
be no specific site of localization. This could be attributed to the existence of a
phytate-protein complex (Pernollet 1978). However, Lott and Buttrose (1978)
employed transmission electron microscopy and revealed the presence of electron-
dense globoid crystals from soybean protein bodies. Their energy dispersive x-
ray analysis showed that the globoids are very rich in phosphorus, suggesting
that the inclusion contains mainly phytate. This finding was later confirmed by
Prattley and Stanley (1983).
2. Nutritional Implications
The physiological function of phytate in plant seeds has been generally assumed
to serve as a source of phosphorus for germination. We are interested in the
substance mainly because of its effect on mineral bioavailability and protein
solubility when present in animal feed or human food. It is well documented
that the requirement for certain metals in experimental animals is increased when
soybeans are used as a source of protein in their diet. The effect has been attributed
to the ability of phytic acid to chelate with di- and trivalent metal ions, such as
Ca2+, Mg2+, Zn 2+, and Fe 3+. to form poorly soluble compounds that are not readily
absorbed from the intestine. Such a conclusion is based not only on animal
studies (Weaver et al. 1984) but also human experiments (Walker et al. 1948,
Young and Janghorbani 1981) and in vitro studies (Zemel 1984, Sandberg et
al. 1989).
However, results among different studies do not always agree. For example.
Mason et al. (1993) found that although removal of phytate resulted in the soy
flour with the most available calcium, calcium absorption by rats from the various
flours was all high and the differences were generally small. Thus. varying the
concenlration of endogenous phytate in soybeans seemed to have little effect on
calcium absorption in rats. Direct experiments with human subjects have also
given conflicting results (Young and Janghorbani 1981). The discrepancy may
result from the complexity of diets as well as different responses among animals
and humans. It is known that other components in the diet. such as fiber, can
also play a role in mineral nutrition. In addition, unlike humans, rats have
intestinal phytase (E.C.3.1.3.8), an enzyme hydrolyzing phytic acid to inositol
and phosphoric acid and thereby removing the metal chelating properties of
phytic ;lcid.
Phytate is also capable of forming complexes with negatively charged protein
molecules at alkaline pH through calcium and magnesium binding mechanisms
and with positively charged protein molecules at pH values below their isoelectric
point by charge neutralization. As a consequence of this nonselective binding to
proteins, phytate has been shown not only to inhibit the action of a number of
enzymes important in digestion (Knuckles and Betschart 1987, Vaintraub and

Bulmaga 1991) but also to affect the isoelectric point, solubility, and functionality
of soy proteins (O'Dell and deB oland 1976, Chen and MOff 1985), The phytate-
reduced soy protein extract exhibited minimum solubility at pH 4,8-5.0, com-
pared to 4.2-4.5 for the control (Fig. 2.19). It had greater solubility below pH
4-4.5 than the control. In addition, the phytate-reduced soy protein isolate was
most soluble and functional at pH values below its isoelectric point (pH 3),
whereas control and commercial soy isolates were generally most soluble and
functional at pH values above their isoelectric points (pH 6 and 9, respectively).
3. Effects on Cooking Quality
Phytate has been shown to affect cooking quality of seeds from some other
legume species, including cowpeas, common beans, lima beans. Such relationship
has been derived mainly from studies on the hard-to-cook phenomenon, i.e.,
certain legume seeds require longer cooking time after prolonged storage at high
temperatureihigh humidity conditions. It has been shown that a decrease in phytate
content during storage of these legume seeds is accompanied by an increase in
the hard-to-cook defect. Because cooking of beans involves dissolution of pectic
substances in the middle lamella of cells, postulated theory holds that divalent
cations released from phytate would bind cell wall pectin and make it less soluble
and the beans are hard-to-cook (Kon and Sanshuck 1981, Jones and Boulter
1983). However, there are conflicting reports against the involvement of phytate
in legume cooking quality (Bemal-Lugo et al. 1991). A comprehensive review
of the cellular, biological, and physicochemical basis for the hard-to-cook defect
in legume seeds is recently available (Liu 1995).
Little is known, however, about the relationship between phytate content and
soybean cookability. Soybeans require considerably more cooking time than most
other edible beans such as cowpeas and common beans in order to have a palatable
texture. Such long cooking time discourages the use of soybeans as a dietary item.
Also, unlike most other legumes, the storage-induced hard-to-cook phenomenon is
not obvious in soybeans.
4. Elimination
Because of the nutritional implications of phytate, extensive research has

centered around its reduction and/or removal in soy products. Unfortunately,
phytate is fairly heat stable. Liu (1986) reported that steaming or cooking for 20
min had a slight decreasing effect, but the change was not significant at a p =
0.05 level. However, surprisingly, when beans were soaked and then cooked,
there was a significant increase in phytate content. One explanation might be
that soaking plus cooking liberates phytate from its bound forms with proteins
and leads to higher extractability. In the case of soy isolate, Rackis (1974) reported
80
70
PHYTATE-REDUCED
~
~
EXTRACT
60
50
....E
~
.;
....
"f
~
0
'"
"'-
w
-'
co
--
~
0
'" 30
20
10
o~ ______L -______~ ______- L_ _ _ _ _ _ ~ ______ ~~ _ _ _ _ _ _L - - - J
3. 0 3.5 4.0 4.5 5.0 5.5 6.0

pH
Figure 2.19. Protein solubility of control and phytate-reduced soy extracts as a function
of pH. Extracts were adjusted to pH values indicated with HCl and NaOH and then
centrifuged at 11,000 x g before being assayed for soluble protein by a dye-binding
procedure. From Chen and Morr (1985).
82 / Soyheans: Chemisfly, Technology, and Utilization
that autoclaving for 4 hr at 115°C is required to destroy most of the phytic acid.
However, from a nutritional view point, such extensive heat treatment would be
unacceptable due to amino acid destruction.
Although soaking has little or not effect in removing phytate from soybeans
(Liu 1986), apparently due to limited activity of endogenous phytase present in
dormant soybeans, increasing the soaking temperature has been found to improve
its effectiveness; the average reduced amount of phytic phosphorus was 26%
and the maximum reduction was 36.1 % for cotyledons soaked at 50°C for 16 hr
(Beleia et al. 1993). The effect was attributed to potentiation of endogenous
phytase as a result of the destruction of heat-sensitive cell membranes (Chang
et al. 1977) and/or leakage of phytate into the soaking water (Beleia et al. 1993).
Another way of taking advantage of endogenous phytase is to allow soybeans
to germinate. Reports have shown that there is an increase in phytase activity,
accompanied by a corresponding decrease in phytate content (Chen and Pan
1977, Bau and Debry 1979, Sattar et al. 1990). Using exogenous phytase has
been another effective way to reduce phytate from soybeans. About Y3 to 2/3 of
the phytate was hydrolyzed during tempeh fermentation by the mold Rhizopus
oligosporus (Sudarmadji and Markakis 1977, Sutard and Buckle 1985). The
treatment of soy meal with the culture filtrate of Aspergillus ficcum (Simons et
al. 1990) or crude phytase preparation (Nelson et al. 1968) significantly reduced
the phytate content and effected a noticeable improvement in the availability of
phosphorus in broilers and pigs.
5. Assay Methods
Most methods for the determination of phytate are based on the insolubility
of its ferric salt. The phytate concentration is not measured directly, but indirectly
from the iron (Makower 1970, Wheeler and Ferrel 1971) or phosphate content
of the precipitate (Thompson and Erdman 1982, Sandberg et al. 1989). For iron
analysis methods, sodium hydroxide is often used to convert the precipitate to
sodium phytate and ferric hydroxide. Iron is then measured after taking up the
ferric hydroxide in acid. The concentration ofphytate is derived from the assumed
iron-to-phosphorus (Fe:P) molar ratio in the precipitate. The proposed value for
the Fe:P molar ratio has ranged from 3:6 to 4:6, but the most frequently used
ratio is the 4:6 model in which 8 ferric iron molecules interact with one phytate
molecule. However, Liu (1986) postulated that in the presence of excess Na+
and SO~- during extraction, additional four ferric iron molecules. could be incorpo-
rated into one phytate molecule, resulting in a 6:6 ratio. Because such assumed
molar ratio is unreliable, phosphate analysis has been recommended (Thompson
and Erdman 1982). Recently, Blatny et al. (1995) described a method that deter-
mines phytic acid directly in cereal grains, legumes, and feeds by capillary
isotachophoresis. Because the procedure bypasses the complex sample prepara-
tion found in most other methods, they claimed that the method is simple and
accurate, with a relative standard deviation of 3.8%.
D. lsofiavones
Isoflavones belong to a group of compounds that share a basic structure consisting
oftwo benzyl rings joined by a three-carbon bridge, which mayor may not be closed
in a pyran ring (Fig. 2.20). The structure is generally simplified as C 6 -C r C 6 . This
group of compounds is known as flavonoids, which include by far the largest and
most diverse range of plant phenolics. Besides isoflavones, other subclasses of fla-
vonoids include red and blue anthocyanin pigments, flavones, flavonols, flavanols,
aurones, and chalcones (Deshpande et al. 1984). Isoflavones differ from flavones
in that the benzyl ring B is joined at position 3 instead of position 2.
For years, soybeans have been primarily identified with their high oil and high
protein content. However, during the past several years, there has been much
interest among clinicians and researchers in the potential role of soyfoods in
preventing and treating chronic diseases. Increasing evidence has suggested that
the isoftavones in soybeans might be the contributing factors (Akiyama et al.
1987, Adlercreutz et al. 1992, Cassidy et al. 1994, Anthony et al. 1996). Conse-
quently, there has been an upsurge in our interest in soybeans and soy product
in recent years. Among the focuses of the most recent research are assays for
contents and bioavailability of isoflavones in soyfoods and the use of soybeans
and isoflavones in preventing and treating cancers and other chronic diseases.
This section provides some updated information regarding soybean isoflavones
in term of their occurrences, effects of food processing, and assay methodology.
The physiological role of soy isoflavones is also briefly reviewed; a detailed
discussion on the subject is provided in Chapter 10.
1. Occurrences
Although flavonoids are found in various plant families at different tissues,
isoflavones arc present in just a few botanical families, because of the limited
distribution of the enzyme chalcone isomerase. which converts 2(R)-naringinen.
a flavone precursor. into 2-hydroxydaidzein (Coward et al. 1993). The soybean
is unique in that it contains the highest amount of isoflavones. being up to 3 mg/g
dry weight. (Walter 1941, Eldridge and Kwolek 1983, Kudou et al. 1991). The
isoflavones in soybeans and soy products are of three types, with each type being
present in four chemical forms (Fig. 2.21). Therefore there are 12 isomers of
6
5 4
Figure 2.20. General structure of flavonoids.
Rl R2 Compounds
H H daldzem
OH H genistein
H OCH3 glyclteln
Rl R2 R3 Compounds
H H H daldzin
OH H H genlstin
H OCH3 H glycltin
H H COCH3 6"·O·Aceryldrudzln
OH H COCH3 5° ·O·AcetylgenISlin
H OCH3 COCH3 5°·0·AcetylglycllIn
H H COCH2COOH 5"·O·Malonyldrudzin
OH H COCH2COOH 5° ·O·MalonylgenISlin
H OCH3 COCH2COOH 6"·O·MalonYlgIYCltin
Figure 2.21. Chemical structures of 12 isoflavone isomers found in soybeans.
isofiavones. The tenns, daidzein, genistein, and glycitein refer to the aglycon(e)s
of soybean isofiavones. In the ~-glucoside form, they become genistin, daidzin,
and glycitin. In the acetylglucoside form, soybean isoflavones are named as 6"-
O-acetyldaidzin, 6"-O-acetylgenistin, and 6"-O-acetylglycitin. In the malonylglu-
coside form, the corresponding names are 6"-O-malonyldaidzin, 6"-O-malonyl-
genistin, and 6"-O-malonylglycitin.
Recently Wang and Murphy (1994a) measured isoflavones in several Japanese
and American soybean varieties grown in Iowa. Some of their data (Table 2.11)
indicate that the soybean isofiavone contents depend on not only genetics but
also crop year and growing location, with the effect of crop year being greater
than location. The total isoflavone content in the tested soybean varieties ranged
Table 2.11. lsoflavone Contents (Ilg/g) in Three Japanese and Four American Soybean Varieties Grown in Iowa, USA"
Keburi Kuro diazu Raiden American varieties, 1989
lsoflavone 1991 1992 1991 1992 1991 1992 Prize HP204 LS301 XL72
Daidzin I 4 tI"" tf tr tr 38 4 10 12
Genistein 9 8 8 7 II 8 33 15 16 45
Glycitein 21 19 tr 12 22 20 20 19 19 21
Daidzin 91 96 80 37 115 53 780 196 442 148
Genistin 179 136 174 128 237 148 806 330 562 481
Glycitin 68 50 66 42 96 73 68 63 64 97
00
v, 6"-O-malonyldaidzin 562 322 375 222 407 242 709 349 752 198
6"-O-malonylgenistin 1232 670 1187 717 1191 723 1342 945 1558 1042
6" -O-malonyglycitin 127 70 III 60 183 III 87 94 92 118
6" -O-acetyldaidzin 12 tr tf 2 tf tr tr tr tr
6"-O-acetylgenistin tr 2 tr tf tr 4 1 I 2
6" -O-acetylglycitin 41 33 37 35 40 34 tf 36 33 37
Total 2343 1411 2041 1261 2305 1417 3886 2053 3551 2201
Source: Data adapted from Wang and Murphy (1994a).

"Samples measured in triplicate.
b tr, trace.
from 1.261-3.886 mg/g seed. Among the 12 isomers, 6"-O-malonylgenistin,

genistin, 6"-O-malonyldaidzin, and daidzin are predominant. There is a distinct
distribution pattern of isomers between American and Japanese soybeans; Japa-
nese soybeans had higher 6"-O-malonylglycitin contents and higher ratios of
malonyldaidzin to daidzin and malonylgenistin to genistin. Tsukamoto et al.
(1995) not only confirmed the effects of variety, growing location, and planting
season on the contents of isoftavones in soybean seeds, but also found that
isoftavone contents decreased in all varieties grown at high temperatures.
In addition, the concentration as well as composition of isoftavones vary greatly
with structural parts within a soybean seed (Table 2.12). The concentration of
the total isoftavones in soybean hypocotyl is 5.5-6 times higher than that in
cotyledons. Glycitein and its three derivatives occur exclusively in the hypocotyl.
Isoftavones are almost absent in seed coats. Tsukamoto et al. (1995) also found
that the hypocotyl had a higher concentration of isoftavones compared with
cotyledons. Yet, of the total seed isofiavones, 80-90% were located in cotyledons,
apparently due to cotyledons being the highest proportion in the seed.
2. Effects of Processing
Processing significantly affects the retention and distribution of isoftavone

isomers in soyfoods. Coward et al. (1993) analyzed isofiavone ~-glucoside conju-
gates and aglycones in various foods and ingredients derived from soybeans.
Their results reveal that most Asian soyfoods (Table 2.13) as well as Western
Table 2.12. Concentrations of Isojiavone Isomers in Two Structural Parts of Soybean

Seeds and the Effect of Extraction Temperatures a
Room temp., 24 hr 80°C, 15 hr
Isoflavone Isomers Hypocotyl Cotyledon Hypocotyl Cotyledon
Daidzein 102 33 35 11
Glycitein ndh nd 15 nd
Genistein 35 48 16 14
Daidzin 320 45 838 145
Glycitin 485 nd 1004 nd
Genistin 118 80 246 210
6"-O-malonyl daidzin 423 70 8 3
6"-O-malonyl glycitin 445 nd 1l nd
6"-O-malonyl genistin 144 117 4 nd
6"-O-acetyl daidzin 2 2 57 8
6"-O-acetyl glycitin 6 nd 89 nd
6"-O-acetyl genistin 105 1 39 1
Total concentration 2185 396 2362 392
Source: Data adapted from Kudou et al. (1991).

"Expressed as mg/l OOg dry weight.
bnd. not detected.
Chl'lnislrv and Nutritional Value of Soybean Components / R7
soy ingredients (Table 2.14), when not diluted by the addition of nonsoybean
components or extracted with aqueous alcohol, have total isoflavone concentra-
tions in the range of 1.33-3.83 mg/g of dry weight. Therefore, these levels are
close to those found in the intact soybeans. On the other hand, isoflavones were
not found in soybean oil, indicating that they go with defatted soyflakes during
oil extraction. Fermented soyfoods, which are usually prepared by mixing soy
with other components such as barley, rice, and wheat, contained isoflavones at
lower concentrations, ranging from 0.36-1.38 mg/g of dry weight. Other soy-
based products, such as soy sauce and frozen flavored soymilk, had much lower
concentrations of isoflavones, with a range of 0.02-0.36 mg/g dry matter. In
addition, Asian fermented soyfoods contain predominantly isoflavone aglycones,
whereas in non fermented soyfoods or ingredients of both American and Asian
origin isoflavones are present mainly as ~-glycoside conjugates. These findings
were confirmed by Wang and Murphy (1994b) who quantified 12 isoflavone
isomers in 29 commercial soyfoods, and Fukutake et al. (1996) who measured
genistein and genistin in Japanese soybean and soy products. Furthermore, on
the basis of their analytical data and the average annual consumption of soybeans
and related product, Fukutake et al (1996) calculated that daily intake of genistein
and genistin by the Japanese is 1.5-4.1 and 6.3-8.3 mg/person, respectively.
These levels are much higher than those for Americans or Western Europeans,
whose mortality rates from breast, colon, and prostate cancers are greater than
the Japanese.
Also within nonfermented soyfoods, Barnes et al. (1994) found that soybeans
and defatted soy flour, each of which had been minimally heated during their
preparation, contained mostly isoflavone 6"-0-malonylglucoside conjugates. Soy-
milk, tofu, and soy molasses, each of which had been heated to lOOoe during
preparation, contained mostly isoflavone ~-glucosides. Toasted soy flour and
isolated soy protein had moderate amounts of each of the isoflavone conjugates.
Apparently, malonylglucoside conjugates are thermally unstable and are con-
verted to their corresponding isoflavone glycosides at a high temperature. The
de-esterifying reaction was presumably a result of transesterification of the ester
linkage between the malonate or acetate carboxyl group and the 6"-0-hydroxyl
group of the glucose moiety, yielding methyl malonate or methyl acetate and
the isoftavone glucoside (Barnes et al. 1994).
Perhaps the most significant approach in examining the effects of processing
methods on the retention, distribution, and transformation of isoflavones is the
recent mass balance study of Wang and Murphy (1996). In this study, contents
of individual isomers as well total isoflavones were monitored in products after
each step of processing during preparation of soymilk and tofu, tempeh and soy
protein isolate. They found that the processing steps causing significant losses
(p <0.05) of isoflavones are coagulation (44%) in tofu processing, soaking (12%),
and heating (49%) in tempeh production, and alkaline extraction (53%) in soy
protein isolate preparation (Table 2.15). In contrast, fermentation, defatting, and
Table 2.13. Isoflavone Concentrations in Various Soyfoods a
Conjugated Aglucones Aglucones (%)
Soy Product Basis Genistin Daidzin Genistein Daidzein Total D/G Ratiod Genistein Daidzein
Asian Primary Soy Materials

Soy milk g 0.130 ± 0.004 0.103 ± 0.006 0.007 ± 0.000 0.011 ± 0.002 0.252 ± 0.012
g dry wt 1.680 ± 0.060 1.337 ± 0.087 0.098 ± 0.002 0.141 ± 0.019 3.256 ± 0.168 0.83 5 10
Tofu b g 0.249 ± 0.028 0.121 ± 0.010 0.031 ± 0.001 0.016 ± 0.001 0.417 ± 0.036
g dry wt 1.215 ± 0.137 0.591 ± 0.046 0.151 ± 0.006 0.077 ± 0.005 2.031 ± 0.171 0.49 11 12
Tofu' g 0.269 ± 0.004 0.200 ± 0.008 0.015 ± 0.001 0.015 ± 0.000 0.494 ± 0.011
g dry wt 2.087 ± 0.030 1.513 ± 0.019 0.116 ± 0.004 0.113 ± 0.000 3.827 ± 0.045 0.74 5 7
Soy flour g 0.741 ± 0.100 0.582 ± 0.077 0.015 ± 0.002 nd 1.338 ± 0.178 0.77 2 0
Soy powder g 1.148 ± 0.103 0.582 ± 0.054 0.014 ± 0.001 nd 1.748 ± 0.156 0.50 I 0
Soy nuts g 1.390 ± 0.039 0.853 ± 0.022 0.066 ± 0.001 0.054 ± 0.001 2.363 ± 0.061 0.62 5 6
00 Processed or Fermented Asian Soy Products
00
Tempeh g 0.113 ± 0.028 0.040 ± 0.013 0.164 ± 0.004 0.113 ± 0.007 0.430 ± 0.005
g dry wt 0.296 ± 0.063 0.103 ± 0.029 0.434 ± 0.005 0.298 ± 0.009 1.130 ± 0.096 0.55 59 74
Miso g 0.043 ± 0.004 0.035 ± 0.D25 0.497 ± 0.029 0.345 ± 0.013 0.920 ± 0.070
g dry wt 0.064 ± 0.007 0.054 ± 0.038 0.745 ± 0.068 0.516 ± 0.036 1.379 ± 0.149 0.70 92 91
Rice miso g 0.198 ± 0.011 nd 0.136 ± 0.000 0.071 ± 0.002 0.404 ± 0.009
g dry wt 0.353 ± 0.018 nt 0.242 ± 0.001 0.127 ± 0.003 0.721 ± 0.014 0.21 41 100
Barley miso g 0.155 ± 0.020 0.142 ± 0.025 0.239 ± 0.008 0.185 ± 0.007 0.721 ± 0.053
g dry wt 0.258 ± 0.032 0.235 ± 0.042 0.396 ± 0.012 0.306 ± 0.009 1.195 ± 0.084 0.83 61 57
Shiromiso soup mix g 0.267 ± 0.020 0.163 ± 0.028 0.170 ± 0.006 0.108 ± 0.008 0.708 ± 0.059 0.62 39 40
Akamiso soup mix g 0.319 ± 0.025 0.54 ± 0.044 0.173 ± 0.005 0.136 ± 0.008 0.882 ± 0.080 0.79 35 35
Soybean paste g 0.078 ± 0.014 0.044 ± 0.040 0.251 ± 0.008 0.197 ± 0.009 0.570 ± 0.071
g dry wt 0.160 ± 0.030 0.090 ± 0.081 0.514 ± 0.016 0.404 ± 0.019 1.168 ± 0.146 0.73 76 82
Soybean paste/rice g 0.066 ± 0.029 0.085 ± 0.016 0.108 ± 0.004 0.103 ± 0.006 0.362 ± 0.041
g dry wt 0.106 ± 0.045 0.136 ± 0.026 0.174 ± 0.008 0.166 ± 0.008 0.582 ± 0.061 1.08 62 55
Continued
Table 2.13. Continued
Conjugated Aglucones Aglucones (%)
Soy Product Basis Genistin Daidzin Genistein Daidzein Total D/G Ratiod Genistein Daidzein
Soybean paste/wheat g 0.110 ± 0.008 0.094 ± 0.026 0.124 ± 0.014 0.105 ± O.OOJ 0.433 ± 0.032
g dry wI 0.220 ± 0.015 0.189 ± 0.052 0.248 ± 0.028 0.210 ± 0.003 0.867 ± 0.063 0.85 53 53
Other Soy Foods
Soy sauce g nd nd 0.009 ± 0.002 0.014 ± 0.001 (Um ± 0.003
g dry wI nd nd 0.036 ± 0.014 0.054 ± O.OJ 3 0.090 ± 0.026 1.50 100 100
Soy cheese g 0.028 ± 0.00 I 0.021 ± 0.00 I 0.002 ± 0.00 I 0.001 ± 0.001 0.050 ± 0.003
g dry wI 0.057 ± 0.001 0.043 ± 0.001 0.005 ± 0.001 0.00 I ± 0.002 0.105 ± 0.003 0.71 8 2
Tofutti g 0.002 ± 0.00 I 0.004 ± 0.006 0.004 ± 0.000 0.001 ± 0.002 0.032 ± 0.008
g dry wt 0.064 ± 0.00 I 0.012 ± 0.016 0.014 ± 0.001 0.003 ± 0.004 0.092 ± 0.020 0.19 18 20
Ice Bean g 0.060 ± 0.006 0.055 ± 0.007 0.001 ± 0.000 O.OOJ ± 0.002 0.117 ± 0.014
00 g dry wt 0.184 ± 0.016 0.167 ± 0.022 0.004 ± 0.002 0.004 ± 0.006 0.360 ± 0.004 0.91 2 2
\Q
Source: Data adapted from Coward et a!. (1993).

Note: "Expressed as mg/g wet weight or mg/g dry weight; mean ±I SD of triplicate analyses.
nd, not detected.
"Tree of Life tofu.
'Mori-Nu tofu.
dD/G ratio = (daidzin + daidzein/(genislin + genistein).
Table 2.14. Isoflavone Concentrations in Various Soy Ingredients"
Conjugated Aglucones Total Aglucones (%)
Soy Product Genistin Daidzin Genistein Daidzein DryWt Protein DIG Ratiob Genistein Daidzein
Soybean chips 0.356 ± 0.074 0.331 ± 0.058 0.052 ± 0.019 0.065 ± 0.023 0.802 ± 0.172 2.1 1 1 ± 0.455 0.97 13 16
Soy flours
Nutrisoy 1.448 ± 0.026 1.161 ± 0.003 0.034 ± 0.000 0.033 ± 0.002 2.678 ± 0.027 5.356 ± 0.054 0.58 2 3
Nutrisoy 7B 1.318 ± 0.009 1.112 ± 0.005 0.053 ± 0.002 0.044 ± 0.001 2.527 ± 0.004 5.054 ± 0.008 0.60 4 4
Baker's Nutrisoy 1.300 ± 0.176 1.046 ± 0.117 0.024 ± 0.020 0.019 ± 0.019 2.389 ± 0.332 4.778 ± 0.664 0.56 2 2
10 Toasted Nutrisoy 1.385 ± 0.066 1.093 ± 0.D25 0.044 ± 0.011 0.040 ± 0.005 2.561 ± 0.076 5.122 ± 0.152 0.57
a
Soy concentrates
Water extracted 1.404 ± 0.103 1.180 ± 0.082 0.033 ± 0.002 0.039 ± 0.002 2.656 ± 0.182 3.794 ± 0.256 0.61 2
Alcohol extracted
Arcon F 0.087 ± 0.014 0.064 ± 0.007 0.004 ± 0.001 0.004 ± 0.000 0.159 ± 0.022 0.244 ± 0.034 0.75 4 6
Areon S 0.227 ± 0.059 0.1 02 ± 0.019 0.069 ± 0.001 0.045 ± 0.002 0.443 ± 0.D75 0.682 ± 0.115 0.50 23 31
Soy isolate 0.430 ± 0.138 0.232 ± 0.105 0.105 ± 0.011 0.073 ± 0.004 0.848 ± 0.228 0.931 ± 0.250 0.41 20 24
Soy isolate 0.589 ± 0.004 0.278 ± 0.001 0.189 ± 0.012 0.102 ± 0.005 1.158 ± 0.012 1.273 ± 0.013 0.35 24 27
Soy fiber 0.154 ± 0.004 0.141 ± 0.001 0.114 ± 0.006 0.085 ± 0.002 0.494 ± 0.012 0.84 43 38
Source: From Coward et al. (1993).

aExpressed as mg/g dry weight or mg/g of protein; means ± SD of triplicate analyses.
bD/G ratio = (daidzin + daidzeinl(genistin + genistein).
Chemistry and Nutritional Value C!l Soybean Components / 91
Table 2.15. Mass Balance of lsojiavones during Preparation of Soymilk and Tofu,
Tempeh, and Soy Protein Isolate
Total Isoflavones"
Type of Soyfood Step Processing Resulting Products (mg)
Soymilk and tofu

Raw soybeans 217.2±63.0"
Soaking Soaked soybeans 196.8±80.0b
Soaking water I.O±O.Y
Cooking Cooked slurry 227.2±41.0"
Filtrating Soymilk 194.8±30.0"
Okara 2S.9±0.7,d
Coagulating Tofu 70.8±4.7'
Whey 9S.4±3.2'
Tempeh
Raw soybeans 117.9±0.S6"
Soaking Soaked soybeans I 04.I±S.3'"
Soaking water trace
Dehulling Dehulled soybeans 97.0±14.0'
Seed hulls 1.6±0.1'·
Cooking Cooked soybeans 39.3±4.0d
Cooking water trace
Fermentation Tempeh 27.9±l2.0d
Soy protein isolate
Soybean flour 30.0±1.7"
Defatting Defatted soy flour 31.7±3.2b
Soybean oil 0.7±0.6"
Alkaline extraction Aklaine soluble 17.8±S.3'
Alkaline insoluble IS.8±4.9'
Acid precipitation Protein isolate l4.S±1.5'
Whey 3.3±0.ld
Source: Data adapted from Wang and Murphy (1996).

Note: All types of soyfood were produced from Vinton 81 variety. Initial raw sample weight for
preparation of soymilk and tofu. tempeh. and soy protein isolate was 600g. I (lOg. and SOg. respectively.
'Individual isoflavone glucosides and aglycon forms were normalized for their molecular weight
differences and summed. Values represent the mean ± standard deviation; n = 3.
" ' Column values with different superscripts differ significantly at p <; 0.05.
dehulling did not cause significant loss of isofiavones. The observation that
isofiavone loss was not significant related to okara during tofu making suggests
that the compound is mainly associated with soluble proteins rather than insolu-
ble carbohydrates.
Regarding the transformation of isofiavones during processing, Wang and
Murphy (1996) found that in the production of tempeh, soy milk, and tofu, malo-
nyldaidzin and malonylgenistin decreased after soaking and cooking. This was
accompanied by increases in acetyldaidzin and acetylgenistin. Tempeh fermenta-
tion caused increases in daidzein and genistein, apparently resulting from fungal
enzymatic hydrolysis of isoftavone glucosides, In protein isolate processing,

alkaline extraction also led to hydrolysis of isoftavone glucosides, resulting in
not only loss of total isoftavones but also increases in daidzein and genistein.
In addition, aqueous alcohol (70-90%) extraction is commercially used for
certain brands of soy protein concentrate (SPC) to remove soluble carbohydrates
from defatted soy ftakes (see Chapter 8 for details). Since aqueous alcohols are
excellent solvents for isoftavones, the process would lead to considerable loss
of isoftavones. According to Coward et al. (1993), SPC, when prepared by hot
aqueous ethanol (65%) extraction, had 1O-20-fold lower isoftavone concentra-
tions than most other soy foods. By contract, SPC prepared by hot water extraction
at neutral pH retained most of the isoftavones present in original defatted soy
ftour. This observation indicates that isoftavones have strong protein binding and
low aqueous solubility.
3. Physiological Effects on Humans and Animals
The major soybean isoftavone aglycones, genistein and daidzein, have been
identified for many decades (Walter 1941). Original research regarding the physio-
logical effects of isoftavones was limited to their estrogenic activity (Wong and
Flux 1962), interference with mineral metabolism, and growth inhibition shown
in rats (Magee 1963). Later research, however, has led to a dilemma. On one
hand, isoftavones have been shown to be partially responsible for an objectionable
aftertaste associated with consumption of soy-based products (Huang et al. 1981,
Kudou et al. 1991, and Okubo et al. 1992). Such aftertaste is characterized as
being sour, bitter, and/or astringent. From this aspect, the presence of isoftavones
is undesirable and they should be eliminated or reduced in soy products (Tsuka-
moto et al. 1991).
On the other hand, isoftavones have also been shown to possess antioxidant
and antifungal activity (Nairn et al. 1976, Pratt and Birac 1979, Fleury et al.
1992), and more importantly, to act as anticarcinogens (Bartholomew and Ryan
1980, Verdeal et al. 1980). In one example, Peterson and Barnes (1993) showed
that genistein acts as an in vitro inhibitor of protein tyrosine kinases, many of
which form part of growth factor-stimulated signal transduction cascades in
normal and transformed cancer cells. There is an apparent association between
increases in soyfood consumption and reduction of cancer risk, and the isoftavone
has been identified to be the contributing factor accounting for such relationship
(Messina and Barnes 1991, Adlercreutz et al. 1992, Messina et al. 1994). Conse-
quently, isoftavones, together with certain other trace compounds present in
plants, have been dubbed as phytochemicals. Although they are not classified
officially as nutrients, these compounds reportedly affect human health as much
as vitamins and minerals do (Caragay 1992, Messina et al. 1994). Further discus-
sion on this subject is found in Chapter 10.
4. Assay Methods
Isofiavones are commonly determined by high pressure liquid chromatography
(HPLC) after extraction from test samples with an aqueous organic solvent
(Kudou et al. 1991, Wang and Murphy 1994a, Barnes et al. 1994). A reverse-
phase HPLC column and an ultraviolet detector are normally required, along
with a gradient solvent solution as a mobile phase. Alternatively, isoftavonoid
aglycons can be analyzed by gas chromatography (Fenner 1996).
There have been variations in extraction conditions among studies. Extractants
which have been used include 70% aqueous ethanol (Kudou et al. 1991), 80%
aqueous methanol (Barnes et al. 1994), and 80% aqueous acetonitrile containing
0.1 % Hel (Barnes et al. 1994, Wang and Murphy 1994a). The extraction time
has ranged from 2 to 24 hr and the extraction temperature from room temperature
to 80°e.
Barnes et al. (1994) reported that maximum recovery of the isoftavones from
soyfood samples was obtained by tumbling for 2 hr at a room temperature or
60°C and that there was no significant differences between the use of 80%
aqueous methanol and 80% aqueous acetonitrile containing 0.1 % HC!. However,
among the variables with extraction, temperature has been shown to exert a
significant effect on final results with respect to both total isoftavone content and
isomer composition. According to Kudou et al (1991), when the samples were
extracted at 80°C instead of room temperature, malonylated isoftavone glycosides
in 70% alcohol extracts from both soybean hypocotyl and cotyledons decreased
significantly as glycosides increased (Table 2.12). Later, Barnes et al. (1994)
confirmed the finding and recommended that extraction at higher temperatures
be avoided. They attributed this observed effect of extraction temperature prior
to sample analysis on the content and composition of isoftavones to the heat-
induced de-esterifying reaction of malonylglucoside conjugates.
VI. Soy Hulls
The soybean hull is also known as the seed coat. On a dry weight basis, hulls
constitute about 8% of the total seed, depending on variety and the seed size. In
general, the larger the seed, the lower the proportion of hulls. The hull of dry
mature soybeans contain about 7-8% moisture (Ibrahim et al. 1990). It is a hard,
water-resistant material and thus protects the cotyledons and hypocotyl from
damage. In an intact seed, the hull is well attached to the cotyledons and relatively
difficult to remove from them. However, when seeds are dried and cracked into
several pieces, as in the soybean mill processing, the hull detaches readily and
is separated from the cotyledons by aspiration.
Dry soy hulls contain about 85.7% carbohydrates, 9% protein, 4.3% ash, and
1% lipids (Table 2.1). The fatty acid composition of hull lipids was recently
found to be significantly different from those of cotyledons and hypocotyl axis
(Liu et a!. 1995b). For 6 genotypes tested, the average percentages plus standard
deviation of major fatty acids for soy hulls are palmitic acid, 23.2 ± 3.0; stearic,
14.8 ± 2.5; oleic, 14.9 ± 3.4; linoleic, 22.7 ± 5.4; and linolenic, 8.5 ± 2.4. Soybean
hulls also contain three plant sterols, campestrol, stigmasterol, and ~-sitosterol.
Their ratios were found to be 1:1.5:2 (Ibrahim et al. 1990).
There are at least three aspects regarding significance of soy hulls. First, soy
hulls affect the seed hydration rate prior to germination or processing. Second,
soybean hulls serve as valuable feedstuff. And third, they possess some potential
as a source of dietary fiber and iron for human consumption. During imbibition
(soaking) prior to germination or processing into various soyfoods, some soybeans
do not absorb water or enlarge significantly. They are known as hard beans. The
occurrence of hard beans is an important defect because hard beans either fail
to germinate or affect the quality of soy products. Regarding the water resistant
mechanism of hard soybeans, Smith and Nash (1961) observed that the seed coat
was the principal barrier to water imbibition and that hard beans usually were
smaller and drier than soybeans that imbibe normally. Saio (1976) examined
hard beans by proximate analysis, light microscopy, and scanning electron micros-
copy. He found that compared with normal beans hard beans had higher fiber
and calcium content and denser and tougher seed coat. He also found that the
micropyle of hard beans was covered with the outside palisade cells. These data
suggest that fiber and calcium content, seed coat surface and micropyle structure
are related to water absorption of soybeans. Later, Arechavaleta-Medina and
Snyder (1981) reasoned that cuticle was most likely the site of the water barrier
in the seed coat of soybeans because soaking hard beans in methanol or ethanol
for 24 hr at 20 0 e made them permeable to water.
Soybean hulls are considered a by-product of soy processing. After separation
from seeds during rolling or flaking, they are toasted and ground, then blended
back with defatted soy meal to make a meal containing 44% protein. For a high
protein (47-49%) meal, the hull is not blended in, but rather it is disposed of
separately. At present, soy hulls are primarily used for animal feed. Since soy
carbohydrates are mainly composed of a-cellulose and hemicellulose, being low
in lignin, they are easily digested by animals. In fact, they are so highly digested
that their digestible energy content is essentially equal to grains. In addition, when
used in high forage diets, soy hulls eliminate the risk of acidosis (Klopfenstein and
Owen 1988).
For the last two decades, some new uses of soy hulls have been explored,
with an emphasis as a source of human food. Like dietary fiber from other
sources, soy hulls have been shown to reduce blood serum cholesterol levels
(Mahalko et al. 1984). So, soybeans hulls have been used as a fiber supplement
for bakery products at a level up to 10% (Johnson et al. 1985). Soy hulls are
also rich in iron; approximately, 32% of the total seed iron is in soy hulls (Levine
et al. 1982). Thus, they can be used as a iron supplement for such food as bakery
and breakfast cereals (Johnson et al. 1985, Lykken et al. 1987).
Chf'mistry and Nutritional Value of Soybean Components / 95
VII. HypocotyJ Axis
In addition to hulls and cotyledons, the third structural part of the soybean seed
is the hypocotyl axis, or germ. Upon germination, it will grow into a new soybean
plant. By weight, the hypocotyl axis is about 2.0% of the seed. In general, the
axis has a protein content similar to that of cotyledons but contains about 10%
less fat and 10% more insoluble carbohydrate than the cotyledon part (Table
2.1). After cracking the soybeans, the first step of soybean processing, the axis
may be separated with the cotyledon or with the hull, depending on which
structural part it adheres to.
Recently, Liu et al. (l995b) investigated the fatty acid composition of the axis
and its relationship with that of cotyledons. They found that the seed axis has
the lowest relative percentages of stearic and oleic acids and the highest of
linoleic and linolenic acids. Furthermore, regardless of the great variation in the
fatty acid profile of the whole seeds among six selected soybean genotypes, the
ratios in five major fatty acids between axis and cotyledons is highly conserved.
The average ratio of six genotypes for palmitic is 1.38 ± 0 .15; stearic, 0.86 ±
0.11; oleic, 0.27 ± 0.02; linoleic, 1.17 ± 0.12; and linolenic, 2.22 ± 0.22. This
finding has two significant implications. First, it implies that lipid metabolism
may be correlated in both axis and cotyledon tissues during seed development.
And second, by giving the fatty acid composition of one tissue, one can predict
that of the other. For example, if cotyledons are found to have 7.5% linolenic
acid relative to total fatty acids from the tissue, the relative percentage of linolenic
acid in the axis tissue would be around 16.7% (= 7.5% x 2.22).
It has been speculated that the hypocotyl axis is the source of the beany
flavor and undesirable taste in soy products. One explanation is that the soybean
hypocotyl has the highest percentage of polyunsaturated fatty acids (Liu et al.
1995b) and the highest concentration of isoflavones (Kudou 1991). This also
explains why some soyfood processors have succeeded in reducing the off flavor
in their products by removing the axis and hulls during processing (Tsukamoto
et al. 1991).
Because of the low proportion of the hypocotyl axis in the whole seed and
difficulty in separating it from the other parts commercially, relatively few studies
have been conducted on its food value. However, this does not necessarily mean
that the axis is insignificant. Considering the fact that isoflavone concentration
in the hypocotyl is about 5-6 times higher than in cotyledons (Kudou 1991).
there may be a potential for a new use of the soybean axis.
References
AACC, 1983. American Association of Cereal Chemists, Approved Methods. Method 71-
10, approved Nov. 1973, St. Paul, MN.
Abbott. T.P., Nabetani, H., Sessa, D.J., and Wold, W.J. 1996. Effects of bound water on
FTIR spectra of glycinin. J. Agric. Food Chern. 44:2220-2224.
Abdullah, A., Baldwin, R.E., and Minor, H. 1984. Germination effects on flatus-causing
factors and anti nutrients of mung beans and two strains of small-seeded soybeans. J.
Food Prot. 47:441.
Adlercreutz, H., Mousave, Y., Clark, J., Hocherstedt, K., HamaHiinen, E., Wahalii, K.,
MakeHi, T., and Hase, T. 1992. Dietary phytoestrogens and cancer: in vitro and in vivo
studies. J. Steroid Biochem. Mol. BioI. 41 :331-337.
Ahrens, E.H., Hirsch, J., Insull, W., Tsaltas, T.T., Blomstrand, R., and Peterson, M.L.
1957. The influence of dietary fats on serum lipid levels in man. Lancet 1:943-953.
Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S., Itoch, N., Shibuya,
M., and Fukami, Y. 1987. Genistein, a specific inhibitor of tyrosine protein kinase. J.
Bio!. Chern. 262:5592-5595.
Anderson, J.T., Grande, F., and Keys, A. 1961. Hydrogenated fats in the diet and lipids
in the serum of man. J. Nutr. 75:388.
Anderson, J.W., Johnstone, B.M., and Cook-Newell, M.L. 1995. Meta-analysis of the
effects of soy protein intake on serum lipids. N. Engl. J. Med. 333:276.
Anderson, R.L. 1992. Effect of steaming on soybean proteins and trypsin inhibitors. J.
Am. Oil Chern. Soc. 69: 1170-1176.
Anonymous 1991. The trouble with margarine. Consumer Report. March, p. 196-197.
Anthony, M.S., Clarkson, T.B., Hughes, c.L., Jr., Morgan, T.M., and Burke, G.L. 1996.
Soybean isoflavones improve cardiovascular risk factors without affecting the reproduc-
tive system of prepubertal Rhesus monkeys. J. Nutr. 126:43.
Applewhite, T.H. (Ed.) 1989. Proceedings of the World Congress: Vegetable Protein
Utilization in Human Foods and Animal Feedstuffs. American Oil Chemists' Society,
Champaign, IL.
Arechavaleta-Medina, F. and Snyder, H.E. 1981. Water imbibition by normal and hard
soybeans. J. Am. Oil Chern. Soc. 58:976.
Argos, P., Narayana, S. V.L., and Nielsen, N.C. 1985. Structural similarity between legumin
and vicilin storage proteins from legumes. EMBO J. 4:1111-1117.
ASCN/AIN. 1996. Position paper on trans fatty acids, American Society for Clinical
Nutrition/American Institute of Nutrition Task Force on Trans Fatty Acids. Am. J.
Clin. Nutr. 63:663-670.
Aspinall, G.O. 1988. Chemistry of soybean carbohydrates. In Proceedings of Soybean
Utilization Alternatives, L. McCann (Ed.), pp. 117-129. University of Minnesota, St.
Paul, MN. Feb. 16-18.
Astwood, J.D. and Fuchs, R.L. 1996. Preventing food allergy: emerging technologies.
Trends Food Sci. Technol. 7:219-226.
Axelrod, B., Cheesbrough, T.M., and Laakso, S. 1981. Lipoxygenase from soybeans. Ch.
53. In Methods in Enzymology, vol. 71, p.441. Academic Press, New York.
Badley, R.A., Atkinson, D., Hauser, H., Oldani, D., Green, J.P., and Stubbs, J.M. 1975.
The structural, physical and chemical properties of soybean protein glycinin. Biochim.
Biophys. Acta 412:214-228.
Chemistr.\' and Nutritional Value 0/ Sovbean Components / 97
Baker, E.C. and Mustakas, G.c. 1973. Heat inactivation of trypsin inhibitor, lipoxygenase,
and urase in soybeans: effect of acid and base additives. J. Am. Oil Chern. Soc. 50: 137.
Barnes, S., Kirk, M., and Coward, L. 1994. Isofiavones and their conjugates in soy
foods: extraction conditions and analysis by HPLC-mass spectrometry. J. Agric. Food
Chern. 42:2466-2474.
Bartholomew, R.M. and Ryan, D.S. 1980. Lack of mutagenicity of some phytoestrogens
in the Salmonella/mammalian microsome assay. Mutat. Res . 78:317-320.
Bates, R.P. and Matthews, R.F. 1975. Ascorbic acid and ~-carotene in soybeans as
influenced by maturity, sprouting, processing and storage. Proc. Fla. State Hort. Soc.
88:266-271.
Bau, H.M. and Debry, G. 1979. Germinated soybean protein products: chemical and
nutritional evaluation. J. Am. Oil Chern. Soc. 56: 160.
Beleia, A. , Thu Thao, L.T., and Ida, E.I. 1993. Lowering phytic phosphorus by hydration
of soybeans. J. Food Sci. 58:375-388.
Ben-Aziz, A. , Crossman, S., Ascarelli, I., and Budowski, P. 1970. Linoleate oxidation
induced by lipoxygenase and heme proteins: a direct spectrophotometric assay. Anal.
Biochem.34:88-100.
Bernal-Lugo, I., Castillo, A., Diaz de Leon, F., Moreno, E. , and Ramirez, J. 1991. Does
phytic acid influence cooking rate in common beans? J. Food Biochem. 15:367-374.
Blatny, P. , Kvasnicka, E, and Kenndler, E. 1995. Determination of phytic acid in cereal
grains, legumes, and feeds by capillary isotachophoresis. J. Agric. Food Chem. 43: 129-
133.
Birk, Y. 1961. Purification and some properties of a highly active inhibitor of trypsin and
a-chymotrypsin inhibitor from soya beans. Biochem. Biophys. Acta 54:378- 381.
Bodwell, C.E., and Marable, N.L. 1981. Effectiveness of methods for evaluating the
nutritional quality of soybean protein. J. Am. Oil Chem. Soc. 58:475-483 .
Bonanome. A. and Grundy. S.M. 1988. Effect of dietary stearic acid on plasma cholesterol
and lipoprotein levels. N. Engl. J. Med. 318: 1244.
Borchers. R., Mange. L.D .. Nelson. S.O .. and Stetson. L.E. 1982. Rapid improvement in
nutritional quality of soybeans by dielectric heating. 1. Food Sci. 37:333.
Borhan, M. and Snyder, H.E. 1979. Lipoxygenase destruction in wholc beans by combina-
tions of heating and soaking in ethanol. J. Food Sci. 44:586.
Bowman. D.E. 1944. Fractions derived from soybeans and navy beans which retard tryptic
digestions of casein. Proc. Soc. Exp. BioI. Med. 57: 139- 140.
Brown, BD .. Wei, L.S .. Steinberg. M.P., and Villota, R. 1982. Minimizing protein insolubi-
lization during thermal inactivation of lipoxygenase in soybean cotyledons. 1. Am. Oil
Chem. Soc . 59:88.
Burkitt, D.P. 1971. Epidemiology of cancer of the colon and rectum. Cancer 28:3 .
Burkitt, D.P. and Trowell . H.C. (Ed.) 1975. Refined Carbohydrate Foods and Disease.
Some Implications of Dietary .fiber. Academic Press, London.
Burks, A. W. , Williams, L.W., Helm, R.M., Thresher, W. , Brooks, 1.R., and Sampson, H.A.
1991. Ch. 22, Identification of soy protein allergens in patients with atopic dermatitis
and positive soy challenges; determination of change in allergenicity after heating or

enzymatic digestion. In Nutritional and Toxicological Consequences of Food Process-
ing, M. Friedman (Ed.). pp. 295-307. Plenum Press, New York.
Calderon de la Barca, A.M., Vazquez-Moreno, L., and Robles-Burgueno, M.R. 1991.
Active soybean lectin in foods: isolation and quantitation. Food Chern. 39:321.
Calloway, D.H. , Hickey, e.A., and Murphy, E.L. 1971. Reduction of intestinal gas-forming
properties of legumes by traditional and experimental food processing methods. J. Food
Sci. 36:25l.
Caragay, A.B. 1992. Cancer preventive foods and ingredients. Food Techno!. 46(4):65.
Carroll, K.K. and Kurowska, E. M. 1995. Soy consumption and cholesterol reduction:
review of animal and human studies. J. NutL (suppl.) 125 (3S):594S-597S.
Cassidy, A. Bingham, S., and Setchell, H.D.R. 1994. Biological effects of a diet of soy
protein rich in isoflavones on the menstrual cycle of premenopausal women. Am J.
Clin. NUtL 60:333.
Catsimpoolas, N. 1969. Isolation of glycinin subunit by isoelectric focusing in urea-
mercaptoethanol. FEBS Lett. 4:259-26l.
Catsimpoolas, N. and Ekenstam, e. 1969. Isolation of a, 13, and 't conglycinins . Arch.
Biochem. Biophys. 129:490.
Catsimpoolas, N., Campbell, T.G., and Meyer, E.W. 1969. Association-dissociation phe-
nomenon in glycinin. Arch. Biochem. Biophys. 131:577.
Cavins, J.F., Kwolek, W.E., Inglett, G.E., and Cowan, J.e. 1972. Amino acid analysis of
soybean meal: interlaboratory study. J. Asso. Off. Anal. Chern. 55:686-691.
Chang, R., Schwimmer, S. and Burr, H.K. 1977. Phytate: removal from whole dry beans
by enzymatic hydrolysis and diffusion. J. Food Sci. 42: 1098-1101.
Chapkin, R.S. 1992. Reappraisal of the essential fatty acids. Ch. 18. In Fatty Acids and
Their Health Implications, e.K. Chow (Ed.), pp. 429-435. Marcel Dekker, New York.
Che Man, Y.B., Wei, L.S., and Nelson, A.I. 1989. Acid inactivation of soybean lipoxygen-
ase with retention of protein solubility. J. Food Sci. 54(4):963.
Chen, B.H-Y. and Morr, e.V. 1985. Solubility and forming properties of phytate-reduced
soy protein isolate. J. Food Sci. 50: 1139-1142.
Chen, L.H. and Pan, S.H. 1977. Decrease of phytates during germination of pea seeds
(Pisum sativum). Nutr. Rep. Int. 16: 125.
Chernick, S.S., Lepkovsky, S.S., and Chaikoff, I.L. 1948. A dietary factor regulating the
enzyme content of the pancreas: changes induced in size and proteolytic activity of the
chick pancreas by the ingestion of raw soybean meal. Am. J. Physiol. 155:33.
Cheryan, M. 1980. Phytic acid interactions in food systems. CRC Crit. Rev. Food Sci.
Nutr. 13:297-335.
Chow, C.K. (Ed.). 1992. Fatty Acids and Their Health Implications. Marcel Dekker,
New York.
Christopher, J.P. and Axelrod, B. 1971. On the different positional specificities of peroxida-
tion of linoleate shown by two isozymes of soybean lipoxygenase. Biochem. Biophys.
Res. Commun. 44:731 - 736.
Chemistry and Nutritional Value o{ Soybean Components / 99
Coates, 1.B., Medeiros, 1.S., Thanh, V.H., and Nielsen, N.e . 1985. Characterization of
the subunits of ~-conglycinin. Arch. Biochem. Biophys. 243:184-189.
Chu, Y.-H. and Lin, J.-Y. 1993. Factors affecting the content of tocopherol in soybean
oil. 1. Am. Oil Chern. Soc. 70: 1263-1268.
Coon, e., Akavanichan, 0., and Cheng, T. 1988. The effect of oligosaccharides on the
nutritive value of soybean meal. In Proceedings of Soybean Utilization Alternatives,
L. McCann (Ed.), p. 203-214. University of Minn. St. Paul, MN.
Cooper, J.B ., Chen, J.A. , van Holst, G.-J. and Varner, J.E. 1987. Hydroxyproline-rich
glycoproteins of plant cell walls. TIES 12:24-27.
Coward, L. , Barnes, N.C., Setchell, K.D.R., and Barnes, S. 1993. Genistein, daidzein.
and their ~-glycoside conjugates: antitumor isoflavones in soybean foods from American
and Asian diets. 1. Agric. Food Chern. 41:1961-1967.
Cristofaro, E. , Mottu, F., and Wuhrmann, 1.J. 1974. Involvement of the raffinose family
of oligosaccharides in flatulence. Ch. 20. In Sugar in Nutrition, H.L. Sipple and K.W.
McNutt (Ed.). Academic Press, New York.
Cruz, R., Batistela, J.e., and Wosiaski, G. 1982. Microbial a-galactosidase for soymilk
processing. J. Food Sci. 46:1196.
de Muelenaere, H.J .H. 1964. Effect of heat treatment on the hemagglutinating activity of
legumes. Nature 201:1029.
Deshpande, S.S ., Chery an, M., and Salunkhe, D.K. 1984. Tannin analysis of food products.
CRC Crit. Rev. Food Sci. Nutr. 24:401-449 .
Desikachar, H.S .R. and De, S.S. 1947. Role of inhibitors in soybeans. Science 106:421.
DiPietro, C.M. and Liener, I.E. 1989. Heat inactivation of the Kunitz and Bowman-Birk
soybean protease inhibitors. J. Agric. Food Chern. 37:39-44.
Donatucci, D.A. 1983. The role of lectins in the nutritional toxicity of raw legumes. Ph.D.
dissertation, University of Minnesota, St. Paul, MN.
Dornbos. D.L. Jr. and Mullen, R.E. 1992. Soybean seed protein and oil contents and
fatty acid composition adjustments by drought and temperature. 1. Am. Oil Chern.
Soc. 69(3):228-231.
Eggum. B.O., and Beames. R.M. 1983. The nutritive value of seed proteins. In Seed
Proteins. Biochemistry. Genetics, Nutritive Value. E. Gottschalk and H.P. Muller (Ed.).
pp. 499-531. Martinus NijhoflJunk, Hague.
Eldridge, A. and Kwolek, W. 1983. Soybean isoflavones: effect of environment and variety
on composition. J. Agric. Food Chern. 31:394-396.
Eldridge. A.C., Warner, K .. and Wolf, W.J. 1977. Alcohol treatment of soybeans and
soybean protein products. Cereal Chern. 54: 1229.
Enig, M.G., Atal, S., Keeney, M., and Sampugna, 1. 1990. Isometric trans fatty acids in
the U.S. diet. 1. Am. Coli. Nutr. 9:471.
Erdman, 1.W. Jr. 1979. Oilseed phytates: nutritional implications. J. Am. Oil Chern .
Soc. 56:736-741.
Evans, e.D., McConnell , D.G., List, G.R., and Scholfield, e.R. 1969. Structure of unsatu-
rated vegetable oil glycerides: direct calculation from fatty acid composition. J. Am.
Oil Chern. Soc. 46:421.
FAO/WHO, 1985. Energy and Protein Requirements, FAOIWHO Nutrition Meetings,

Report Series 724. Food and Agriculture Organization/World Health Organization,
Geneva.
FAOIWHO, 1990. Protein Quality Evaluation. FAOIWHO Nutrition Meetings, Report
Series 5 l. Food and Agriculture Organization/World Health Organization, Rome.
Fenner, G.P. 1996. Low-temperature treatment of soybean (Glycine max) isoflavonoid
aglycon extracts improves gas chromatographic resolution. 1. Agric. Food Chern.
44(12):3727-3729.
Fernando, S.M. and Murphy, P.A. 1990. HPLC determination of thiamin and riboflavin
in soybeans and tofu. 1. Agric. Food Chern. 38:163-167.
Fleury, Y., Welti, D.H., Phillippossian, G., and Magnolato, D. 1992. Soybean (malonyl)
isoflavones characterization and antioxidant properties. In Phenolic Compounds in Food
and Their Effects on Health, Vol. II, M.-T. Huang, C.-T. Ho, and c.y. Lee, (Ed.), pp.
98-113. American Chemical Society, Washington, DC.
Fontes, E.P.B., Moreira, M.A., Davies, C.S., and Nielsen, N.C. 1984. Urea-elicited changes
in relative electrophoretic mobility of certain glycinin and [3-conglycinin subunits. Plant
Physiol. 76:840-843.
Frazier, P.l. 1979. Lipoxygenase action and lipid binding in breadmaking. Baker's Dig.
53(6):8-14.
Friedman, M. and Gumbmann, M.R., 1986. Nutritional improvement of soy flour through
inactivation of trypsin inhibitors by sodium sulfite. 1. Food Sci. 51(5):1239-1241.
Friedman, M., Brandon, D.L., Bates, A.H., and Hymowitz, T. 1991. Comparison of
commercial soybean cultivar and an isoline lacking the Kunitz trypsin inhibitor: compo-
sition, nutritional value, and effects of heating. 1. Agric. Food Chern. 39:327.
Fujimaki, M., Arai, S., Kirigaya, N., and Sakurai, Y. 1965. Studies of flavor compounds
in soybean. Part 1. aliphatic carbonyl compounds. Agric. BioI. Chern. 29:855-858 .
Fukushima, D. 1968. Internal structure of 7S and lIS globulin molecules in soybean
proteins. Cereal Chern. 45:203.
Fukutake, M., Takahashi, M., Ishida, K., Kawamura, H., Sugimura, T., and Wakabayashi,
K. 1996. Quantification of genistein and genistin in soybeans and soybean products.
Food Chern. Toxico!. 34(5):457-461.
Gauthier, S.F. , Vachon, c., lones, 1.D., and Savoie, L. 1982. Assessment of protein
digestibility by in vitro enzymatic hydrolysis with simultaneous dialysis. J. Nutr.
112:1718.
German, B., Damodaran, S., and Kinsella, J.E. 1982. Thermal dissociation behavior of
soy proteins. 1. Agric. Food Chern. 30:807-811.
Green, G.M., Olds, B.A., Mathews, G., and Lyman, R.L. 1973. Protein as a regulator of
pancreatic enzyme secretion in the rat. Proc. Soc. Exp. BioI. Med. 142:1162-1167.
Green, G.M. and Lyman, R.L. 1972. Feedback regulation of pancreatic enzyme secretion
in rats. Proc. Sci. Exp. BioI. Med. 140:6-12.
Gumbmann, M.R., Spangler, W.L., Dugan, G.M., and Rackis, J.l. 1986. Safety of trypsin
inhibitors in the diet: effects on the rat pancreas of long-term feeding of soy flour and
Chemistry and Nutritional Value of Soybean Components / JOJ
soy protein isolate. In Nutritional and Toxicological Significance of Enzyme Inhibitors

in Foods, M. Friedman (Ed.), p. 33. Plenum Press, New York.
Guzman, G .J. and Murphy, P.A. 1986. Tocopherols of soybean seeds and soybean curd
(tofu). J. Agric. Food Chern. 34:791-795.
Hajika, M., Igita, K., and Nakazawa, Y. 1991. A line lacking all the seed Jipoxygenase
isozymes in soybean [Glycine max (L.) Merrill] induced by gamma-ray irradiation.
Japan. J. Breed. 41 :507-509.
Hammerstrand, G.E., Black, L.T., and Glover, J.D. 1981. Trypsin inhibitors in soy products:
modification of the standard analysis procedure. Cereal Chern. 58:42-45.
Hammond, E.G. and Glatz, B.A. 1989. Biotechnology applied to fats and oils. Ch. 6. In
Developments in Food Biotechnology, R. King and P.S.J. Cheetham (Ed.), Vo!. 2 pp.
173-217. John Wiley & Sons, New York.
Harada, K., Toyokawa, Y., and Kitamura, K. 1983. Genetic analysis of the most acidic
lIs globulin subunit and related characters in soybean seeds. Japan. J. Breed. 33:23-30.
Hashizume, K., Nakamura, N. , and Watanabe, T. 1975. Influence of ionic strength on
conformation changes of soybean proteins caused by heating, and relationship of its
conformation changes to gel formation. Agric. BioI. Chern. 39:1339-1347.
Hefle, S.L. 1996. The Chemistry and Biology of Food Allergens. Food Techno!.
50(3)86-92.
Hegsted, D.M., McGandy, R.B., Myers, M.L., and Stare, FJ. 1965. Quantitative effects
of dietary fat on serum cholesterol in man. Am. J. Clin. Nutr. 17:281-295.
Hessing, M., Bleeker, H., van Biert, M., Vlooswijk, R.A.A. , and van Oort, M.G. 1994.
Antigenicity of legume proteins. Food Agric. Immunol. 6:315-320.
Hettiarachchy, N.S. and Ziegler, G.R. (Ed.) 1994. Protein Functionality in Food Systems.
1FT Basic Symposium Series. Marcel Dekker, Inc. New York, NY.
Hildebrand. D.F. and Hymowitz. T . 1981. Two soybean genotypes lacking lipoxygenase-I.
J. Am. Oil. Chern. Soc. 58 :583-586.
Hildebrand. D.F.. Hamilton-Kemp, T.R .. Legg. C.S., and Bookjans. G. 1988. Plant lipoxy-
genases: occurrence. properties. and possible functions. Current Topics Plant Biochem.
Physiol. 7:201-219.
Howard, P.A. , Lehnhardt, W.F. , and Orthoefer, F.T. 1983 . 7S and lIS vegetable protein
fraction and isolation. U.S. Patent 4,368,151, Jan . ll.
Hsu, H.W. , Vavak, D.L. . Satterlee, L.D. , and Miller, G.A. 1977. A multienzyme technique
for estimating protein digestibility. J. Food Sci. 42:1269.
Huang, A.S., Hsieh, O.A.L. , and Chang, S.S. 1981. Characterization of the nonvolatile
minor constituents responsible for the objectionable taste of defatted soybean flour. J.
Food. Sci. 47: 19.
Hurburgh, C.R. Jr. 1994. Long-term soybean composition patterns and their effect on
processing. J. Am. Oil Chern. Soc. 71(12): 1425-1427.
Hutton, K. and Foxcroft, p.o. 1975. Effect of processing temperature on some indices of
nutritional significance for micronized soya beans. Proc. Nutr. Soc. 34:49A.
Hymowitz, T., Collins, F.I., Panczner, J., and Walker, W .M. 1972. Relationship between
the content of oil, protein, and sugar in soybean seed. Agron . J. 64:613-616.
Hwang, K.M., Murphreee, S.A., and Sartorelli, AC. 1974. A quantitative spectrophotomet-
ric method to measure plant lectin-induced cell agglutination. Cancer Res. 34:3396.
Ibrahim, N. , Puri, R.K., Kapila, S., and Unklesbay, N. 1990. Plant sterols in soybean
hulls. J. Food Sci. 55:271-272.
Ikeda, K., Matsuda, Y., Katsumaru, A., Teranishi, M. , Yamanoto, T., and Kishida, M.
1995. Factors affecting protein digestibility in soybean foods . Cereal Chern. 72(4):401.
Jacks, T.J., Yatsu, L.Y., and Altschul, A.M. 1967. Isolation and characterization of peanut
spherosome. Plant Physiol. 42:585-597.
Johnson, C.D., Berry, M.F., and Weaver, C.M. 1985. Soybean hulls as an iron source for
bread enrichment. J. Food Sci. 50:1275-1277.
Jones, P.M .B . and Boulter, D. 1983 . The cause of reduced cooking rate in Phaseolus
vulgaris following adverse storage conditions. J. Food Sci. 48:623-627.
JSNFS, 1984. Rice, soybean, and fish-Scientific approach to staple foods. In Japanese
Society of Nutrition and Food Science. M. Fujimaki , G. Inoue, and T. Tanaka, (Ed.),
Kouseikan Publishing, Tokyo.
Kakade, M.l., Simons, N., and Liener, I.E. 1969. An evaluation of natural vs . synthetic
substances for measuring the antitryptic activity of soybean samples. Cereal Chern.
46:518-526.
Kakade, M.L. , Hoffa, D.E., and Liener, I.E. 1973. Contribution of trypsin inhibitors to
the deleterious effects of unheated soybeans fed to rats. J. Nutr. 103: 1172.
Kaul, R., Read, J., and Mattiasson, B. 1991. Screening for plant lectins by latex agglutina-
tion test. Phytochemistry 30:4005.
Kellor, R.L. 1974. Defatted soy flour and grits. J. Am. Oil Chern Soc. 51:77A-80A.
Kennedy, A.R. 1993. Overview: anticarcinogenic activity of protease inhibitors. In Prote-
ase Inhibitors as Cancer Chemopreventive Agents, W. Troll and A.R. Kennedy (Ed.),
pp. 9-64. Plenum Publishing, New York.
Kennedy, AR. 1994. Prevention of carcinogenesis by protease inhibitors. Cancer Res .
(Suppl.) 54: 1999s-2005s.
Keys, A, Anderson, J.T., and Grade, F. 1957. Prediction of serum-cholesterol responses
of man to changes in fats in diet. Lancet 2:955-966.
Kikuchi, T.S. , IshiI, S., Fukushima, D. , and Yokotsuka, T. 1971. Food chemical studies
on soybean polysaccharides. Part I. Chemical and physical properties of soybean cell
wall polysaccharides and their changes during cooking. J. Agric. Chern. Soc. 45:228 .
Kim, Y.A and Barbeau, W.E. 1991. Evaluation of SDS-PAGE method for estimating
protein digestibility. J. Food Sci. 56(4):1082-1086.
Kim, C.-S., Kamiya, S., Sato, T., Utsumi, S., and Kito, M. 1990. Improvement of nutritional
value and functional properties of soybean glycinin by protein engineering. Protein
Eng. 3:725-73l.
Kinney, AJ. 1996. Soybean biotechnology: improving soybean seed quality by genetic
Cht'lnislry and Nutritional Value o{ Soybean Components / 103
engineering. Paper No. 21A, presented at 87th American Oil Chemists' Society Annual
Meeting & Expo, Indianapolis, IN. April 28-May 1.
Kitamura, K. 1984. Biochemical characterization of lipoxygenase lacking mutants, L-I-
less, L-2-less, and L-3-less soybeans. Agric. BioI. Chern. 48:2339-2346.
Kitamura, K. 1995. Genetic improvement of nutritional and food processing quality in
soybean. Jap. Agric. Res. Quart. 29: 1-8.
Kitamura, K., Toyokawa, Y., and Harada, K. 1980. Polymorphism of glycinin in soybean
seeds. Phytochemistry 19: 1841-1843.
Kitamura, K., Davies, C.S .. Katzuma, N., and Nielsen, N.C. 1983. Genetic analysis of a
null-allele for lipoxygenase-3 in soybean seed. Crop. Sci. 23:924-927.
Klopfenstein, T. and Owen, F. 1988. Feeding soybean hulls to cattle. In Proceedings on
Soybean Utilization Alternatives, L. McCann (Ed.), Feb. 16-18. University of Minne-
sota, St. PauL MN.
Knuckles, B.E. and Betschart, A.A. 1987. Effect of phytate and other myo-inositol phos-
phate esters on amylase digestion of starch. J. Food Sci. 52:719.
Kobayashi, A., Tsuda, Y., Hirata. N., Kubota, K., and Kitamura, K. 1995. Aroma constit-
uents of soybean [Glycine max (L.) Merrill] milk lacking lipoxygenase isozymes. J.
Agric. Food Chern. 43:2449-2452.
Kohle, H. and Kauss, H. 1980. Improved analysis of hemagglutination assays for quantita-
tion of lectin activity. Anal. Biochem. 103:227.
Kohyama, K. and Nishinari, K. 1993. Rheological studies on the gelation process of
soybean 7S and II S proteins in the presence of glucono-8-lactone. J. Agric. Food
Chern. 41:8-14.
Koide, T .. Tsunasawa. S., and Ikenaka. T. 1973. Studies on soybean trypsin inhibitors
and amino acid sequence around the reactive site of soybean trypsin inhibitor (Kunitz).
Eur. J. Biochem. 32:408-416.
Kon, S. and Sanshuck. D. 1981. Phytate content and its effect on cooking quality of
beans. J. Food Sci. 44: 1329-1335.
Kris-Etherton. P.M. and Nicolosi. R.J. 1995. TrailS Fauy Acids and Coronar\' Heart
Disease Risk. International Life Sciences Institute (lLSI) Press. Washington. D.C.
pp. 1-24.
Krogdahl, A. and Holm, H. 1981. Soybean proteinase inhibitors and human proteolytic
enzymes: selective inactivation of inhibitors by treatment with human gastric juice . .J .
Nutr. III :2045.
Kudou, 5 .. Fleury, Y .. Welti. D., Magnolato, D .. Uchida, T .. Kitamura. K., and Okubo.
K. 1991. Malonyl isofiavone glucosides in soybean seeds (Glycine max Merrill). Agric.
BioI. Chern. 55:2227-2233.
Kunitz, M. 1945. Crystallization of a trypsin inhibitor from soybeans. Science 101 :668-
669.
Lalles, J.P., Dreau, D., Salmon, H., and Toullec, R. 1996. Identification of soyabean
allergens and immune mechanisms of dietary sensitivities in preruminant calves. Res.
Vet. Sci. 60:111-116.
Lalles, J.P. and Peltre, G. 1996. Biochemical features of grain legume allergens in humans
and animals. Nutri. Rev. 54(4):101-107.
Lehnhardt, W.F., Gibson, P.W., and Orthoefer, F.T. 1983. Fraction and isolation of 7S
and 11 S protein from isoelectrically precipitated vegetable protein mixtures. U.S. patent
4,370,267, Jan. 25.
Lei, M.-G. and Reeck, G.R. 1987. Two-dimensional electrophoretic analysis of the proteins
of isolated soybean bodies and of the glycosylation of soybean proteins. J. Agric. Food
Chern. 35:296-300.
Levine, S.E., Weaver, C.M., and Kirleis, A.W. 1982. Accumulation of selected trace
elements in hydroponically grown soybeans and distribution of the elements in processed
soybean fractions. J. Food Sci. 47: 1283.
Liener, I.E. 1953. Soyin, a toxic protein from the soybean. I. Inhibition of rat growth. 1.
Nutr. 49:527.
Liener, I.E. 1955. The photometric determination of the hemagglutinating activity of soyin
and crude soybean extracts. Arch. Biochem. Biophys. 54:223.
Liener, I.E. 1994. Implications of antinutritional components in soybean foods. CRC Crit.
Rev. Food Sci. Nutr. 34(1):31-67.
Liener, I.E., Deuel, H.J. Jr., and Frevold, H.L. 1949. The effect of supplemental methionine
on the nutritive value of diets containing concentrates of the soybean trypsin inhibitor.
J. Nutr. 39:325.
Liener, I.E., Goodale, R.L., Deshmukh, A., Satterberg, T.L., Ward, G., DiPietro, C.M.,
Bankey, P.E., and Borner, J.W. 1988. Effect of a trypsin inhibitor from soybeans
(Bowman-Birk) on the secretory activity of the human pancreas. Gastroenterology
94:419-427.
Liener, I.E. and Kakade, M. 1980. Protease inhibitors. In Toxic Constituents of Plant
Foodstuffs, I.E. Liener (Ed.), pp. 7-71. Academic Press, New York.
List, G.R., Evans, C.D., Warner, K., Beal, R.E., Kwolek, W.F., Black, L.T., and Moulton,
K.J. 1977. Quality of oil from damaged soybeans. J. Am. Oil. Chern. Soc. 54:8.
Liu, K.S. 1986. Effects of processing and maturation on certain antinutritional factors in
soybeans. M.S. thesis, Michigan State University, East Lansing, MI.
Liu, K.S. 1995. Cellular, biological and physicochemical basis for the hard-to-cook defect
in legume seeds. CRC Crit. Rev. Food Sci. Nutr. 35(4):263-298.
Liu, K.S. and Markakis, P. 1987. Effect of maturity and processing on the trypsin inhibitor
and oligosaccharides of soybeans. J. Food Sci. 52(1):222-223, 225.
Liu, K.S. and Markakis, P. 1989a. Trypsin inhibition assay as related to limited hydrolysis
of inhibitors. Anal. Biochem. 178:159-165.
Liu, K.S. and Markakis, P. 1989b. An improved colorimetric method for determining
antitryptic activity in soybean products. Cereal Chern. 66:415-422.
Liu, K.S. and Markakis, P. 1990. Effect of the rcactant mixing sequcnce on the chymotryp-
sin inhibition assay. Analyst 115:1143-1145.
Liu, K.S. and Markakis, P. 1991. Aqueous ethanol extraction of soybean trypsin inhibitors
and characterization of a calcium-sensitive fraction. J. Food Biochem. 15:159-168.
Liu, K.S., Markakis, P., and Smith, D. 1990. Trypsin inhibition by free fatty acids and
stearoyl-CoA. J. Agric. Food Chem. 38: 1475-1478.
Liu, KS., Orthoefer, F., and Brown, E.A. I 995a. Association of seed size with genotypic
variation in the chemical constituents of soybeans. J. Am. Oil Chem. Soc. 72(2): 191.
Liu, KS., Brown, E.A., and Orthoefer, F. 1995b. Fatty acid composition within each
structural part and section of a soybean seed. J. Agric. Food Chem. 43:381-383.
Lolas, G.M., Palamidas, N., and Markakis, P. 1976. The phytic acid-total phosphorus
relationship in barley, oats, soybeans, and wheat. Cereal Chern. 53:876.
Lotan, R.H., Sieggelman, W., Lit, H., and Sharon, N. 1974. Subunit structure of soybean
agglutinin. J. BioI. Chem. 249: 1219.
Lott, J.N.A. and Buttrose, M.S. 1978. Glohoids in protein bodies oflegume seed cotyledons.
Aust. J. Plant Physiol. 5:89-111.
Lykken, G.I., Hunt, J.R., Nielsen, E.J., and Dintzis, F.R. 1987. Availability of soybean
hull iron fed to humans in a mixed Western meal. J. Food Sci. 52:1545-1548.
MacLeod, G. and Ames, J. 1988. Soy flavor and its improvement. CRC Crit. Rev. Food
Sci. Nutr. 27(4):219-401.
McNiven, M.A., Grimmelt, B., MacLeod, J.A., and Voldeng, H. 1992. Biological character-
ization of a low trypsin inhibitor soybean. J. Food Sci. 57(6):1375-1377.
Maga, J.A. 1982. Phytate: its chemistry, occurrence, food interactions, nutritional signifi-
cance, and methods of analysis. J. Agric. Food Chern. 30:1-9.
Magee, A.C. 1963. Biological responses of young rats fed diets containing genistin and
genistein. J. Nutr. 80:151-156.
Mahalko, J.R., Sandstead, H.H., Johnson, L.K., Inman, L.F., Milne, D.B., Warner, R.e..
and Haubnz, E.A. 1984. Effect of consuming fiber from com bran, soy hulls, or
apple powder on glucose tolerance and plasma lipids in type II diabetes. Am. J. Clin.
Nutr. 39:25.
Makower, R.U. 1970. Extraction and determination of phytic acid in beans (Phaseolus
I'ulgaris). Ccreal Chem. 47:288.
Marczy. 1.S., Simon. M.L.. Mozsik, L.. and Szajani. B. 1995. Comparative study on the
lipoxygenase activities of some soybean cultivars. J. Agric. Food Chem. 43(2):313.
Martin. M.L Hulley, S.B .. Browner, W.S., Kuller, L.H., and Wentworth, D. 1986. Serum
cholesterol, blood pressure. and mortality: implications from a cohort of 361,662 men.
Lancet 2:933-936.
Masai, T., Wada, K, Hayakawa, K, Yoshihara, \., and Mitsuoka, T. 1987. Effects 01
soybean oligosaccharides on human intestinal flora and metabolic activities. Japan].
Bacteriol. 42(1 ):313.
Mason, A.C., Weaver, e.M .. Kimmel, S., and Brown, R.K. 1993. Effect of soybean
phytate content on calcium bioavailability in mature and immature rats. J. Agric. Food
Chem. 41:246-249.
Mattson, F.M. and Grundy, S.M. 1985. Comparison of effects of dietary saturated, monosa-
turated, and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. 1.
Lipid Res. 26: 194-202.
iVlensink, R,P, and Katan, M.B. 1989. Effect of a diet enriched with monounsaturated or
polyunsaturated fatty acids on levels of low-density and high-density lipoproteins in
healthy women and men. New Eng!. J. Med. 321:436-44l.
Mensink, R.P. and Katan, M.B. 1990. Effect of dietary trans fatty acids on high-density
and low density lipoprotein cholesterol levels in healthy subjects. N. Eng!. J. Med.
323:439-445.
Messina, M. and Barnes, S. 1991. The role of soy products in reducing cancer risk. J.
Nat!. Cancer Inst. 83:541-546.
Messina, M., Messina, V., and Setchell, KD.R. 1994. The Simple Soybean and Your
Health. Avery Publishing Group, Garden City Park, New York.
Mittal, B.K and Steinkraus, KH. 1975. Utilization of oliogosaccharides by lactic acid
bacteria during fermentation of soymilk. J. Food Sci. 40: 114.
Montelongo, J,-L., Chassy, B.M., and McCord, J.D. 1993. Lactobacillus salivarius for
conversion of soy molasses into lactic acid. J. Food Sci. 58(4):863-866.
Moreira, M.A., Hermodson, M.A., Larkins, B.A., and Nielsen, N.C. 1979. Partial character-
ization of the acidic and basic polypeptides of glycinin. J. Bio!. Chern. 254:9921-9926.
Moreira, M.A, Tavares, S.R., Ramos, V., and de Barros, E.G. 1993. Hexanal production
and TBA number are reduced in soybean [Glycine max (L.) Merr.J seeds lacking
lipoxygenase isozymes 2 and 3. J. Agric. Food Chern. 41: 103-1 06.
Morita, S., Fukase, M., Hoshino, K, Fukuda, Y., Yamaguchi, M., and Morita, Y. 1994.
A serine protease in soybean seeds that acts specifically on the native ex subunit of ~
conglycinin. Plant Cell Physio!. 35(7): 1049-1056.
Morita, S., Fukase, M., Yamaguchi, M., Fukuda, Y., and Morita, Y. 1996. Purification,
characterization, and crystallization of single molecular species of ~-conglycinin from
soybean seeds. Biosci. Biotech. Biochem. 60(5): 866-873.
Moroz, L.A. and Yang, W.H. 1980. Kunitz soybean trypsin inhibitor. A specific allergen
in food anaphylaxis. New Eng!. J. Med. 15:1126-1128.
Murasawa, H., Sakamoto, A., Sasaki, H., and Harada, K. 1991. The effect of glycinin
subunit on tofu-making. In Japan Part Proceedings of the International Conference
on Soybean Processing and Utilization, K Okubo (Ed.). pp. 53-57.
Murphy, P.A. 1985. Structural characteristics of soybean glycinin and ~-conglycinin. In
World Soybean Research Conference III: Proceedings, R. Shibles (Ed.), pp. 143-151.
Westview Press, Boulder, CO.
Murphy, P.A. and Resurrection, AP. 1984. Varietal and environmental differences in
soybean glycinin and ~-conglycinin concentration. J. Agric. Food Chern. 32:911.
Mustakas, G.c., Albrecht, W.J., McGhee, J.E., Black, L.T., Bookwalter, G.N., and Griffin,
E.L. Jf. 1969. Lipoxidase deactivation to improve stability, odor and flavor of full-fat
soy flours. J. Am. Oil Chern. Soc. 46:623.
Nairn, M., Gestetner, B., Bondi, A, and Birk, Y. 1976. Antioxidative and antihemolytic
activities of soybean isoflavones. J. Agric. Food Chern. 24:1174-1177.
Naismith, W.E.F. 1955. Ultracentrifuge studies on soya bean protein. Biochim. Biophys.
Acta 16:203-210.
Nakamura, T., Utsumi, S., and Mori, T. 1986. Mechanism of heat-induced gelation and
gel properties of soybean 7S globulin. Agric. BioI. Chern. 50: 1287-1293.
Neff, W.E., Selke, E., Mounts, T.L., Rinsch, W., Frankel, E.N., and Zeitoun, M.A.M.
1992. Effect of triacylglycerol composition and structures on oxidative stability of oils
from selected soybean germplasm. 1. Am. Oil Chern. Soc. 69(2): 111-118.
Nelson, T.S., Shieh, T.R., Wodzinski, R.L., and Ware, I.H. 1968. The availability of
phytate phosphorus in soybean meal before and after treatment with a mold phytase.
Poult. Sci. 47:1842.
Nelson, A.I., Steinberg, M.P., and Wei, L.S. 1976. Illinois process for separation of
soymilk. 1. Food Sci. 41 :57-61.
Nielson, N.C. 1985a. Structure of soy proteins. Ch. 2. In New Protein Foods, Vol 5. Seed
Storage Proteins, A.M. Altschul and H.L. Wilcke (Ed.), pp. 27-64. Academic Press,
Orlando, FL.
Nielsen, N.C. 1985b. Structure and complexity of the lIS polypeptides in soybeans. 1.
Am. Oil Chem. Soc. 62(12):1680-1685.
Nishiba, Y., Furuta, S., Hajika, M., Igita, K., and Suda, I. 1995. Hexanal accumulation
and DETBA value in homogenate of soybean seeds lacking two or three lipoxygenase
isozymes. 1. Agric. Food Chern. 43(3):738-741.
Nonaka, M., Toiguchi, S .. Sakamoto, H., Kawajiri, H., Soeda, T., and Motoki, M. 1994.
Changes caused by microbial transglutaminase on physical properties of thermally
induced soy protein gels. Food Hydrocolloids, 8( I): 1-8.
NRC-NAS. 1972. Nutritional Requirements of Laboratory Animals. No. 10. 2nd Rev. Ed.
Obata, A., Matsuura, M., and Kitamura, K. 1996. Degradation of sulfhydryl groups in
soymilk by lipoxygenases during soybean grinding. Biosci. Biotech. Biochem.
60(8):1229-1232.
Odani, S. and Ikenaka, T. 1973. Studies on soybean trypsin inhibitors VIII. Disulfide
bridges in soybean Bowman-Birk protease inhibitors. J. Biochem. (Tokyo) 74:697.
O·Dell. B.L. 1979. Effect of soy protein on trace mineral availability. In Soy Protein and
Human Nutrition. H.L. Wilckc, D.R. Hopkins, and D.H. Waggle, (Ed.) Academic Press,
New York.
O'Dell. B.L. and deBoland, A. 1976. Complexation of phytate with proteins and cations
in com germ and oilseed meals. 1. Agric. Food Chem. 24:804.
Ogawa, L Bando, N .. Tsuji, H., Okajima, H., Nishikawa, K. and Sasaoka, K. 1991.
Investigation of the IgE-binding proteins in soybeans by immunoblotting with the sera
of the soybean-sensitive patients with atopic dermatitis. J. Nutri. Sci. Vitaminol. 37:555-
565.
Ogawa, T., Tsuji, H., Bando, N., Kitamura, K., Zhu, Y.L., Hirano, H., and Nishikawa,
K. 1993. Identification of the soybean allergenic protein, Gly m Bd 30 K, with the soybean
seed 34-kDa oil-body-associated protein. Biosci. Biotech. Biochem. 57: I 030-1 033.
Okubo, K., Iijima, M., Kobayashi, Y., Yoshikoshi, M., Uchida, T., and Kudou, S. 1992.
Components responsible for the undesirable taste of soybean seeds. Biosci. Biotechnol.
Biochem.56:99-103.
J08 / Soybeans: Chemistry, Technology, and Utilization
Olson, A., Gray, G.M., and Chiu, M.-C. 1987. Chemistry and analysis of soluble dietary
fiber. Food Techno!., Feb. pp. 71-80.
Orf, J.H. 1988. Modifying soybean composition by plant breeding. In Proceedings: Soy-
bean Utilization Alternatives, L. McCann (Ed.). p. 131. University of Minnesota, St.
Paul, MN. Feb. 16-18.
Orf, J.H. and Hymowitz, T.H. 1979. Inheritance of the absence of the Kunitz inhibitor
in seed protein of soybeans. Crop Sci. 19: 107.
Osborne, T.B. and Mendel, L.B. 1917. The use of soybean as food. J. Bio!. Chern. 32:369.
Ozawa, K. and Laskowski, M. Jr. 1966. The reactive site of trypsin inhibitor. J. Bio!.
Chern. 241:3955-3961.
Pallansch, M.J. and Liener, I.E. 1953. Soyin, a toxic protein from the soybean. II. Physical
characterization. Arch. Biochem. Biophys. 45:366.
Park, D.K., Terao, J., and Matsushita, S. 1983. Influence of the positions of unsaturated
acyl groups in glycerides on autoxidation. Agric. BioI. Chern. 47:2251-2255.
Perkins, E. G. 1995. Composition of soybeans and soy products. In Practical Handbook
of Soybean Processing and Utilization, D.R. Erickson (Ed.), pp. 9-28. AOCS Press,
Champaign, IL.
Pernollet, J-c. 1978. Protein bodies of seeds: ultrastructure, biochemistry, biosynthesis
and degradation. Phytochemistry 17:1473-1480.
Pernollet, J-c. and Mosse, J. 1983. Structure and location of legume and cereal seed
storage proteins. No. 20. In Seed Proteins, J. Daussant, J. Mosse, and J. Vaughan (Ed.),
Annual Proceedings of the Phytochemical Society of Europe, pp. 155-190. Academic
Press, London.
Peters, J. and Czakor, B. 1989. Effect of extrusion cooking on trypsin inhibitor activity.
Nahrung 33:275.
Peterson, T.G. and Barnes, S. 1993. Genistein and biochanin A inhibit the growth of
human prostate cancer cells, but not epidermal growth factor receptor tyrosine autophos-
phorylation. Prostate 22:335-345.
Pour-El, A., Nelson, S.O., Peck, E.E., Tjhio, B., and Stetson, L.E. 1981. Biological
properties of VHF- and microwave-heated soybeans. J. Food Sci. 46:880-885, 895.
Pratt, D.E. and Birac, P. M. 1979. Source of antioxidant activity of soybean and soy
products. J. Food Sci. 44: 1720-1722.
Prattley, c.A. and Stanley, D.W. 1983. Protein-phytate interactions in soybeans. I. Local-
ization of phytate in protein bodies and globoids. J. Food Biochem. 6:243-253.
Pryde, E.H. 1980. Composition of soybean oil. Ch. 2. In Handbook of Soy Oil Processing
and Utilization. S.R. Erickson, E.H. Pryde, O.L. Brekke, T.L. Mounts, and R.A. Falb,
(Ed.), p. 13. American Oil Chemists' Society, Champaign, IL.
Pull, S.P., Pueppke, S.G., Hymowitz, T., and Orf, H.H. 1978. Soybean lines lacking
120,OOO-Da seed lectin. Science 200:1277 .
Rackis, J.J. 1974. Biological and physical factors in soybeans. J. Am. Oil Chern. Soc.
51:161A.
Rados, J.J., Honig, D.H., Sessa, D.I., and Steggerda, F.R. 1970. Flavor and flatulence
factors in soybean protein products. I. Agric. Food Chern. 18:977.
Read, R.J. and lame, M.N.D. 1986. Introduction to the protease inhibitors: X-ray crystallog-
raphy. In Protease Inhibitors, A.I. Barrett and G. Salvesen (Ed.). Elsevier, Amsterdam!
New York.
Reseland, I.E., Holm, H., Jacobsen, M.B., Jenssen, T.G., and Hanssen, L.E. 1996. Protein-
ase inhibitors induce selective stimulation of human trypsin and chymotrypsin secretion.
I. Nutr. 126(3):634-642.
Rios-Iriarte, B.J. and Barnes, R.H. 1966. The effect of overheating on certain nutritional
properties of the protein of soybeans. Food Techno!. 20:836.
Ritter, M.A., Morr, C.V., and Thomas, R.L. 1987. In vitro digestibility ofphytate-reduced
and phenolics-reduced soy protein isolates. J. Food Sci. 52(2):325-341.
Robinson, D.S., Wu, Z., Domoney, c., and Casey, R. 1995. Lipoxygenases and the quality
of foods. Food Chern. 54:33-43.
Saio, K. 1976. Soybeans resistant to water absorption. Cereal Foods World. 21: 168.
Saio, K. and Watanabe, T. 1966. Preliminary investigation on protein bodies of soybean
seeds. Agric. Bio!. Chern. 30: 1133-1138.
Saio, K. and Watanabe, T. 1968. Observation of soybean foods under electron microscope.
J. lap. Soc. Food Sci. Techno!. 15:290-296.
Saio, K., Kamiya, M., and Watanabe, T. 1969. Food processing characteristics of soybean
11 Sand 7S proteins. Part I. Effect of difference of protein component among soybean
varieties on formation of tofu-gel. Agric. Bio!. Chern. 33:1301-1308.
Saio, K., Yamagishi, T. and Yamauchi, F. 1986. Quantitative analysis of soybean proteins
by densitometry on gel electrophoresis. Cereal Chern. 63(6):493-496.
Salunkhe, D.K., Sathe, S.K., and Reddy, N.R. 1983. Legume lipids. In Chemistry and
Biochemistry of Legumes, S.K. Arora (Ed.) Edward Arnold Puh. Ltd., London.
Samoto, M., Akasaka, T., Mori, H., Manabe, M., Ookura, T., and Kawamura, Y. 1994.
A simple and efficient procedure for removing the 34 KD allergenic soybean protein,
Gly m I, from defatted soy milk. Biosci. Biotech. Biochem. 58(1 I ):2123-2125.
Samoto, M., Miyazaki, C., Akasaka, T., Mori, H .. and Kawamura. Y. 1996. Specific
binding of allergenic soybean protein Gly m Bd 30 K with a'- and a-subunits of
conglycinin in soy milk. Biosci. Biotech. Biochem. 60(6):1006-1010.
Sampson, H.A. and McCaskill, c.c. 1985. Food hypersensitivity and dermatitis: evaluation
of I 13 patients. I. Pediatr. 107:669-675.
Sandberg, A.S., Carlsson, N.G., and Svanberg, U. 1989. Effect of inositol, tri-, tetra-.
penta-, and hexaphosphates on in vitro estimation of iron availability. J. Food Sci.
54(1): 159-161.
Sattar, A., Neelofar, and Akhtar, M.A. 1990. Irradiation and germination effects on phytate,
protein, and amino acids of soybeans. Qua!. Plant. Plant Foods Hum. Nutr. 40: 185.
Satterlee, L.D., Kendrick, J.G., and Miller, G.A. 1977. Rapid in-vitro methods of measuring
protein quality. Food Techno!. 31(6):77.
Senti, F.R., (Ed.) 1985. Health Aspects of Dietary Trans Fatty Acids. Life Sciences
110 / Suybeans: Chemistry, Technulogy, and Utilization
Research Office, Federation of American Societies for Experimental Biology,

Bethesda, MD,
Sharon, N. and Lis, H. 1972. Lectins: cell-agglutinating and sugar-specific proteins. Sci-
ence 177:949-959.
Shibasaki, M., Suzuki, S., Tajima, S., Nemoto, H., and Kuroume T. 1980. Allergenicity
of major components of soybean. Int. Arch. Allergy AppI. Immun. 61:441-448.
Simons, P.C.M., Verseegh, H.AJ., Jongbloed, A.W., Kemme, P.A., Slump, P., Bos, K.D.,
Wolters, M.G.E., Beudeker, RF., and Verschoor, G.J. 1990. Improvement of phosphorus
availability by microbial phytase in broilers and pigs. J. Nutr. 64:225.
Simopoulos, AP. 1989. Executive summary, in Dietary w3 and w6 Fatty Acids-Biological
Effects and Nutritional Essentiality, C. Galli and A.P. Simopoulos (Ed.), pp. 391-404.
Plenum, New York.
Sirtori, C.R., Even, R, and Lovati, M.R. 1993. Soybean protein diet and plasma cholesterol:
from therapy to molecular mechanisms. Ann. N. Y. Acad. Sci. 676: 188.
Sissions, J.W. and Tolman, H. 1991. Anti-nutritional properties of soyabean antigens in
calves. In Toxic Factors in Crop Plants. Proceedings of the Secund Spring Conference,
March 22, lP.F. D'Mello and C.M. Duffus (Ed.). pp. 62-85. Edinburgh.
Skrede, A. and Krogdahl, A 1985. Heat affects nutritional characteristics of soybean meal
and excretion of proteinases in mink and chicks. Nutr. Rep. Int. 32:479.
Smith, A.K. and Circle, S.l. (Ed.) 1972. Soybeans: Chemistry and Technology. AVI
Publishing, Westport, CT.
Smith, AK. and Nash, A.M. 1961. Water absorption of soybeans. J. Am. Oil Chem.
Soc. 38: 120.
Smith, R.J. and Gallon, J.R 1993. Nitrogen Fixation. Ch. 6. In Plant Biochemistry and
Molecular Biology, P.J. Lea and R.C. Leegood (Ed.), pp. 129-153. John Wiley &
Sons, Chichester.
Smith, c., Megen, W.V., Twaalfhoven, L., and Hitchcock, C. 1980. The determination
of trypsin inhibitor levels in foodstuffs. J. Sci. Food Agric. 31:341-350.
Staswick, P.E. and Nielsen, N.C. 1983. Characterization of a soybean cultivar lacking
certain glycinin subunits. Arch. Biochem. Biophys. 223: 1-8.
Staswick, P.E., Hermodson, M.A, and Nielsen, N.C. 1981. Identification of the acidic
and basic subunit complexes of glycinin. J. BioI. Chern. 256:8752-8755.
Steiner, R.F. and Frattali, V. 1969. Purification and properties of soybean protein inhibitors
of proteolytic enzymes. J. Agric. Food Chem. 17:513-518.
Streggerda, F.R., Richards, E.A, and Rackis, J.J. 1966. Effects of various soybean products
on flatulence in the adult man. Proc. Soc. Expt. BioI. Med. 121:1235-1239.
Suda, I., Hajika, M., Nishiba, Y., Furuta, S., and Igita, K. 1995. Simple and rapid method
for the selective detection of individual lip oxygenase isozymes in soybean seeds. J.
Agric. Food Chem. 43:742-747.
Sudarmadji, S. and Markakis, P. 1977. The phytate and phytase of soybean tempeh. J.
Sci. Food Agric. 28:381-383.
Chemistry and Nutritional Value of Soybean Components / J J J
Sutardi and Buckle, KA 1985. Reduction in phytic acid levels in soybeans during tempeh
production, storage and frying. J. Food Sci. 50:260.
Sykes, G.E. and Gayler, KR. 1981. Detection and characterization of a new [3-conglycinin
from soybean seeds. Arch. Biochem. Biophys. 210:525.
Takasoye, M., Inoue, N. and Knuma, C 1990. Clinical investigation offeces improvements
by soybean oligosaccharides. Rinsho Toh Kenkyu (Clinics and Research), (in Japa-
nese) 67:304-310.
Thanh, V.H. and Shibasaki, K. 1976a. Major proteins of soybean seeds. A straightforward
fraction and their characterization. J. Agric. Food Chern. 24(6): 1117-1121.
Thanh, V.H. and Shibasaki, K. 1976b. Heterogeneity of [3-congJycinin. Biochim. Biophys.
Acta 439:326.
Thanh, V.H. and Shibasaki, K. 1979. Major proteins of soybean seeds. Reconstitution of
[3-conglycinin from its subunits. J. Agric. Food Chern. 26:695.
Thanh, V.H., Okubo, K., and Shihasaki, K. 1975a. Isolation and characterization of the
multiple 7S globulins of soybean proteins. Plant Physiol. 56: 19-22.
Thanh, V.H., Okubo, K, and Shibasaki, K. 1975b. The heterogeneity of the 7S soybean
protein by Sepharose gel chromatography and disc gel electrophoresis. Agric. BioI.
Chem. 39(7): 1501-1503.
Thompson, D.B. and Erdman, J.W. 1982. Phytic acid determination in soybeans. J. Food
Sci. 47:513-517.
Tombs, M.P. 1967. Protein bodies of soybeans. Plant PhysioJ. 42:797-813.
Tomomatsu, H. 1994. Health effects of oligosaccharides. Food Technol., Oct., pp. 61-65.
Torun, B., Vitery, F.E., and Young, V.R. 1981. Nutritional role of soya protein for humans.
J. Am. Oil Chem. Soc. 58:400-406.
Tsien, H.C., Jack, M.A, Schmidt, E.L., and Wold, F. 1983. Lectin in five soybean cultivars
previously considered to be lectin-negative. P1anta 158: 128.
Tsukamoto, C. Kawasakim. Y .• Iwasaki. T., and Okubo. K. 1991. A process for the
removal of glycosides during tofu production and an evaluation of the marketability
of the final product. In Japan part o{ Proceedings o{ the International Con{erence on
Soybean Processing and Utilization, K. Okubo (Ed.). pp. 47-51.
Tsukamoto. C. Shimada. S., Igita, K., Kudou, S., Kokubun, M .. Okubo. K., and Kitamura.
K. 1995. Factors affecting isofiavone content in soybean seeds: changes in isofiavones.
saponins, and composition of fatty acids at different temperatures during seed develop-
ment. J. Agric. Food Chern. 43:1184-1192.
Turner, R.H. and Liener, I.E. 1975. The effect of the selective removal of hemagglutinins
on the nutritive value of soybeans. J. Agric. Food Chern. 23:484.
Utsumi, S. and Kinsella, J.E. 1985. Forces involved in soy protein gelation: effect of
various reagents on the formation, hardness and solubility of heat-induced gels made
from 7S, l1S, and soy isolate. J. Food Sci. 50:1278-1282.
Utsumi, S., Gidamis, AB., Kanamori, J., Kang, J., and Kito, M. 1993. Effects of deletion
of disulfide bonds by protein engineering on the confonnation and functional properties
of soybean proglycinin. J. Agric. Food Chern. 41:687-691.
Vahouny, G. and Kritchevsky, D. (Ed.) 1986. Dietary Fibers Basic and Clinical Aspects.
Plenum Press, New York.
Vaintraub, LA. and Bulmaga, V.P. 1991. Effect of phytate on the in vitro activity of
digestive enzymes. J. Agric. Food Chern. 39:859.
Verdeal, K., Brown, R.R., Richardson, T., and Ryan, D.S. 1980. Affinity ofphytoestrogens
for estradiol binding proteins and effect of coumesterol on growth of 7, 12-dimethylbenz-
(a)anthracene-induced rat mammary tumors. J. Nat!. Cancer Inst. 64:285-290.
Vessby, B. 1994. Implications of long-chain fatty acid studies. INFORM 5(2):182-185.
Walker, A.R.P., Fox, F.W., and Irving, J.T. 1948. Human mineral metabolism. 1. Effect
of bread rich in phytate phosphorus on the metabolism of certain mineral salts with
special reference to calcium. J. Biochem. 42:452-462.
Walter, E.D. 1941. Genistin (an isoflavone glycoside) and its aglucone, genistein from
soybeans. J. Am. Chern. Soc. 63:3273-3276.
Wang, H.-J. and Murphy, P.A. 1994a. Isoflavone composition of American and Japanese
soybeans in Iowa: effects of variety, crop year and location. J. Agric. Food Chern.
42: 1674-1677.
Wang, H.-J. and Murphy, P.A. 1994b. Isoflavone content in commercial soybean foods.
J. Agric. Food Chern. 42:1666-1673.
Wang, H.I. and Murphy, P.A. 1996. Mass balance study of isoflavones during soybean
processing. J. Agric. Food Chern. 44:2377-2383.
Wang, H.L., Swain, E.W., Wallen, L.L., and Hasseltine, C. W. Free fatty acids identified
as antitryptic factor in soybeans fermented by Rhizopus oligosporus. J. Nutr. 105: 1351-
1355.
Wardlaw, G.M. and Snook, J.T. 1990. Effect of diets high in butter, com oil, or high-
oleic acid sunflower oil on serum lipids and apolipoproteins in man. Am. J. Clin.
Nutr. 51 :815-821.
Watanabe, M. 1993. Hypoallergenic rice as a physiologically functional food. Trends
Food Sci. Techno!. 4: 125-128.
Weaver, C.M., Nelson, N., and Elliott, J.G. 1984. Bioavailabi1ity of iron to rats from
processed soybean fractions determined by intrinsic and extrinsic labeling techniques.
J. Nutr. 114:1042-1048.
Wheeler, E.L. and Ferrel, R.E. 1971. A method for phytic acid determination in wheat
and wheat fraction. Cereal Chern. 48:312.
Whitaker, J.R. 1991. Lipoxygenase. In Oxidative Enzymes in Foods, D.S. Robinson and
N.A.M. Eskin (Ed.). pp. 175-215. Elsevier Applied Sci., London.
Wilkens, W.F., Mattick, L.R., and Hand, D.E. 1967. Effect of processing method on
oxidative off-flavor of soybean milk. Food Techno!. 21:86.
Wilson, L.A. 1996. Comparison of lip oxygenase-null and lipoxygenase-containing soy-
beans for foods. Ch. 12. In Lipoxygenase and Lipoxygenase Pathway Enzymes, G.J.
Piazza (Ed.), pp. 209-225. AOCS Press, Champaign, IL.
Wolf, W.J. 1969. Soybean protein nomenclature: a progress report. Cereal Sci. Today,
14:75, 76, 78, 129.
Wolf, W.J. 1993. Sulfuydryl content of glycinin: effect of reducing agents. J. Agric. Food
Chern. 41: 168- 176.
Wolf, W.J. and Briggs, D.R. 1959. Purification and characterization of the lIS component
of soybean proteins. Arch. Biochem. Biophys. 85:186-199.
Wolf, W.J. and Cowan, J.e. (Ed.). 1975. Soybeans as a Food Source. CRC Press, Cleve-
land,OH.
Wolf, W.J., Babcock, G .E., and Smith, A.K. 1961. Ultracentrifugal differences in soybean
protein composition. Nature 191: 1395-1396.
Wong, E. and Flux, D.S . 1962. Estrogenic activity of red clover isoflavones and some of
their degradation products. J. Endocrinol. 24:341-348.
Woodham, A.A. and Deans. P.S. 1975 . Amino acid requirements of growing chickens.
Bf. Poult: Sci. 16:269-287.
Wu, Y.V. and Sessa, D.J. 1994. Conformation of Bowman-Birk inhibitor. J. Agric. Food
Chern. 42:2136-2138.
Yabuuchi, S., Lister, R.M., Axelrod, B., Wilcox, J.R., and Nielsen, N.C. 1982. Enzyme-
linked immunosorbent assay for the determination of lipoxygenase isoenzymes in soy-
bean. Crop Sci. 22:333-337.
Yamauchi, F., Sato, M. , Sato, W., Kamata, Y. , and Shibasaki , K. 1981. Isolation and
identification of a new type of ~-conglycinin in soybean globulins. Agric. BioI.
Chern. 45:2863-2868.
Yamauchi, F., Yamagishi, T., and Iwabuchi, S. 1991 . Molecular understanding of soybean
protein. Food Rev. Int. 7:283-322.
Young, V.R. 1991. Soy protein in relation to human protein and amino acid nutrition. J.
Am. Diet. Asso. 91(7):828-835.
Young, V.R. and langhorbani, M. 1981. Soy proteins in human diets in relation to
bioavailability of iron and zinc: a brief review. Cereal Chern. 58:12.
Young, V.R. and Scrimshaw. N .S. 1978. Nutritional evaluation of protein and protein
requirements. In Protein Resources and Technology, M. Milner. N.S. Scrimshaw, and
D.I.e. Wang (Ed.). A VI Publishing, Westport, CT.
Young, V.R .. Bier. D.M .. and Pellett, P.L. 1989. A theoretical basis for increasing current
cstimates of the amino acid requirements in adult man, with experimental support. Am.
J. Clin. Nutr. 50:80-92.
Zarkadas, C.G., Yu, Z., Voldeng, H.D., and Minero-Amador, A. 1993. Assessment of the
protein quality of a new high-protein soybean cultivar by amino acid analysis. J. Agric.
Food Chem. 41:616-623.
Zemel, M. 1984. In I'itro evaluation of the effect of ortho- , tripoly- and hexametaphosphate
on zinc, iron and calcium bioavailability. 1. Food Sci. 49:1562-1565.
Zock, P.L. and Katan, M .B. 1992. Hydrogenation alternatives: effects of trans fatty acids
and stearic acid versus linoleic acid on serum lipids and lipoproteins in humans. Lipid
Res. 33:399.
Zuo, Y., Fahey, G.e., Merchen, N.R., and Bajjalieh, N.L. 1996. Digestion responses to
low oligosaccharide soybean meal by ileally-cannulated dogs. J. Animal Sci.
74(10):2441 - 2449.

Liu 1997

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Liu 1997

Uploaded by

Copyright:

Available Formats

2

Chemistry and Nutritional Value of

Table 2.1 . Proximate Composition of Soybeans and Their Structural Parts

Source: Data adapted from Wolf and Cowan (1975).

are also referred to as lipid bodies, spherosomes, oleosomes, or lipid-containing

Figure 2.1. Electron micrograph of soybean cotyledon cells. PB = protein bodies, S =

Table 2.2. Typical Composition of Crude and Refined Soybean Oil

Source: Data adapted from Pryde (1980).

I. Fatty Acid Composition

tion/processing, In terms of physical properties, the greater the chain length of

cis Isomer trans Isomer

Distribution of fatty acids at the glycerol molecule of a triglyceride is another

Glycerol (sn-glycerol) Triglyceride (triacylglycerol)

composition, interesterification generally increases crystallization tendencies

C. Nutritional Value of Soybean Oil

1. Essential Fattv Acids

Depressed growth, scaly dermatoses, increased permeability of skin, fatty liver,

2. Health Implications of Individual Fatty Acids

Table 2.4. Estimated Effects of Different

Source: From Vessby (1994).

3. Health Implications of trans Fatty Acids

Trans isomers are formed mainly during hydrogenation of polyunsaturated

The component present in the greatest amount in soybeans is protein, averaging

A. Protein Classification and Nomenclature

J.I =0.5 J.I = 0.1

Figure 2.5. Effect of ionic strength on the ultracentrifuge pattern of water-extractable

speaking, ~-conglycinin is only a component of the 7S fraction. It should also

B. Isolation of Major Storage Proteins

Localized in organelles known as protein bodies, storage proteins constitute the

Extraction by dilute salt solutions and separation of soybean proteins using

Defatted soy meal

Whole buffer extract

by adjusting to pH 4.8. The prepared 7S globulin is washed with pH 6.2 Tris

C. Characterization of Major Storage Proteins

1. f3-Conglycinin (7S Globulin)

All three major subunits are rich in aspartate/asparagine, glutamate/glutamine,

2. G/ycinin (llS Fraction)

6(A-SS-8) • 6(A) + 6(8)

(Murphy and Resurrection 1984). In contrast to ~-conglycinin, only a small

Source: Data adapted from Nielsen (l985a).

3. Differences between 75 and 115 Globulins

I. Heat instability of the constituent subunits of a protein, such as glycinin,

2. Hydrophobicity is an important factor in the emulsifying properties.

The significance of soybean trypsin inhibitors lies in their nutritional implica-

in vivo systems (Kennedy 1993). At least one inhibitor in soybeans, the BB

Although the nutritional effect of protein protease inhibitors on humans is the

Raw soybeans 171.0±3.4 100.0

Source: Data adapted from Liu and Markakis (l989b).

Because TI inactivation by heat treatments parallels the improvement in nutri-

c The E-Iast test

activity in soybean products is proposed (Liu and Markakis ] 989b). Compared

Lectins, also known as hemagglutinins, are proteins and possess a remarkable

linked to an enlargement of the pancreas, a lowering of blood insulin levels, an

acids, producing conjugated unsaturated fatty acid hydroperoxides. The enzyme

2. Oxidative Reaction and Off-flavor Formation

LOX catalyzes hydroperoxidation of linoleic acid and other polyunsaturated

LOX also catalyzes cooxidation of such pigments as carotenoids and chloro-

Hexanal 12-0xo-cis-9-dodecenoic acid

13-Hydroxy-12-oxo-cis-9-octadecenoic acid (a-ketol)

9-Hydroxy-12-oxo-trans-1O-octadecenoic acid (y-ketoll

9-Hydroperoxy-trans-l 0, 13-cis-octadecadienoic acid

cis-3-Nonenal 9-0xo-nonanoic acid

due to higher lipoxygenase activity in damaged beans. Therefore, most heat

Figure 2.15. Gas chromatogram of volatile substances extracted by simultaneous distilla-

o~ L -~ __- L_ _ _ _ _ _ ~ ______ ~~ _ _ _ _ _ _L - - - J