Title: Factors affecting the nutritional quality of soybean products fed to swine and chicks
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00102767/00001
 Material Information
Title: Factors affecting the nutritional quality of soybean products fed to swine and chicks
Physical Description: Book
Language: English
Creator: Campbell, Donnie Ray, 1949-
Copyright Date: 1986
 Record Information
Bibliographic ID: UF00102767
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - AEK6033
oclc - 15512335

Full Text







The author dedicates this dissertation in memory of his mother, the

late Verna Dean Campbell, and to his father, Joe Campbell, and his

entire family. Their support, patience and thoughtfulness are grate-

fully acknowledged.


There were many people who assisted in the completion of this

project. The encouragement, assistance, friendship and professional

guidance of Dr. C. E. White (chairman), Dr. G. E. Combs (cochairman),

Dr. R. D. Miles and Dr. R. M. Shireman are gratefully acknowledged.

The helpful assistance and advice (such as teaching the art of cane

pole fishing) of Dane Bernis were desperately needed to complete this

project. Through numerous ichthyological expeditions, Mike Harrison

is acknowledged for giving the author the pleasure of always sur-

passing Mike's aquatic skills. Likewise, the author is grateful for

Scot Williams, the bull Gator, who provided chum for the marine

species as needed. The author also extends appreciation to his fellow

graduate students and friends (Larry, Joe, Tom, Dewie, Kelly, Amy,

Britt and Bill), the swine unit crew (Tom, Denny, Shep, Kenny, Barry

and John), laboratory (A1, Pami and Nancy) and secretarial (Kathy and

Sharon) staff.

For teaching the author two important lessons the following

individuals are additionally acknowledged: M~ike Harrison, who

demonstrated so many times that there is one step beyond knowing

something like the back of your hand, and Dr. Combs, who taught the

author the way to increase the number of fish caught per cast; cast

once and troll all day or until a fish is caught.

The author extends deepest appreciation to Wendy Jo who had a

large part in the completion of this project. Her assistance,

patience, understanding and friendship will always be remembered.


ACKNOWLEDGEMENTS ** ** ** *. . . ... iii

ABSTRACT ** ** ** ** ** ** *. .. ..vi


I INTRODUCTION * * * * * * * . . . . . 1

II LITERATURE REVIEW ** ** ** *. .. . . 5

Trypsin Inhibitors . .. . *** *. . .. . 5
Protein Digestibility . .. ... . .. .. .. 17
Fat Absorption and Energy Digestibility .. ... .. 20
Antibiotic Supplem~entation .. . .... . .. 21
Mineral and Vitamin Supplementation . . . 25
Other Anti-Nutritional Factors ... .. .. ... 27
Soybean Processing *. . .. . 28
Effects of Overheating Soybeans ... .. .. .. .. 30
Improvement of the Nutritive Value of Soybeans by Heat
Processing ** ** ** ** ** ** *. . . 32
Other Methods of Processing Soybeans ** *. .. . 36
Swine Feeding Trials ** ** ** ** ***. .. . 39
Poultry Feeding Trials ** *. . .. .. .. .. 44
Composition of Commercial Soybean Meal ... .. .. 45
Variation Among Soybean Varieties .. .. . ... .. 53
Effect of Storage on Soybeans and Soybean Products . .. 57
Effect of Maturity on Composition of Soybeans .. .. 58
Influence of Growing Conditions on Soybean Composition . 59


Introduction * ** ** ** *. . . . 61
Materials and Methods ** ** ** ** ***. . . 62
Results and Discussion * **** *. . . . 64


Introduction * ** ** ***. . .. 72
Materials and Methods ********** .... .74
Trial 1 * * * * * * . .. 74
Trial 2 * * * * * * * * . . . 75

Trial 3 .. .. . 7
Results and Discussion .. .. .. .. .. .. ... 79
Trial 1 .. .. .. .. . .. .. 7
Trial 2 .. ... .. 8
Trial 3 .. .. .. .. 8
Grower period .. .. ... .. .. .. .. 87
Finisher period .. .. .. .. .. . 90
Overall .. .. .. .. ... .. .. .. 90


Introduction ... .. .. ... .. .. .. . .. 94
Materials and Methods ... .. .. .. .. ... 96
Trial 1: Chick Trial ... .. .. .. .. .. 96
Trial 2: Starter Period . .. . .. .. ... 99
Trial 3: Grower Period .. .. .. .. .. .. .. 99
Trial 4: Finisher Period. ... . .. .. .. .. 99
Trials 5, 6 and 7: Digestibility Trials .. .. 100
Results and Discussion . .. ... ... . .. 101
Trial 1: Chick Trial .. .. .. ... .. .. 101
Trials 2-4: Swine Trials .. .. ... .. .. 104
Starter period .. .. .. .. ... . .. 104
Grower period . ... .. ... .. .. 107
Finisher period .. .. .. .. .. .. .. 107
Trials 5, 6 and 7: Digestibility Trials .. .. 109
Starter period .. .. .. .. .. ... 109
Grower period .. . ... .. .. ... 109
Finisher period .. .. .. .. .. ... 113

VI CONCLUSIONS ... .. .. .. ... .. .. . .. 116




LITERATURE CITED .. .. ... .. ... .. .. .. .. 162

BIOGRAPHICAL SKETCH .. .. ... .. .. ... .. .. 184

Abstract of Dissertation Presented to the Graduate
School of the University of Florida in Partial
Fulfillment of the Requirements for the
Degree of Doctor of Philosophy



Donnie Ray Campbell

December, 1986

Chairman: Dr. C. E. White
Cochai rman: Dr. G. E. Combs
Major Department: Animal Science

Eleven experiments were conducted to evaluate several factors

that could influence the nutritional quality of soybean products fed

to swine and chicks. The effects of increasing moisture content of

soybeans prior to roasting and varying roasting temperature on the

nutritional value of soybeans fed to starting swine (5 kg) were

determined by feeding diets containing soybean meal or "full-fat"

soybeans (unheated or roasted at 110 or 125 C with 0 or 10% water

added prior to roasting) (Exp. 1). Growth of pigs fed the diet

containing soybean meal was superior (P<.05) compared to the other

treatments groups (whole soybeans). Increasing the roasting

temperature from 110 to 125 C or adding 10% moisture prior to roasting

soybeans augmented (P<.05) feed intake and weight gain of weanling


Comparison of the varietal differences of soybeans indicated wide

variation in the fat and protein contents and trypsin inhibitor (TI)

and urease activities (UA; Exp. 2 and 3). Fat concentration was

negatively correlated (P<.05) with the protein concentration and

positively correlated (P<.05) with the UA of the unheated soybeans.

The effect of these varietal variations on the performance of growing-

finishing swine (40 kg initially) was evaluated (Exp. 4). During the

grower period, pigs fed the raw or roasted soybean (Bragg or Davis

varieties) diets gained more slowly (P<.05) than those fed soybean

meal. Growth of pigs was increased (P<.05) by roasting either variety

of soybeans at 110 C. During the finisher periods utilization of

roasted soybeans of either variety and raw Davis soybeans permitted

pig growth and feed efficiency equal (P>.05) to pigs fed soybean

meal. In addition, six trials involving crossbred pigs and one 21-day

trial using day-old chicks were conducted to evaluate the utilization

of defatted soybean flakes which had been subjected to varying heating

times. Total gains, feed intake and feed efficiency of chicks were

not influenced (P>.05) by feeding defatted soybean flakes cooked for

16, 18, 20 or 22 minutes whereas growth rate and feed intake of

weanling pigs and average daily gains of growing pigs fed defatted

soybean flakes cooked for 16 minutes or less were adversely affected

(P<.05) when compared to pigs fed defatted soybean flakes cooked for

22 minutes. During the grower period, feed intake and efficiency

were reduced (P<.05) compared to pigs fed defatted soybean flakes

cooked for 22 minutes. Finishing pigs fed defatted soybean flakes

cooked for 6, 12, 16 or 22 minutes grew at a similar (P>.05) rate and

required equal (P>.05) amounts of feed per body weight gain. Amino

acid digestibilities at the end of the small intestine were also

determined at each age group with pigs fed the defatted soy diets.



The soybeans Glycine max. (L.) Merr., is a member of the family

Leguminosae and subfamily Papilionoideae and has been consumed in the

Orient since ancient times. The earliest Chinese records which

mention soybeans date back to about the time of the building of the

Egyptian pyramids. The Buddhist religions because of the exclusion of

meat from the diet of its people, was a major influence in the

development of soybeans for food in the Oriental countries. However

the use of soybeans in the United States covers a comparatively short

period with the first commercial oilmill processing plant established

in 1922. The meal was regarded as a by-product initially and had

little value when compared to the high quality oil.

Soybean meal (SBM) is currently the most widely used source of

supplemental protein in livestock diets. This extensive usage may be

attributed to the excellent amino acid profiler dependable supply and

competitive price. Approximately 80% of the SBM produced in the

United States is used in swine and poultry diets (Smithr 1977). When

formulating present day swine and poultry diets using computer

least-cost feed programs, excess nutrients may be kept to a minimum.

This practice has increased the awareness of including only high

quality ingredients, or at least having an analysis with the true

nutritional quality of the feedstuff. With SBM it is critical to

monitor the nutritional value as well as ensure that the heat labile

anti-nutritional factors found in raw soybeans have been denatured by

an optimum level of heat processing.

The level of heat processing required to denature the heat labile

anti-nutritional factors found in soybeans before they are marketed

requires energy which appears destined to become more costly in the

future. Dada (1983) estimated that conventional fuel for boilers

accounts for more than half of the energy used to manufacture SBM.

The desolventizer-toasting process and final drying step accounts for

50 to 70% of the total steam consumption. It is the desolventizer-

toasting process that is the most sensitive step in controlling the

nutritional quality of SBM (Mustakas et al., 1981). The amount of

heat processing of soybean products required for optimal animal

performance and still ensure economical situations for both the

commercial soybean meal processor and swine producers could vary in

the future.

Two different SBM's (44.0 and 48.5% protein) are currently being

marketed depending on whether the soybean hulls are added back to the

meal after being heat processed. However, both commonly receive the

same heat processing and are currently being marketed for all animal

species regardless of age. There is some indication that older

maturer) animals might efficiently utilize SBM having less heat

processing (Combs and Wallace, 1969). If older animals are more

efficient than younger animals in their ability to utilize

underprocessed SBMr then it may be miore economical to also market a

SBM having less heat treatment. This practice could result in a

reduction of energy usage and lower the price of the SBM fed to older


During periods of low demand for soybean oil or low price for

whole soybeans, it is economical to include whole "full-fat" soybeans

in nonruminant diets. Whole soybeans, if properly processed, contain

not only high quality protein but are also a rich source of energy due

to their high oil content. Therefore whole soybeans have the

potential of suppling major quantities of both energy and protein in

nonruminant diets. Two methods of heat processing whole soybeans

(roasting and extrusion) currently available on the farm have

increased the feasibility of including whole soybeans in diets for

nonruminant animals. Adding fat to increase dietary caloric density

has been used during periods of elevated ambient temperature to

increase daily energy intake. In addition, recent studies have

evaluated the effect of adding fat to diets fed to sows during late

gestation and lactation in an attempt to increase piglet

survivability. However liquid fat is difficult to manage without

proper storage, mixing and handling facilities. These difficulties

are less of a problem when adding ground whole full-fat soybeans in

the diet to increase the caloric density. Also, soybean oil contains

high levels of unsaturated fatty acids, and when fed to chicks (Porter

and Britton, 1974) and swine (Seerley et al., 1974; Wahlstrom et al.,

1971) their carcasses contain an increased quantity of unsaturated

fatty acids while still maintaining favorable organoleptic qualities.

This type of meat product may be more attractive to the consumer today

due to the current trend to reduce saturated fat in the human diet.

Numerous new varieties of soybeans with improved agronomic

advantages over existing varieties are released annually. In soybean

breeding programs soybeans can be selected for increased protein or

increased fat composition but not increases in both traits

concurrently (Hartwigs 1979). Most of the anti-nutritional factors,

such as trypsin inhibitors in raw soybeans are proteins which also

contain a large percentage of the sulphur amino acids. Consequently

varieties containing an increased quantity of trypsin inhibitors may

contain higher contents of protein and sulphur amino acids when

compared to varieties with a lower concentration of trypsin

inhibitors. If the soybeans are subjected to heat processing the

higher content of trypsin inhibitors may not be detrimental, and when

denatured, would provide increased amounts of available amino acids.

The objectives of this research were:

1. Assess the effect of increasing the moisture content of

raw soybeans prior to roasting at different temr-peratures on their

nutritional quality when fed to weanling pigs.

2. Quantitate varietal differences in nutrient composition

and assess their effect on performance of growing-finishing pigs.

3. Determine if weanling pigs and day-old chicks could

attain maximniumn performance when fed raw defatted soybean flakes

subjected to varying levels of heat processing.

4. Determine the influence on pig performance of varying

the level of heat processing of defatted soybean flakes as the pig



Osborne and Mendel (1917) were the first to recognize that

unheated soybeans were inferior to heated soybeans. They reported

that heating the soybean protein improved the flavor and resulted in

improved growth rate when fed to rats. Subsequent research by

Shrewsbury et al. (1932) substantiated that heating improved the

nutritional value of soybeans when fed to pigs and rats and postulated

the presence of toxic factors in raw soybeans that could be removed or

destroyed by heating. Since then, many workers have reported that

cooking or roasting improves the nutritional value of soybeans.

Several factors have been shown to contribute to the unsatisfactory

performance of animals fed raw soybeans. Among these were trypsin

inhibitors (Bowman, 1944), lectins (phytobemagglutinins) (Liener and

Pallansch, 1952), saponins (Potter and Kummerow, 1954), goitrogenic

substance (Patton et al., 1939), decreased fat absorption (Nesheim et

al., 1962), and decreased amino acid availability (Borcherss 1961).

Trypsin Inhibitors

Twenty-seven years after Osborne and Mendel's initial work,

soybeans were found to contain proteolytic inhibitors (Ham and

Sandstedt, 1944; Bowman, 1944). Subsequently, a compound which

inhibited trypsin was isolated, crystallized, and characterized from

raw soybeans (Kunitz, 1945). After further investigations another

acetone insoluble trypsin Inhibitor was isolated which Is now referred

to as the Bowman-Birk inhibitor (Birk et al., 1963). Obara and

Wajanabe (1971) confirmed that various trypsin inhibitor fractions

existed and determined that they have different susceptibilities to

heat inactivation. The different trypsin inhibitors are comprised of

a complex mixture of proteins and have been classified broadly into

two main groups (Liener and Tomlinson, 1981). One group, of which the

Kunitz soybean trypsin inhibitor is the best known examples has a

molecular weight in the range of 20,000 to 25s000 daltons,

specifically inhibits trypsin, and is relatively heat labile. The

other groups Bowman-Birk inhibitors consists of a family of proteins

having molecular weights of approximately 8,000 daltons. Because of

their high cystine content, the proteins found within this family are

generally considered to be relatively heat stable. The Bowman-Birk

inhibitor is unique in that it Inhibits chymotrypsin as well as

trypsin at two independent binding sites. Further research has

indicated that the Bowman-Birk group consists of ten isoinhibitors

(Tan-Wilson et al., 1985) and the Kunitz group has three isoinhibitors

(Orf and Hymowitzs 1979). Because of these inhibitor specificities,

raw soybeans contain twice the trypsin inhibitor activity compared to

the activity for inhibition of chymotrypsin (Baintner, 1981). During

heat processing trypsin inhibitor capacity is partially Inactivated

prior to initial inactivation of the chymotrypsin inhibitor activity.

Studies with the Bowman-Birk trypsin inhibitor indicated that most of

this inhibitor is degraded during its passage through the stomach and

small intestine of chicks and that there Is negligible absorption of

the native inhibitor with most of the degradation products being

excreted in the feces (Madar et al., 1979). In addition to the

protein trypsin inhibitors, Hafez and Mohamed (1983) reported that

soybeans also contain nonprotein trypsin inhibitors.

Proteolytic inhibitors are not unique to soybeans but are

somewhat ubiquitous in nature and have distinct roles. For example,

In mammals trypsin inhibitors are a component of colostrum and help

prevent proteolysis of antibodies. The pancreas also secretes trypsin

inhibitor to prevent activation of the proteolytic enzymes of the

pancreatic juices until they are secreted into the small Intestine.

Likewise researchers postulate the functions of the proteinase

Inhibitors in soybeans (seeds and plant) are to (1) maintain dormancy

by preventing autolysisr (2) regulate protein synthesis and metabolism

and (3) prevent attack by predatory insects (Smith and Circler 1972).

Initial studies in which partially purified preparations of

soybean trypsin inhibitor were fed to rats and chicks resulted in no

significant effect on their growth rate (Borchers et al.r 1948).

However, subsequent research has well documented the growth depression

obtained by Including raw soybeans In place of heated soybeans in the

diet of rats (Liener et al., 1949; Borchersr 1961), chicks (Alumot and

Nitsanr 1961; Nesheim et al., 1962), and pigs (Pekas, 1966; Hooks et

al.r 1967a). Trypsin inhibitor concentrate or raw soybean meal also

causes pancreatic hypertrophy concurrent with the significantly slower

growth in chicks (Nesheim et al., 1962; Salmon et al., 1967) and rats

(Brambila et al., 1961; Borchers, 1964) but not in pigs (Hooks et al.,

1965; Pekasr 1966). Liener et al. (1949) fed diets containing raw

soybeans, heated soybeans, and heated soybeans + 1.8% trypsin

inhibitor to rats and reported a depressed protein efficiency ratio

value for the heated diet containing trypsin inhibitor compared to the

heated soybean diet but a higher protein efficiency ratio value than

obtained for the diet containing raw soybeans. These data provided

the Initial indication that trypsin inhibitors were not the exclusive

anti-nutritional constituent of soybeans. The Kunitz inhibitor can

account for all the pancreatic hypertrophy effects but for only about

30 to 60% of the growth-inhibiting properties of raw soybean meal fed

to rats (Rackis, 1965). More recent research by Kakade et al. (1973)

confirmed the 40% growth depression but also reported only 40% of the

pancreatic enlargement in rats produced by the ingestion of raw

soybeans was accounted for by the trypsin inhibitors.

Contrasting research has been presented to account for the

pancreatic enlargement. Kakade et al. (1967) observed hyperplasia

(increase in cell number) of the pancreatic acinar cells while Konijn

and Guggenheim (1967) reported hypertrophy or increased cell size. No

histopathological damage was observed in rat pancreas hypertrophied

for six months, and pancreatic hypertrophy was reversible in rats

(Booth et al., 1964) and in chicks (Salmon and McGinnis, 1969). The

ability to compensate for the proteolytic activity of raw soybeans was

greater with increasing age of chicks (Nitsan and Alumotr 1964).

The efficacy of supplementing rat, chick and pig diets with

trypsin to overcome the adverse effect of feeding raw soybeans has

been studied. Inclusion of 5% dietary trypsin powder in a raw soybean

diet improved rat growth (Borchers and Ackerson, 1951). Autoclaving

the trypsin powder destroyed Its proteolytic activity but did not

reduce growth. Brambila et al. (1961) added a crystalline and a crude

trypsin preparation to chick diets containing raw soybeans but still

obtained depressed growth. Pancreatic hypertrophy was reduced with

the crude trypsin preparation but crystalline trypsin did not prevent

the pancreatic enlargement. Similarly, inclusion of .5% trypsin in

diets containing raw soybeans fed to pigs at 3, 9 or 16 weeks of age

did not Improve the rate and efficiency of gain or dry matter, protein

and ether extract digestibilities (Combs and Wallaces 1969). Plasma

glucose and urea nitrogen also did not differ when pigs were fed .5%

trypsin in the diet.

Feeding raw soybean meal to chicks resulted in hypertrophic

pancreas with higher trypsin and lower amylase specific activities in

this organ (Pubols et al.r 1964). Supplemental dietary methionine

increased the ratio of amylolytic to proteolytic enzymes (Nitsan and

Gertler, 1972). Total secretion of trypsin nearly doubled in chicks

fed raw soybean meals whereas amylase, lipase and chymotrypsin

activities were not significantly different from that of chicks fed

autoclaved soybean meal (Dal Borgo et al., 1967). Further research

data collected by Gertler and Nitsan (1970) indicated increased levels

of trypsin, chymotrypsinr and pancreatopeptidase but decreased levels

of amylase were secreted from the pancreas when raw soybeans were

substituted for heated soybeans in a chick diet. However the

addition of a trypsin inhibitor In the heated soybean diet increased

the quantities of all four enzymes. Soybean trypsin inhibitor

activity was correlated with pancreatic trypsin activity in the chick

but was not correlated with the activities of amylase, chymotrypsin or

pancreatic trypsin inhibitor (Pubols et al., 1985). In addition,

feeding raw soybean meal was reported to Inhibit the synthesis of

pancreatic lipase while stimulating excessive secretion of intestinal

lipase in chickens (Lepkovsky and Furuta, 1970).

In other studies, feeding raw soybean meal to rats also

stimulated bypersecretion of pancreatic enzymes (Borchers, 1964).

Konijn et al. (1970) and Temler et al. (1984) reported increased

trypsin and chymotrypsin activities and an enlarged pancreas In rats

by feeding diets containing .72 and 1.087o trypsin inhibitor. Amylaser

elastase and lipase activities, feed intake and body weight were not

Influenced. The Increase in pancreatic trypsin activity due to

feeding soybean trypsin inhibitor was confirmed by Fushiki et al.

(1984) but the researchers also noted an Increase in pancreatic lipase

activity. The trypsin activity was highly correlated with the total

protein output In the bile-pancreatic juice. Feeding of raw soybeans

or crystalline trypsin inhibitor immediately Increased amylase and

lipase activities in the intestine of rats and after three hours the

activities were increased three to fourfold (Lyman and Lepkovskyr

1957). The initial intestinal trypsin activity was lowr apparently

inactivated by the trypsin Inhibitor, but increased after six hours.

These researchers noted that pepsin secretion was unaffected by

feeding raw soybeans. The findings by Borchers (1964) indicated that

kidney transaminase activity was also reduced when rats were fed raw

soybean meal but other tissue enzymes showed no change.

The physiological changes In pigs due to feeding raw soybeans do

not concur with the data obtained with chicks and rats. Pancreatic

juice secretion was reduced in pigs fed raw soybean meal compared to

pigs fed heated soybean meal diets (Pekas, 1966). However, the

secretary response obtained by feeding raw soybean meal is related to

pig maturity (Hooks et al., 1965). Weanling pigs fed raw soybean meal

had reduced pancreas weight, nitrogen content of the pancreas and

lipase activity of the intestinal fluids and pancreas compared to

weanling pigs fed heated soybean meal. Protease activity of the

pancreas and intestinal fluids did not differ. When growing pigs were

fed raw soybean meal, their pancreatic weights and protease and lipase

activities of the pancreas and Intestinal fluids were similar to pigs

fed heated soybean meal. In contrast to weanling pigs, nitrogen

content of the pancreas was increased when raw soybean meal was fed to

growing pigs. Cell structure and zymogen content of pancreatic tissue

did not differ between pigs fed heat processed or unprocessed soybean

meal nor between the different ages of pigs. In another study both

raw soybean and Kunitz soybean trypsin inhibitor decreased pancreatic

trypsin and chymotrypsin activities of growing pigs (Yen et al.,

1977). However inhibition of intestinal trypsin and chymotrypsin

activities was greater in the pigs fed the raw soybean diet. Data

from a more recent study using growing pigs fitted with a pancreatic

cannula indicated that feeding an unheated commercial soybean product

increased the protein secretion and total activities of trypsin and

chymotrypsin of the pancreas compared to pigs fed a heated soybean

product (0zimek and Sauer, 1985). In contrast, Corring et al. (1985)

fed a diet containing raw soybeans to growing pigs and observed that

the total protein output from the pancreas was not affected. However,

the volume of pancreatic juice secreted was increased as well

as the plasma levels of secretin when a diet containing raw soybeans

was fed. Zebrowska et al. (1985) confirmed the work of Corring and

also reported no differences in total and specific activities of

trypsin, chymotrypsinr carboxypeptidases A and B and amylase in the

pancreatic juice of growing pigs fed raw or adequately heat processed

soybean meals.

Seventy percent of the protein digestion of heat treated soybean

meal in the chick was found to occur In the duodenum and 20% occurred

in the upper jejunum (Bieloral et al.r 1973). When raw soybean meal

was fed, 70% of the protein digestion took place in the duodenum and

no further digestion occurred in the remaining segments of the

digestive tract. The increased secretion of enzymes and bile

constituents into the duodenum in response to feeding raw soybeans was

inactivated in the jejunum by the anti-nutritional factors in the

soybeans and further digestion did not occur. Bieloral et al. (1973)

suggested that the growth depression from feeding raw soybean meal

could result from the 20% reduced protein absorption and the small

amount of energy needed for the increased production and secretion of

enzymes into the duodenum. This helps explain why a lower

concentration of free amino acids in the intestinal contents with

chicks fed raw soybeans as compared with those fed heated soybeans,

was observed In an earlier study by the same group (Bielorai et al.,


Numerous factors have been proposed as being responsible for the

reduced animal performance when raw soybeans are fed. Alumot and

Nitsan (1961) concluded that the growth retardation was attributed to

a combination of low availability of dietary protein and an increased

protein requirement resulting from stimulation of the pancreatic

activity to increase enzyme production to overcome the trypsin

inhibitors. The excessive amount of protein secreted in the pancreas

of rats was ultimately lost In the feces (Haines and Lymanr 1961). In

a later study, Muelenaere (1964) reported that feeding rats a diet

containing 5% trypsin inhibitor caused a slower rate of stomach

emptying and Interfered with amino acid absorption through the

Intestinal wall. Lanchantin et al. (1969) found that the Kunitz

trypsin Inhibitor reacted with trypsin almost instantaneously to form

a complex with an extremely low dissociation constants thus decreasing

the quantity of protein being hydrolyzed. Lepkovsky et al. (1971)r

using rats and chicks, suggested that the trypsin Inhibitor combines

with protein in the intestine to form complexes which escape digestion

and are subsequently lost in the feces. Green et al. (1977)

substantiated that native undenatured soybean protein is capable of

binding trypsin by forming an enzyme-substrate complex which can

remove feedback inhibition of pancreatic secretion by trypsin.

Additional research by Lepkovsky et al. (1970) has indicated that

after feeding raw soybeans to chicks, the quantity of enterokinase in

the intestinal juice could not be measured which could limit the

quantity of trypsinogen being converted to trypsin. Singh and

Krikorian (1982) reported In a more recent study that low levels of

phytic acid found in raw soybeans in vitro can also inhibit trypsin


Lyman et al. (1974) and Schneeman et al. (1977) attributed the

stimulatory effect of raw soybean flour on the secretary activity of

the pancreas to a negative feed-back regulation. The pancreas was

induced to increase the output of enzymes when the levels of trypsin

and chymotrypsin in the intestine fall below certain threshold values

due to complexes formed with trypsin inhibitors or dietary protein.

It was believed that the mediating agent between trypsin and the

pancreas is the hormone cholecystokinin. The feeding of diets

containing trypsin Inhibitor at 6% of the protein level to rats

stimulated an increase in the production of pancreatic proteases

compared to the feeding of diets containing either proteins or

peptides (Temler et al., 1984). Although trypsin inhibitors are

potent stimulators of the secretion of cholecystokininr they can also

stimulate secretion of other unidentified gastrointestinal factors.

Struthers et al. (1983) likewise previously noted that the increased

secretion of cholecystokinin was not the sole mediator of effects

produced by feeding raw soybean flour.

Richardson (1981) and Liener (1981) reviewed previous research

and summarized the deleterious effects of proteinase inhibitors in raw

soybeans and developed the following scheme. The proteinase

inhibitors and undenatured soybean protein are only partially

inactivated by pepsin and entering the small intestines form a complex

with trypsin and chymotrypsin. The resulting lower concentrations of

trypsin and chymotrypsin reduce proteolytic hydrolysis and stimulate

secretion of cholecystokinin. The undigested protein is lost through

the feces (exogenous loss). Increased secretion of pancreatic enzymes

caused the protein from body tissue to be broken down and used in

increased synthesis of proteinases. Methionine in particular Is used

by its conversion to homocysteine then to cystathionine and cysteine.

This mechanism resulted in increased quantities of proteinases in the

intestine. The sulphur containing amino acids are then degraded by

the micoflora in the colon and lost in the feces endogenouss loss).

The control mechanisms for synthesis of pancreatic nucleic acids

and enzyme proteins are dissimilar within different mammalian species

(Struthers et al., 1983). Feeding raw soybean products containing

large quantities of trypsin inhibitors produced enlargement of the

pancreas in rats mice, and chicks; but not in dogs, calves, pigs or

monkeys. Pancreatic hypertrophy and increased secretion were almost

Immediate in rats (Lyman and Lepkovsky, 1957); whereas in chicks,

hypertrophy and pancreatic juice secretions were delayed for three to

eight days following the feeding of soybean trypsin inhibitor (Nitsan

and Alumots 1964; Kakade et al., 1967). The immediate response in the

pancreas of rats was a decrease in protein synthesis followed by

increased selective enzyme synthesis, stimulated by both the Kunitz

and Bowman-Birk inhibitors which resulted ultimately in hypertrophy and

loss of endogenous protein (Konijn et al.s 1970). Animals whose

pancreas weights were greater than .3% of their relative body weight

exhibited pancreatic hypertrophy when fed raw soybeans whereas animals

whose pancreas weights were less then .3% of their relative body

weight did not (Liener, 1977). Although pancreatic hypertrophy does

not occur in all animal species, soybean trypsin inhibitors have been

shown in vitro to inhibit 90 to 100% of trypsin obtained from the rat,

monkey, bovine, humans porcine and mink (Struthers et al., 1983).

Additional investigations have centered on quantifying various

species response to feeding raw and heated soybean products. For

example, Struthers et al. (1983) fed rats, pigs and monkeys raw and

heated soybean flour containing 115 to 130 and 4 to 8 units trypsin

inhibitor per mg protein, respectively. Growth and nitrogen

digestibility were depressed 60 and 5%, 84 and 45%, and 0 and 9% for

rats, pigs, and monkeys, respectively due to feeding the raw soybean

flour. Fecal trypsin concentration was Increased 300 to 400% in rats

but decreased 50% in pigs and monkeys when fed raw soybean flour.

Rats and pigs consumed significantly less of the raw soybean flour

diet compared to the diet containing the heated soybean flour.

However monkeys consumed equal quantities of the two diets. Other

differences due to feeding raw soy flour included changes In RNA per

mg pancreas and pancreatic protein concentrations with changes of +40

and +47% rand +20 and -7% for rats and pigs respectively. These

measured criteria were not altered in the monkeys. In other studies

Hasdai and Liener (1983) reported depressed growth, feed intake, feed

efficiency, and protein digestion when raw soybean flour was Included

in the diet of hamsters. They also obtained increased pancreas and

kidney size and elevated trypsin, chymotrypsin, amylase and lipase

activities in the pancreas. Feeding raw soybeans to the mink resulted

in a 20-fold increase in fecal trypsin activity when compared to that

found in chick excreta (Skrede and Krugdehls 1985). In addition,

chick excreta contained a larger quantity of proteinase inhibitors

compared to mink excreta. Furthermore Gorrill and Thomas (1967)

observed poor growth in young calves fed raw soybean meal, and reduced

trypsin and chymotrypsin secretions but no pancreatic hypertrophy.

Protein Digestibility

As previously inferred, a change in pancreatic proteolytic enzyme

production and consumption of proteolytic inhibitors can combine to

alter protein hydrolysis and thereby reduce the availability of and

increase the requirement for the essential amino acids. Therefore,

adding amino acids to diets containing raw soybeans to improve

performance has been investigated. One such study by Hill et al.

(1953) indicated that the addition of an amino acid mixture failed to

prevent the growth depressing effect of raw soybeans when fed to

chicks. However, in subsequent research, normal growth was obtained

by supplementing a diet containing raw soybeans with another mixture

of amino acids (Fisher and Johnson, 1958). They suggested that

earlier data which did not show improved growth was a result of

inadequate amino acid balance. Increasing the dietary protein content

instead of adding only amino acids to the diet improved the growth

rate of rats (Fisher and Shapiror 1963) but not of pigs (Combs et al.,

1967) fed diets containing unheated soybean meal. Saxena et al.

(1962a) obtained increased growth and feed efficiency when adding a

mixture of amino acids, varying from four to 14 amino acids to raw

soybeans, but the performance was not equal to a diet containing

heated soybean meal with the same amino acids. These researchers also

reported that chicks fed raw soybeans consumed five times the amount

of oxygen and had much lower liver and muscle glycogen content than

chicks fed autoclaved soybean meal. However when these chicks were

fed the raw soybean diets containing amino acid supplements, their

oxygen consumption returned to normal. More specifically

supplementation of diets for weanling rats containing raw soybeans

with methioniner threonine and valine resulted in performance that was

similar to rats fed heated soybeans (Borchers, 1961). Khayambashi and

Lyman (1966) confirmed Borchers' work but also reported that the

increased pancreatic and intestinal protease activities and intestinal

insoluble nitrogen observed in rats fed a diet containing soybean

trypsin inhibitor were not reduced when methioniner threonine and

valine were added. Hooks et al. (1967a) reported that methionine

supplementation in diets fed to rats and chicks improved their

performance but the addition of either threonine or valine to diets

containing raw soybeans did not.

Dietary supplementation with methionine to overcome the growth

depressing effect when raw soybeans were fed has received a great deal

of attention but results were inconclusive. Nickelson et al. (1960)

fed weanling pigs and presented data indicating improved growth with

as low as .1% supplemental methionine. The ameliorated effects of

feeding raw soybeans to rats and chicks by supplementing diets with

methionine was confirmed by Hooks et al. (1967a). However these

researchers did not obtain increased pig performance by including

methionine in raw soybean diets. Similarly, more recent studies have

reported that methionine supplementation of diets containing raw

soybeans did not Improve the performance of finishing pigs (Jensen et

al., 1970) nor pigs at three, nine, or 15 weeks of age (Combs and

Wallace, 1969). Although Combs and Wallace (1969) noted a increase in

dry matter and protein digestibilities when methionine was added to

the raw soybean diet fed to pigs at three weeks of age, these criteria

were not different for pigs at nine or 16 weeks of age,

Almquist et al. (1942) fed chicks raw soybean diets and obtained

increased growth with methionine supplementation but not with

additions of choline or cystine. A study by Nitsan and Gertler (1972)

confirmed that the addition of .3 or .6% methionine to diets

containing raw soybeans would increase chick performance. The increase

in chick growth obtained by adding methionine to a diet containing

inadequately heat processed soybean flakes was confirmed by Miles and

Featherston (1976). These researchers also reported that lysine

supplementation alleviated the adverse effects on growth when chicks

were fed soybean flakes which were overheated.

Likewise methionine supplementation of rat diets containing raw

soybean meal increased growth and feed efficiency equal to that of

rats fed diets containing heated soybean meal but performance was

depressed when compared to rats fed a heated soybean meal diet

containing supplemental methionine (Hensley et al., 1953). In

addition, the inclusion of aureomycin in the diets containing raw and

heated soybean meal supplemented with methionine permitted an

additional increase in growth and feed efficiency. In another study

the addition of .6% methionine improved the protein efficiency ratio

of a raw soybean meal diet similar to the improvement obtained by

heating the raw soybean meal (Liener, 1949). Nitrogen retention as

well as performance of rats were improved when .3% methionine was

included in a diet containing raw soybean meal (Yen et al., 1971).

These findings are in contrast to an earlier study (Carroll et al.,

1953) which indicated that methionine supplementation in rat diets

containing either raw or heated soybean meal did not affect the amount


and site of nitrogen absorption. Subsequent research has indicated

that the Inhibitors in raw soybeans do not affect the availability of

supplemental methionine when included in a rat diet but these

inhibitors only depressed the availability of the intact methionine

(Rao et al., 1971) However, Liener et al. (1949) had previously

reported that a preparation of trypsin inhibitor was capable of

inhibiting rat growth even when Incorporated in diets containing

either predigested protein or free amino acids.

The response from adding methionine to raw soybean diets fed to

rats can be influenced by ambient temperature (Yen et al., 1971). The

largest increase in growth between rats fed either a heated soybean

diet or a diet containing raw soybeans supplemented with methionine

when compared to rats fed a raw soybean diet was obtained when the

ambient temperature was 23 C compared to 7 C. The percentage of raw

soybean meal in a rat diet can also alter the response of rats from

adding methionine (Barnes et al., 1962). Addition of .3% methionine

to diets containing 50% unheated soybean flakes increased growth of

rats but no growth stimulation was obtained by the addition of

methionine to a diet containing 70% unheated soybean flakes.

Fat Absorption and Energy Digestibility

Feeding raw soybean meal has also been shown to reduce fat

absorption in young chicks; this effect was found to decrease with

increasing age (Nesheim et al., 1962). Borchers (1964) fed raw

soybean meal to chicks, rats, and mice and also noted depressed

dietary metabolizable energy and fat absorption. However Garlich and

Nesheim (1966) fed chicks a crude trypsin inhibitor preparation or

Kunitz trypsin inhibitor and reported that each product when added


to a diet containing heated soybean meal depressed dietary

metabolizable energy but had only a small effect on fat absorption.

Providing extra calories in diets containing raw soybean meal has been

reported to Improve rat growth (Fisher and Shapiror 1963) but not the

gains of growing pigs (Combs et al., 1967).

Feeding raw soybean meal has been observed to greatly enhance

secretion of total fatty acids, phospholipids, cholesterol and bile

into the duodenum when compared to feeding heated soybean meal to

chicks (Sklan et al., 1972). The high secretion rate of fatty acids

was counteracted by an increase In their absorption rate; yetr the

overall net absorption of fatty acids was still slightly reduced

compared to chicks fed heated soybean meal. The same protein

fractions that cause pancreatic hypertrophy, also contract the

gallbladder accelerate bile secretion and decrease fat absorption

(Sambeth et al.r 1967). Serafin and Nesheim (1970) reported that

undigested protein in raw soybean meal may also bind bile acids and

elevate the rate of fecal excretion thereby depressing fat

absorption. It had also previously been postulated that the

depression in protein digestibility may also be partially responsible

for the difference in metabolizable energy values between raw and

toasted soybean meals fed to chicks (Nesheim and Garlich, 1966).

Antibiotic Supplementation

The effect of dietary antibiotics in reversing the growth

depression associated with feeding raw soybeans has received

considerable attention. Supplementation of diets containing raw or

heated soybean mieal with chlortetracycline increased the growth rate


and feed efficiency in rats fed either diets although the improvement

was greater with rats fed the raw soybean meal diet (Hensley et al.,

1953). Including aureomycin in raw and heated soybean meal diets also

increased the absorption of nitrogen and the amino acids (1ysines

leucine, methionine and cystine) in both diets (Carroll et al.,

1953). Similar to the work of Hensley et al.s the increased

absorption was greater in pigs fed the raw soybean meal diet.

Likewises the inclusion of .1% procaine penicillin and .1%

streptomycin sulfate in raw soybean diets increased the growth of rats

equal to that obtained by feeding a heated soybean meal diet

(Borcherss 1958).

Braham et al. (1959) noted that the inclusion of procaine

penicillin, chlortetracycline, novabiocins zinc bacitracin or

streptomiyciln increased chick growth in raw soybean diets by 31 to 51%,

but only a 4 to 14% improvement was measured for chicks fed the heated

soybean diets. The raw soybean diets containing the different

antibiotics did not permit equal performance when compared with the

unsupplemented heated soybean diet. In trials with turkey poultss the

beneficial effects on growth by adding antibiotics are negatively

correlated with the amount of raw soybeans contained in the diet

(Linerode et al., 1961).

The inclusion of aureomrycin In a diet containing underprocessed

soybeans fed to pigs from weaning to market weight did not improve

performance compared to pigs fed adequately processed soybean oil meal

(Becker et al., 1953). Sheppard et al. (1967) fed weanling pigs diets

containing raw or heated soybean meal and reported increased growth by

including antibiotics in both diets. However the smallest response


was observed with antibiotic supplementation of the diet with raw


The poor growth associated with feeding raw soybeans has been

suggested by Borchers (1961) to be the result of enhanced deleterious

bacterial activity in the Intestine. The inclusion of antibiotics In

the diet may counteract this condition. Growth inhibition and

increased intestinal nitrogen (protein) were observed to occur In both

conventional and germ-free chicks but Coates et al. (1970) and Hewitt

and Coates (1969) reported growth inhibition in conventional chicks

was significantly greater than in germ-free chicks fed raw soybean

diets. However pancreatic hypertrophy was observed in the germ-free

chicks fed raw soybean meal as well as in the conventional chicks.

They postulated that the intestinal microflora potentiated the

anti-nutritional effects of raw soybeans by the formation of

additional factors resulting from microbial action on heat-labile

components in raw soybean mieal. Strains of Escherichia Cali were

subsequently shown to be responsible for the adverse effects

associated with the consumption of this raw soybean diet

(Jayne-Williams and Hewitt, 1972).

Barnes et al. (1965), working with chicks and rats, attributed

the beneficial influence of antibiotics on overcoming the growth

inhibiting effects of feeding raw soybean meal to the preservation of

sulfur amino acids from degradation by the intestinal microflora.

Trypsin contains 8.7% cystine and the increased trypsin synthesis and

secretion brought about by feeding raw soybeans can account for

one-half of the cystine excreted by the rat (Barnes et al., 1965).


Researchers using methionine labeled with 3S isotope noted that

ingested trypsin inhibitor stimulated the conversion of methionine to

cystine in the pancreas since high concentrations of radioactive

cystine were found in the small intestine (Barnes and Kwongs 1965).

These researchers postulated that the depletion of methionine through

its conversion to cystine was at least one of the mechanisms causing

growth inhibition. Thus, antibiotics, possibly by reducing the

bacterial degradation of cystine in the lower guts increased the

intestinal absorption of cystine sufficiently to meet the requirement

for synthesis of pancreatic enzymes (Kakade et al., 1970).

The microflora spectrum In the intestinal tracts which can be

modified by antibiotic supplementation, will affect flatulence

production (Rackiss 1966). Pazor et al. (1962) reported that

oligosaccharides comprise about 15% of the air-dried weight of

soybeans with sucrose stachyose, and raffinose present in that order

of abundance. Stachyose and raffinose are thought to produce large

amounts of gas in the lower digestive tract. These compounds having

low molecular weights, are comprised of a-galactosidic and a-fructo-

sidic linkages. Since most animals do not have the digestive enzyme

a-galactosidase, the intact oligosaccharides enters the lower

intestine where they are metabolized to such gasses as carbon dioxide,

hydrogen and to a lesser extent methane (Liener, 1981).

Another response observed from antibiotic supplementation was an

alteration in pancreatic enzyme production (Goldberg and Guggenheimr

1964). One hour after feeding, the tryptic and amylolytic activities

of the pancrease were much lower in rats fed raw compared to heated

soybean flour diets. Inclusion of aureomycin or penicillin in the

diet containing raw soybean flour diminished this reduction in

pancreatic enzyme activity that was brought about by feeding raw

soybeans. Addition of penicillin either through the diet or by

subcutaneous injection improved the protein efficiency ratio and

reduced the loss of proteolytic enzymes in the gut. However the

Improvement was not equal to that found with rats fed diets containing

heated soybean flour without antibiotic supplementation.

Mineral and Yitamin .Supplem;entat ipp~o

The supplementation of trace minerals to the diet has also been

shown to improve the nutritional value of raw soybean products. The

addition of raw soybeans to the diet consequently Increased the

dietary requirement of cyanocobalamine (vitamin B12), vitamin D3r

calcium, phosphorus, zincs iron, copper and molybdenum (Rackis,

1981). In addition, Weaver et al. (1984) found that the

bioavailability of iron from defatted soyflour was relatively high and

addition of vitamin C did not significantly enhance absorption of iron

from raw or heated soyflour.

Phytic acid is located in the 7S protein fraction of soybeans In

the form of a soluble protein-phytate salt complex with significant

amounts specifically deposited In the globold inclusions of the

soybean seed (Prattley and Stanley, 1982). Research has also

indicated that the phytic acid content of soybeans was involved in

reducing the availability of calcium, magnesium, zinc, copper and iron

by the formation of complexes which are poorly absorbed (Davis et al.,

1962; Lieners 1981; Ellis and Norris, 1981). However other


researchers have reported that the small amount of phytate present In

the soybean protein does not affect the bioavailability of copper

(Grace et al., 1984), iron (Welch and Van Campen, 1975) or magnesium

(Lo et al., 1980). Recent evidence has indicated that heating raw

soybeans can reduce their phytic acid content (Liener, 1981).

The Inclusion of unheated soybean meal in diets of chicks

increased their susceptibility to rickets unless higher than

recommended levels of vitamin D3 were added to the diet (Carlson et

al., 1964). Autoclaving the soybean meal or supplementation with

calcium and phosphorus also eliminated the occurrence of rickets

(Jensen and Mraz, 1966).

Raw soybeans contain a heat-labile substance that increased the

requirement for vitamin Bl2 in rats (Edelstein and Guggenheimr
1970a). The metabolites associated with enzymes that require vitamin

812 as a coenzyme are also increased (Edelstein and Guggenheimr
1970b). Supplementation of raw soybean meal diets with vitamin Bl2

stimulated growth of rats (Rackis, 1981). However, Ward et al. (1986)

recently reported that raw soybeans did not enhance Bl2 turnover or

Impair B12 absorption in chicks.

Supplementing diets containing raw soybeans with other trace

minerals has been shown to Improve animal performance. The addition

of 125 ppm copper sulfate to diets containing raw soybeans increased

average daily gain and feed efficiency of market hogs (Young et al.,

1970). The researchers did not speculate whether the response was due

to the copper per se or indirectly from the bacteriostatic effect of

the copper sulfate. Linerode et al. (1961) obtained no nutritional


benefit from adding a zinc supplement in a turkey diet containing raw

soybeans. Other studies utilizing lodine-deficient diets have

Indicated that feeding raw soybeans caused marked enlargement of the

thyroid glands of rats and chicks, an effect which could be counter-

acted by administration of potassium iodide or partially eliminated by

heat processing soybeans (Patton et al., 1939; Block et al., 1961).

Other Anti-Nutritional factors

Other constituents of soybeans have been Imrplicated as being

responsible for the reduced performance commonly associated with

feeding raw soybeans. For examples hemagglutinins, a glycoprotein

(Lis et al., 1966), was isolated from soybeans in 1952 (Liener and

Pallensch, 1952) and later determined to comprise an estimated I to 3%

of the protein of defatted soybean flour (Liener and Roses 1953).

Hemagglutinins, also known as lectinss appear to function similarly to

trypsin inhibitors in the soybean plant as defensive proteins that

protect the plant from insect invasion and are appropriately located

on the surface of the plant cell (Lehningerr 1982). Lectins bind to

certain carbohydrate groups, D-galactose and N-acetyl-D-galactosaminer

on the cell surface. Soybean lectins have been shown to be readily

destroyed by heat treatment and their destruction was accompanied by a

marked improvement in the nutritive value of the protein when fed to

chicks (Liener and Hills 1953). However rats fed soybean extracts,

from which the lectins had been removed by affinity chromatography,

grew just as poorly as those consuming the original crude soybean

extract (Liener, 1981). Therefore, it appears that lectins do not

play a major role In reducing the nutritional quality of soybean


Another anti-nutritional factor, saponins which has been found in

some plants to have an adverse effect on animal growths is also a

constituent of soybeans. However, feeding chicks, rats, and mice

diets supplemented with three times the level of saponin found in

soybean flour was not detrimental to performance (Ishaaya et al.s


Soybean Processing

As previously discussed, heat processing inactivates several

anti-nutritional factors in soybeans, such as trypsin inhibitors,

hemagglutinins, goitrogens, antivitamins and phytates (Liener, 1981).

However, soybeans contain other proposed anti-nutritional factors,

such as saponins, estrogenic compounds, and flatulence compounds,

which are heat-stable. Also, some components of raw soybeans

(pyridoxines total and free folacin) which are needed for growth are

reduced when soybeans are heated (Soetrisno et al., 1982). Sugawara

et al. (1985) found that the green or grassy odor of soybeans

disappeared or decreased by heating but the beany odor remained even

if the soybeans were heated for 8 hours.

Water-extractable proteins from soybeans can be separated into

four fractions with approximate sedimentation rates of 2, 7, 11 and

15Ss and comprise 22, 37s 31 and 11% of the total protein in soybeans,

respectively (DeMans 1980). Trypsin inhibitors are located in the 2S

fraction. The 7S fraction contains lectinsl lipoxygenase (the enzyme

that catalyzes the oxidation of lipids) and 7S globulin; whereas, the

11S fraction consists mainly of 11S globulin (glycinin). The 7S and

11S (glycinin) globulins are the major storage proteins of soybean

seeds. The 7S globulin a glycoprotein, is present as a monomer with

molecular weight of 180,000 to 210,000 daltons. Glycinin is a large

molecule of 290,000 to 320,000 dalton molecular weight composed of 12

subunits having a rigid globular conformation (Kitamura et al.,

1976). The 15S fraction has proteins with molecular weights

approximately 600,000 daltons.

The application of heat to soybean protein has been shown to

alter the hydrogen and hydrophobic bonds which resulted in decreased

water solubility of the proteins (DeMan, 1980). Kakade et al. (1973)

suggested that native undenatured soybean protein Is in itself

refactory to enzymatic attack unless denatured by heat. Subsequent

research indicated that glycinin in raw soybeans resists proteolytic

attack (Kamata et al., 1979). Heating glycinin at 100 C increased the

digestibility by denaturation or unfolding the conformation of the

protein (form digested at a fairly high rate). However, heating

glycinin at 120 C caused decreased digestibility suggesting that

excess heat causes refolding of the protein in a new conformation that

is more resistant to enzymatic attack.

The heating of raw soybeans has been shown to increase the

nutritional availability of sulphur and nitrogen in diets fed to rats

(Johnson et al.r 1939). The availabilities of methionine and cystine

in trials with chicks increased until a temperature of 120 C was

obtained; after which the availabilities then decreased (Evans and

McGinnis, 1946). Carroll et al. (1953) reported that proper heat

treatment of soybean meal increased absorption of lysine, leucfne and

methionine in rats but the greatest improvement was with cystine

absorption. In addition, plasma levels of threonine and trytophan in

rats were increased markedly by heating the soybeans, indicating

increased digestibility and absorption of these amino acids (Rao et

al.r 1971).

Effects of Overheating Soybeans

Overheating soybean protein impaired its nutritional quality.

Evans and associates (Evans and McGinnis, 1946; Evans and McGinnis,

1948; Evans et al., 1951; Evans et al., 1962) have conducted extensive

research with overheated soybean protein. .They reported that with

chickens, cystine was the limiting amino acid In overheated soybeans.

Methionine availability was also decreased when soybean protein was

autoclaved at 120 C. These workers indicated that 31% of the total

cystine was destroyed and 25% became inactivated with autoclaving.

The relative availabilities of cystine and methionine by chicks were

increased by autoclaving raw soybeans at 100 C for 30 minutes; but

autoclaving the soybeans at 130 C for 60 minutes reduced the quantity

of cystine utilized to the level equivalent of feeding raw soybeans

while methionine utilization was decreased. In additions adding

sucrose or glucose to soybeans increased inactivation of cystine by

four fold. Subsequent research has shown that some of the soybean

proteins appear to have more heat labile cystine than other soybean

proteins. Iriarte and Barnes (1966) presented data which indicated

that the amount of heat required to destroy the growth inhibitors also

destroyed some of the cystine; such that it was the first limiting

amino acid. More recent results by Crissey and Thomas (1983) further

Illustrated the reduced availabilities of methionine and lysine due to

extreme overheating of soybeans. These researchers fed roosters

commercial soybean meal (SBM) that had received additional

autoclaving for one hour at 121 C then dried the meal for 30 minutes

and reported a 38 and 22% increase In the excretion of methionine and

lysine, respectively.

Lysine destruction can also be used as a indicator in determining

the nutritive value of overheated soybeans. Evans and Butts (1948)

and Evans and McGinnis (1948) reported a greater proportion of the

lysine was made inactive than was destroyed. Their results showed a

greater percentage of lysine inactivated with an increase in heating

temperature. The addition of sucrose or glucose further increased the

binding of lysine to a form which was unavailable to enzymatic

digestion. In addition to cystine, methionine and lysiner other amino

acids including argininer tryptophan histidine and serine were

partially destroyed or denatured by excessive heating of soybean meal

(Lienerr 1958; Skrede and Krugdehlr 1985).

Kasarda and Black (1968) implied that overheating of soybean

protein was a major source of ammonia evolution which could react with

the available reducing sugars to form pyrazine compounds. Scott et

al. (1969) noted that soybean products contain reducing carbohydrates

such as glucose and these sugars can react with the free epsilon amino

group of lysine in the soybean protein. This reaction is termed the

"Maillard" or "browning" reaction. Heat treatment accelerates the

formation of carboydrate-amino group complexes which are resistant to

enzymatic hydrolysis. The result is that amino acids (especially

lysine) become nutritionally unavailable. Knipfel et al. (1975)

autoclaved soybean protein with different carbohydates at 121 C for 0

to 1280 minutes. Soybean protein autoclaved with sucrose, glucose

or fructose reduced rat growths net protein ratio and protein

digestibility more than soybean protein which was autoclaved with

starch and cellulose or without the added carbohydrates. Plasma

lysine was reduced more than plasma methionine when rats were fed

soybean protein heated with glucose fructose and sucrose.

Supplementation of amino acids to the diets containing overheated

soybeans has been studied in trials with chicks by Evans and McGinnis

(1946). The addition of cystines lysine, and methionine to the diet

improved chick growth. Single addition of any of these amino acids

simply did not ameliorate the depressed growth obtained from feeding

overheated soybeans.

Improvement of t~he Nutritive Value of Soybeans by Heat Processing

Many factors such as heating temperature, pressure, and time are

very important In obtaining the maximum nutritional value from heating

soybeans. Optimum chick growth required heating soybeans 30 minutes

at 107 C (Carew et al., 1961). These heating conditions confirmed

earlier work of Evans and McGinnis (1946) which obtained the best

chick performance when feeding soybeans that were also heated for 30

minutes at 100 to 120 C, although they found that no processing

temperature above 120 C further improved protein quality for chicks.

Subsequent research by Arnold (1973) indicated that heating

temperatures above 120 C could produce optimum chick performance. He

reported that optimum performance was obtained by heating soybeans at

149 to 160 C for 10 minutes or 171 to 194 C for 5 minutes.

Featherston and Rogler (1966) reported improved soybean protein

utilization when compared to the other treatments in chicks fed

full-fat soybean flakes heated at 133 C for 20 minutes. Simovic et


al. (1972) studied even higher temperatures and indicated that

temperatures up to 158 C could be used satisfactorily but, above this

temperature, the soybeans became charred. The temperatures were 95.5,

111.0, 127.0 and 158.0 C with heating durations of 3.01 2.5, 1.5 to

2.0, or 1.0 minutes respectively.

Noland et al. (1966) reported the best heating treatment for

soybeans was 60 minutes at 115 C for pigs and 24 to 72 minutes at 115

C for rats. Feeding early weaned pigs, Lawrence (1967) obtained

better heating results when soybeans were autoclaved for 24 to 36

minutes at 110 C than for longer or shorter periods of time.

Subjecting raw defatted soybean flakes to live steami at atmospheric

pressure (approximately 100 C) for 10 or 60 minutes permitted similar

rate and efficiency of growth and feed Intake when fed to weanling

pigs (Clawson et al., 1981). Prior to heating the raw soyflakes were

mixed 5:1 (w/w) with water. Becker et al. (1953) reported superior

growth of pigs from weaning to market when fed diets containing

soybeans toasted in a French cooker for 38 to 55 minutes at 99 to 116

C compared to pigs fed diets containing soybeans toasted for 18 to 20

minutes or 33 to 37 minutes at 99 to 104 C.

Hayward et al. (1936a) Indicated that autoclaving soybeans for

one hour at 1.06 gm/cm2 pressure doubled the nutritive value of the

protein when fed to rats. Westfall and Hauge (1948) demonstrated

that soybean flour heated at 108 C for 15 to 30 minutes resulted In

superior soybean protein when included in a mouse diet. Likewise,

Rackis (1966) achieved maximum protein efficiency and trypsin

inhibitor inactivation of full-fat and defatted soyflakes by steaming


at 100 C for only 15 minutes In trials with rats. Subsequent research

using higher heat processing temperatures suggested that the optimum

heating period for ground whole soybeans fed to rats was 10 to 20

minutes at 120 or 132 C compared to soybeans heated at 100 C= (Seerley

et al., 1974).

The oil from soybeans can be removed by various extraction

processes and the heating time required to produce a nutritionally

optimum soybean meal varied among the different methods (Hayward et

al., 1936b). To produce optimum quality proteins expeller processed

meal needed 2 to 3 minutes at 112 to 150 Cs hydraulic processed meal

needed 90 minutes at 105 to 124 C, and the solvent extracted meal

required 15 minutes at 98 C. Ethano7 extraction of raw soybeans

Improved performance of pigs compared to the feeding of raw

unextracted soybeans, yet the improvement was still lower when

compared to soybeans that were heat-treated (Hancock et al., 1985).

Another method for extraction of the soybean oils supercritical carbon

dioxide at a temperature of 80 C for 8.3 hours, did not denature the

trypsin inhibitor activity of raw soybeans (Pubols et al., 1985).

Toasting at 109 C for 5 minutes was required to inactivate the toxic

factors and permiit optim~um chick growth (McFarland and Pubolss 1982).

Renner and Hill (1960) also noted a variation in nutritive value

between extracted dehulled raw soyflakes, and ground raw soybeans when

fed to chicks. The extracted dehulled raw soyflakes had received

optimum heat processing with a wide range of heating times (10 to 60

minutes) at 107 C whereas the ground soybeans had an optimum cook of

10 minutes at 107 C. These products produced maximum metabolizable

energy values and rate and efficiency of chick growth.

The moisture content of soybeans prior to heat processing greatly

influences the heating time required to produce quality soybean

products (Albrecht et al. 1966). Trypsin inhibitors were readily

destroyed by atmospheric steaming for 20 minutes provided the soybeans

contained about 25% moisture before steaming, whereas soybeans

containing lower inital moisture content required a longer heating

time or higher processing temperatures. However, when moisture

content of the soybeans was elevated to 60% by soaking overnight then

only 5 minutes of atmospheric steaming was required to inactivate the

trypsin inhibitors. Likewise Waldroup et al. (1985) reported that

the addition of 12% water to soybeans prior to heating decreased the

time required to produce adequate protein supplements for chicks.

However, Seerley et al. (1974) fed rats soybeans which had 10 or 20%

moisture added prior to heating at 118 C for 40 minutes and noted no

difference in their performance that could be attributed to initial

moisture level. There is evidence to suggest that moist heat can

reduce the damaging effects of overheating soybean protein (Renner et

al., 1953).

Other methods used to denature the anti-nutritional factors of

soybeans and improve their nutritional value have been studied. For

example the addition of thiols such as cysteine and N-acetyl-cysteine

facilitates heat inactivation in the temperature range of 25 to 85 C

thereby decreasing the damage to heat sensitive amino acid residues

such as cystine, methionine and lysine (Leir et al., 1981; Friedman et

al., 1982). This method also increased the limited sulphur content of

soybeans. It was postulated that the cations of thiols were involved

in the formation of mixed disulfide bonds between the added thiols and

enzyme inhibitors and structural proteins. The addition of cysteine

prior to heating can also facilitate the heat inactivation of trypsin

inhibitors in both purified soybean Kunitz inhibitor and soybean

extracts, thereby indicating increased protein digestibility, protein

efficiency ratio and nutritive value of soybeans (Friedman et al.,


Other Methods of Processing Sqybeans

Germination of whole raw soybeans has also been reported to imp

ove their nutritional quality by increasing the protein dispersibility

index (PDI) and improving the odor and flavor scores as evaluated by

an 8-member panel test (Suberbie et al., 1981). During the early

phases of germination, reserve proteins were mobilized at a steady

rate, but carbohydrates were quickly depleted (McAlister and Kroberr

1951). Mobilization of fat began immediately after depletion of the

carbohydrate in the seed. Thus, the non-protein nitrogen content

doubled (Becker et al., 1940) and the level of offending

oligosaccharides were reduced markedly during germination (Pazor et

al., 1962; East et al., 1972). Sprouting also reduced the energy, dry

matter, total lipids, starch and concentrations of the amino acids

alanine, arginines glutamic acid, threonines glycines lysine, proline

and serine compared to ungerminated soybeans (Peer and Leesons 1985).

These researchers further noted an increased ash, aspartic acid and

leucine but no change in the concentrations of histidiner isoleucine,

methionine, phenylalanine and tyrosine due to germination. Since the

total protein content did not change, the percentage of Individual

amino acids was altered.


The trypsin inhibitor content of soybeans during germination has

been reported to change very little (Desikachar and Des 1947; Collins

and Sanders, 1976). However, other researchers have noted a reduction

in trypsin Inhibitor as well as lipoxygenase activity during

germination (Suberbie et al., 1981). Freed and Ryan (1978a) found a

13% reduction In the Kunitz trypsin inhibitor activity after seeds had

been sprouting for nine days. Bates et al. (1977) obtained a 33%

reduction In total trypsin Inhibitor activity after only four days of

sprouting and Collins and Sanders (1976) reported that up to 13% was

reduced after germinating the soybeans for only three days. Likewise,

Peer and Leeson (1985) reported that trypsin Inhibitor activity was

reduced in a cubic trend with increased sprouting time. Freed and

Ryan (1978b) found that only modified forms of the Kunitz inhibitor

appeared during germination.

The actual benefit of germinating soybeans per se on the

reduction of trypsin inhibitor activity in soybeans miay be limited.

Collins and Sanders (1976) found that soaking four different varieties

of soybeans in water for one hour reduced their trypsin inhibitor

activity between 3.4 to 10.2%6.

Germination increased the nutritive value of soybeans fed to rats

(Desikachar and De, 1947) but not for chicks (Mattingly and Birds

1945). Everson et al. (1944) confirmed that sprouted soybeans had

better feeding value for rats compared to the raw unsprouted

soybeans. However, autoclaying both products provided additional

growth responses.


Fermentation has also been used to improve the nutritional value

of soybeans (Zamora and Veum, 1979). After heating whole soybeans at

121 C for 30 minutes, they were fermented with Aspergillus oryzae or

Rhizopus oliqosporus. During fermentation, dry matter was reduced by

3% and the percentages of lysine, leucine and methionine increased

slightly. Feeding pigs a diet containing soybeans fermented with

Asperqillus oryzae resulted in faster growth than pigs fed an

unfermented (only heated) whole soybean diet. Feed efficiency,

metabolizable energy and nitrogen utilization for pigs fed the

fermented soybean diets were increased compared to pigs fed soybean

meal and unfermented soybean diets.

Treating raw soybean meal with formaldehyde (1% of the mieal

protein content) inactivated 99, 97 and 40% of the trypsin- and

chymotrypsin-inhibiting and urease activities, respectively (Nitsan

and Bruckentalr 1977). When fed to chicks both soybean diets (raw

and heated) containing formaldehyde reduced body weight gain, feed

efficiency and protein efficiency ratio compared to chicks fed either

diet unsupplemented with formaldehyde. The reduction in these

measured parameters due to the Inclusion of formaldehyde in the diet

containing heated soybeans was greater when compared to the diet

containing unsupplemented raw soybeans. However, pancreatic

hypertrophy and pancreatic trypsin, chymotrypsinr lipase and amylase

activities were decreased by adding formaldehyde to the raw soybean

diet. Also, these researchers reported that formaldehyde addition in

the raw soybeans resulted in increased intestinal trypsin content but

decreased amylase in both the small intestine and cecum when compared

to feeding the unsupplemented raw soybean diet. Cecal trypsin and

chymotrypsin activities did not differ significantly between these two


Dal Borgo et al. (1967) reported that young chicks were able to

utilize raw soybeans more efficiently when fed in combination with

glucose or sucrose than when starch was provided as the source of

carbohydrate. These effects were not obtained with heated soybean

meal. Recent research by Pontif et al. (1986a) reported that changing

the dietary carbohydrate source from corn to wheat in raw and heated

soybean diets resulted in reduced average daily gain. No reduction

was noticed when the carbohydrate source was comprised of a mixture of

corn and wheat.

Swine Feeding Trials

Numerous processing methods have been reported to yield quality

soybean products. Among these are autoclaving (Combs et al., 1967)r

roasting (Bairdr 1983), extrusion (Baird, 1983), and the use of

microwaves (Hafez et al., 1985; Fuller and Owings, 1986). The heat

generated during the process of pelleting was not adequate to produce

a quality soybean supplement (Hanke et al., 1972; Hooks et al.,


The superiority of feeding heated soybeans compared to feeding

raw soybeans to pigs is well documented. Performance of 8-week-old

pigs was improved by feeding soybeans heated in an autoclave (Combs et

al., 1967). Roasting and extrusion were also adequate heat processing

methods of soybeans fed weanling pigs (Noland et al.r 1976). However,

Rust et al. (1972) reported that roasting (141 C) and extrusion (125

C) of whole soybeans were inadequate heat treatments compared to SBM


fed to pigs weaned at 21 days of age, although feed efficiency was

similar between pigs fed the SBM and the two heated soybean diets.

Similar adverse effects due to feeding roasted soybeans to weanling

pigs were noted by Crenshaw and Danielson (1985a). Pelleting soybean

meal or raw soybean diets fed to four-week-old pigs did not improve

performance of pigs fed either diet (Crenshaw and Danielson, 1984a).

Adams and Jensen (1985) compared the feeding value of soybeans

that had been extruded or roasted for weanling pigs and reported that

extrusion increased digestibilities of fat, dry matter, energy and

nitrogen and metabolizable energy compared to roasting. In addition,

grinding the roasted soybeans through a fine as opposed through a

coarse screen did not affect their feeding value. Infra-red roasting

of soybeans fed to weanling pigs provided equal growth and feed

efficiency but reduced digestiblities of dry matter, ether extract and

energy compared to pigs fed extruded soybeans (Faber and Zimmerman,

1973). However, the biological value of the two products was equal.

Kinyamu et al. (1985) compared the performance of weanling pigs

fed diets containing SBM, extruded soybeans or Jetsploded soybeans

having trypsin Inhibitor activities of 6.3, 8.1 and 28.0 units per

milligram, respectively. The Jetsploded method requires that the

soybeans be heated at a high enough temperature to cause the soybeans

to burst. The extrusion method uses both steam~ and pressure

(friction) to process the soybeans. The pigs fed the diet containing

Jetsploded soybeans grew slower and were less efficient compared to

pigs fed diets with soybean m~eal or extruded soybeans. Digestion

coefficients for each diet followed a similar trend as the performance



A preponderance of research on feeding soybeans to

growing-finishing pigs has established that raw soybeans are an

Inadequate protein source and require heat processing for maximum

performance of growing pigs. However, research on the necessity to

heat soybeans for finishing pigs is Inconclusive. Jimenez et al.

(1963) found that the growth of growing-finishing pigs was improved by

feeding whole soybeans which had received a heat treatment. Soybeans

have been subjected to various methods of heat processing to produce

adequate soybeans fed to growing-finishing pigs. Among the different

processing methods are infra-red heating (Noland et al.r 1970;

Wahlstrom et al., 1971), dry-roasting (Hanke et al.r 1972; Baird,

1983; Miller et al., 1985), extrusion (Noland et al., 1969; Hanke et

al.r 1972; Baird, 1983), cooking in a six-stack French cooker at 110 C

for 10 minutes (Seerley et al.r 1974) and cooking soybeans for six

hours in a 40 gallon feed cooker mixed in a 1.0:1.7 ratio with water

(Young, 1969).

Chin and Diggs (1986) reported higher values for digestible and

metabolizable energy, nitrogen-corrected metabolizable energy and

digestible nitrogen from feeding roasted soybeans compared to raw

soybeans but the percentage nitrogen retained did not differ between

the two protein supplements.

Roasting temperature is critical to produce a soybean product

having optimum nutritional quality for swine (Campbell et al.r 1984).

These researchers reported that roasting soybeans at 110 C did not

permit similar growth of growing-finishing pigs fed SBMI although a

roasting temperature of 125 C produced soybeans which had equal

nutritional value as that of commercial SBM. Likewise extruding

soybeans at 132 and 143 C provided equal growth to SBM yet improved

growth compared to pigs fed soybeans extruded at 115 C (Seerley et

al., 1974). Crenshaw and Danielson (1985b) substantiated earlier work

which indicate that roasted soybeans were adequate protein supplements

for growing-finishing pigs but also noted that barrows significantly

utilized roasted soybeans more effectively than gilts.

Recent research has provided additional evidence that raw

soybeans are not an adequate protein supplement for growing-finishing

pigs and performance was not improved when soybeans were substituted

in place of SBM on a equal weight basis or when diets were

reformulated to be isonitrogenous diets (Crenshaw and Danielson,

1984b). Likewise, replacing the SBM in the diet fed to growing pigs

with 0, 33, 67, and 100% raw soybeans resulted in a linear decrease in

average daily gain, feed efficiency, feed intake and loin-eye area

(Pontif et al., 1986b).

The age of the pig may influence the extent to which the

anti-nutritional factors in raw soybeans affect animal performance.

The beneficial effects of feeding heated soybeans to swine decreased

with pig maturity (Combs and Wallace, 1969). These researchers

compared effects of feeding diets containing SBM, raw and autoclaved

soybeans to three-r nine- and 16-week-old pigs. Performance of the

three- and nine-week-old pigs was adversely affected by feeding diets

containing raw soybeans. The heated soybean diet was inadequate for

three-week-old pigs when compared to the SBM diet but yielded similar

performance when fed to nine-week-old pigs. Sixteen week-old pigs

were able to utilize heated and raw soybeans as effectively as

properly heated SBM. Recent research indicated that feeding raw

soybeans to growing-finishing pigs resulted in reduced performance

regardless of initial age and weight of pigs (Crenshaw and Danielson,

1985c). Likewise Pontif et al. (1986b) reported that the replacement

of dietary SBM with raw soybeans when pigs weighed 59 kg decreased the

rate and efficiency of growth and feed intake. However the

significance level on the growth data was between .05 and .10%.

Jensen et al. (1970) and Hanke et al. (1972) reported also that raw

soybeans were not an adequate protein supplement for finishing pigs

and the heat generated during pelleting was not sufficient to improve

their nutritional value.

In more mature swine, gilts and sows, the data on feeding raw

soybeans have been more consistent. Gestating swine were able to

reproduce normally through second and third parties, when fed diets

containing raw soybeans as the only source of supplemental protein

(Crenshaw and Danielsons 1983; Crenshaw and Danielson, 1985d) SBM as a

protein source in gestation and lactation diets (Danielson and

Crenshaw, 1984). Allee et al. (1985) confirmed that feeding raw

soybeans to sows would not impair their reproductive performance but

the sows fed raw soybeans lost more weight during lactation than sows

fed SBM. Noland et al. (1971) reported that the whole soybean plant

(pods, leaves and stems) could be pelleted to provide adequate protein

supplementation for gestating swine. The plant was cuts dehydrated

and pelleted when the largest pods were full grown, seeds were full

size and leaves were just beginning to yellow. This soybean product,


supplemented with ground corn, vitamins and minerals provided

satisfactory nutrition for gilts and sows when self-fed through the

gestation period.

Poultry Feeding Trials

By comparison, the inconsistencies of feeding raw soybeans

related to swine maturity are similar to that of feeding raw soybeans

to poultry. Waldroup (1982) presented an excellent review on feeding

whole soybeans to poultry so the following discussion pertains only to

changes in maturity. Similar to weanling pigsr the inclusion of raw

soybeans to the diet of broiler chicks adversely affected growth

compared to chicks fed heated soybeans. Ogundipe and Adams (1974)

reported depressed body weight of pullets when feeding raw unextracted

soybeans from 11 to 20 weeks of age, although Saxena et al. (1963) had

previously reported that chicks 6 to 12 weeks of age could effectively

utilize diets containing raw soybeans. Barnstein and Lipstein (1962)

suggested that chicks of any ages when fed a diet containing raw

soybeans, would have a reduction in growth but as chicks matured, the

time of adaptation to raw soybeans decreased. Several researchers

have indicated that raw soybeans could replace SBM in laying hen diets

without a reduction in egg production (Saxena et al., 1963; Latshaw

and Claytons 1976; Latshaws 1974). However, other researchers have

reported that raw soybeans were not an acceptable protein supplement

for laying hens (Fisher et al., 1957; Rogler and Carricks 1964;

Waldroup and Hazens 1978). Thuss raw soybeans could be processed to

provide an acceptable protein supplement for laying hens by roasting

and extrusion (Waldroup et al., 1969; Waldroup and Hazens 1978)r


adequately supplementing the diet with methionine (Salmon and

McGinnis, 1968) or including methionine and/or vitamin Bl2 In the

raw soybean diet (Fisher et al.s 1957). In contrast, Waldroup et al.

(1969) reported that adding synthetic lysine and methionine to layer

diets containing raw soybeans did not ameliorate the adverse effect on

egg production.

Composition of Commercial Soybean Keal

The National Soybean Processors Association considers that SBM

(48% protein) is of acceptable quality when it has less than 12%

moisture greater then 48% crude protein and less then 3.4%: crude

fiber (Jones, 1984). Hill and Renner (1960) reported limited

variation in protein, fat, fiber, ash, and metabolizable energy

content of SBM. However a more recent study of SBM samples obtained

from 18 different feed companies and five different SBM manufacturers

collected during a 6-year period demonstrated that significant

differences in fat and fiber content of SBM occurred among samples

taken in different years (Joness 1984). Also, moisture, protein and

fiber contents were significantly different between SBM samples

obtained from the different suppliers. The number of samples that did

not attain the minimum acceptable quality standards increased from

39.6%6 In 1976 to a high of 83.3% in 1983. Upon further investigation,

it was noticed that most deficiencies (40 to 60%) occurred in samples

having greater then 12%6 moisture. Furthermore variation has also

been observed In amino acid content of SBM (Whitacre et al., 1984).

The researchers found that the average content: of niethionines lysine

and protein in 175 SBM samples was .6, 2.6 and 46.7%, respectively.

For comparison, the National Research Council (1979) considers the

content of methionine, lysine, and protein to be .7, 3.2 and 48.5%


Although the nutritional composition of soybeans used In the

processing of various soybean meals is very important it is necessary

to subject soybeans to an adequate amount of heat processing.

Longenecker et al. (1964) found marked differences in manufacturing

processes for soybean concentrates and reported that In most instances

heat treatment was not adequate to support maximum growth when

included in diets for weanling rats. Several researchers have

suggested that commercial SBM is heated for much longer periods than

is required to produce optimum quality protein supplements for

weanling pigs (Lawrencer 1967) and rats (Noland et al.r 1966).

Campbell (1974) fed weanling pigs SBM which had been processed in

Arkansas and in Illinois and the rate andu efficiency of growth of pigs

fed the Arkansas SBM were increased compared to pigs fed the Illinois

SBM. However, neither SBM equaled the performance obtained by feeding

adequately processed full-fat soyflakes. Although not necessarily

confirming SBM is overheated research by Clawson et al. (1981)

demonstrated that SBM was not underheated. These researchers fed

weanling pigs SBM with and without 30 minutes of additional heating

with live steam at atmospheric pressure (approximately 100 C)r

reducing the trypsin inhibitor units from 7 to 2 per mgr and reported

no difference in pig performance.

Other researchers have indicated that SBM received Insufficient

heat processing for optimum performance of day-old chicks (McNaughton

et al., 1981). The SBM was autoclaved at a temperature of 110 C in 5

minutes increments. Extra processing was required before the broilers

could attain their genetic potential for growth. The best performance

was obtained by feeding broilers SBM autoclaved an additional 10


As previously discussed heat processing improved the nutritive

value of soybeans by inactivating anti-nutritional factors. Trypsin

inhibitors are considered the most important anti-nutritional factor

in soybeans. However soybeans also contain the enzyme urease which

has little nutritional importance In poultry and swine diets. It is

significant that the urease content is currently used by the SBM

processing plants to evaluate optimum processing conditions. The

analytical method developed by Caskey and Knapp (1944) monitors the

urease-urea relationship in the soybean. In this method, urea is

degraded by urease and the pH of the media rises due to formation of

ammnonia and the resultant change in pH is recorded as the urease

activity (UA). This analysis is easier and quicker to conduct than

determining the trypsin inhibitor content. Since the heat treatment

of soybeans required to denature the urease content parallels the heat

treatment necessary to inactivate trypsin inhibitors the UA Is

considered a good indication of adequate heat processing (Caskey and

Knapp, 1944; Albrecht et al., 1966; MlcNaught~on and Reece, 1980). In

additions the inactivation of trypsin inhibitor and urease activities

during heat processing precedes the destruction of lysine (M~cNaughton

and Reece, 1980). However, Borchers et al. (1947) observed that

urease was more sensitive to heat inactivation than trypsin inhibitors

and should not be regarded as being completely accurate in determining

adequacy of heat processing. These researchers along with Balloun et

al. (1953) reported that the UA is of little value for detecting

overheated soybean meal.

The American Feed Manufacturer Association considers that SBM

having UA in the range of .05 to .20 change in pH units has received

adequate heat processing (Smiths 1977) which is a reduction from a UA

of approximately 2.0 for raw soybeans. However, Harris (1983)

reported a commercial lot of SBM with a UA of .83; thus the quality of

commercial SBM being marketed does vary. Likewise Jones (1984)

assayed 1729 SBM samples taken over three years from four different

suppliers for UA and reported that some SBM samples had UA higher than

1.0. There were significant differences in SBM quality manufactured

by the four different suppliers, UA were .057, .074r .041 and .109,

respectively. The UA value of SBM also increased according to the

year it was produced being .062, .089 and .091 for 1981, 1982 and

1983, respectively. In addition, SBM samples processed in the colder

months (Oct.r Nov. and Dec.) were higher in UA. Soybean meal samples

(277) obtained from one source had 2.88% with UA greater than .91. In

a recent study by Rudolph et al. (1983) comparing the digestibility of

nitrogen and amino acids between different SBM (48.5% and 44.0%) with

soybean flour, the UA of the 48.5% and 44.0% SBM was .46 and .21,

respectively. Also SBM can be overheated as indicated by Noland et

al. (1976). They received SBM having a UA of .02 and reported

unsatisfactory performance when fed to weanling pigs.


Edwards (1983) reported a range In UA between .006 and .208 in

commercial SBM from different processors. One source produced SBM

having consistently higher trypsin inhibitor and UA activities which

concurrently produced a higher incidence of tibial dyschondroplasia

when fed to chicks. However, gains and feed efficiencies were similar

in chicks fed the various SBM.

Early research data indicating optimum heating times and

temperatures without the UA and trypsin inhibitor activities In SBM

are difficult to compare since the actual heat treatment cannot be

determined. Different procedures for estimating trypsin inhibitor

activity were, and are being, reported which hinders using the trypsin

inhibitor activity to compare research data. The UA is a common

analysis and allows more reliable comparison. Rackis (1968) stated

that SBM with UA between 0.05 and 0.15 has been properly heated for

maximum nutritional value. However, Lawrence (1967), feeding early

weaned pigs, reported maximum pig performance when fed SBM with UA of

0.01 to 0.02 as contrasted to either greater or lesser pH change.

Likewise Noland et al. (1976) reported that autoclaved full-fat

soyflakes having a UA of .03 and .05 supported higher feed intake and

better gains of 8.2 kg pigs than soyflakes having UA of .01 and .11

and SBM having UA of .02. In contrast, Hansen et al. (1984a) fed SBM

with and without extra heat processing having UA of .02 and .11,

respectively to pigs weaned at 4 weeks of age and reported no

preference to either diet when given a choice and no difference in pig

performance (Hansen et al., 1984b).


Nitrogen balance and gross energy digestibility were similar in

growing pigs fed SBM with UA ranging between .01 and .19 (Chat-Ju et

al., 1984). Also, apparent ileal nitrogen and individual amino acid

digestibilities did not differ. Likewise, feeding defatted soyflakes

ranging In UA between .05 and .48 to barrows ranging in weight from 25

to 45 kg resulted in no significant differences in nitrogen, amino

acid and energy digestibilities and nitrogen retention between the

different soyflake treatments (Vandergrift et al., 1983). Pig growth

and feed efficiency were not influenced by the imposed dietary

treatment groups. Rudolph et al. (1983) confirmed the high

nutritional value of soybean products (48.5% SBM) having a high UA of

.46 for 38 kg barrows. The barrows in Rudolph's and Vandergrift's

studies were limit-fed. One adverse effect associated with feeding

low quality soy products to swine has been depressed feed intake.

Seerley et al. (1974) fed extruded soybeans to 28.9 kg pigs having a

UA of .53 and reported that the growth of these pigs did not differ

from pigs fed SBM. However, the amount of feed required per unit of

weight gain was increased in pigs fed the soybean product having a UA

of .53. The pigs were full-fed but feed intake was not reported. The

extruded soybeans having a UA of 1.83 In this study were inadequate

protein supplements for growing pigs.

Research on the optimum heat processing of SBM fed to chicks

indicates variation similar to that reported for pigs. McNaughton et

al. (1981) fed day-old chicks commercial SBM containing a UA of .19

compared to the same SBM having additional heating to reduce the UA to

.02. The best performance was obtained by feeding broilers the

autoclaved SBM having a UA of .02. More recent research has indicated

no difference in amino acid availability in SBM with a UA of 0 and .10

(Dale et al., 1986). However SBM having a UA below .05 was

previously thought to be overheated. Also, additional research has

indicated that SBM with UA as high as .50 was an adequate protein

supplement for broiler chicks (Waldroup et al., 1985). Likewise, Mian

and Garlich (1985) found no difference in growth of turkey poults fed

SBM with UA between .14 and .90. However, soybean meals with UA of

.02 and 1.50 were inferior to the other treatment groups.

Some differences In the UA of SBM being marketed may be

attributed to the percent moisture of the soybeans before processing

(McNaughton and Reece, 1980). These researchers reported that

increasing the moisture content of defatted soyflakes enhances the

denaturation of urease and trypsin inhibitors. Mustakes et al. (1981)

reported that the quality of SBM was influenced by heating timer

jacket-steam pressure and moisture content. The moisture content of

SBM prior to entering the toaster was directly affected by the hexane

level remaining after the oil removal process.

Although the trypsin inhibitor activity of soybean meal is

considered important in livestock and poultry feeds, Gillette et al.

(1978) found no correlation between growth and the trypsin inhibitor

activity in commercial SBM being fed to rats. Likewise, Kakade et al.

(1972) reported no correlation between trypsin inhibitor activity and

protein efficiency ratio In rat diets containing over 100 different

varieties of soybeans.

Another assay to facilitate the determilnation of optimum soybean

processing was developed by McNaughton et al. (1981). They reported

that optimum heat processing could be determined by observing the

change In color of the soybeans. These researchers used a Hunterlab

colorimeter on the +a band to predict the trypsin inhibitor content in

autoclaved SBM. However, Rudolph et al. (1983) determined color

values of various soybean products using the procedure described by

McNaughton and reported the +a color values did not coincide well with

the trypsin inhibitor values of the different soybean products.

A method which can be used to determine the protein quality of

soybean meal and other protein sources fed to chicks is measuring the

uric acid excretion (Miles and Featherston, 1976). Protein efficiency

ratio and uric acid excretion were similar indicators of protein

quality. This method is of value for studies in chicks where

separation of urinary and fecal nitrogen determination is extremely


A recent technique to more adequately evaluate the quality of

soybean products has been conducted with pigs having a cannula

inserted at the end of the small intestine (Vandergrift et al., 1983;

Rudolph et al.r 1983; Ozimek and Sauerr 1985). Therefore, a

measurement of the actual uptake of nutrients can be determined.

Vandergrift et al. (1983) presented data showing that a larger

quantity of nutrients from raw soyflakes is digested in the cecum and

large intestine by the microflora compared to adequately heated

soyflakes. The range In differences In the digestibilities of

nitrogen and amino acids was 14.3 to 50.67. for raw soyflakes and .5 to

20.6% for soyflakes heated for 25 minutes at a maximum temperature of

100 C. Van Weerden et al. (1985) compared differences between 11eal

and fecal digestibilities of normal toasted, slightly undertoasted and

slightly overtoasted SBM fed to growing pigs. No differences were

observed in fecal digestibilities among the SBM'sr whereas the 11eal

digestibilities of organic matter, crude proteinr lysine, sulfur amino

acids, threonine and tryptophan were reduced in the under- and

over-toasted SBM's. Ileal digestibility of the carbohydrate In the

normal toasted meal was increased 35 to 50% compared to the under- and

over-toasted meals.

Variation Aongcn Soybean Yor~ietifs

One factor contributing to the variability in SBM can be

associated with different soybean varieties. Significant variation in

both oil and protein contents of soybeans have been reported to be

related to both variety and location grown (Cavinessr 1973; 01oghobo

and Fetugar 1984; Gandhi et al.r 1985). Other researchers have

confirmed the large variation in protein and oil contents due to

variety and found that the two components were negatively correlated

(Hymowitz and Collins, 1974; Krivoruchco et al., 1979; Hartwigr 1979;

Hafezr 1983). Also, the protein content was negatively correlated

with the total sugar (Hymowitz and Collins, 1974) and the sucrose

(Mwandemeler 1985) contents of soybeans but positively correlated with

the stachyose content (Hymowitz and Collins, 1974). In addition,

changes in the protein content of soybeans were not correlated with

the pentosans, crude fiber and ash contents and soybean seed size

(Krober and Cartter, 1962).

Furthermore, the protein content of soybeans was found to be not

correlated with their methionine (Krober and Cartter, 1966) or the

non-protein nitrogen (NPN) content (Becker et al., 1940). Krober and

Gibbons (1962) confirmed the absence of a correlation between NPN and

the protein content and also reported that NPN was not affected by

variety or location grown but appeared to be influenced by weather

conditions. However, Becker et al. (1940) indicated a wide range (2.9

to 7.8%) existed in the NPN content of the 12 varieties of soybeans


Several researchers have reported that the methionine content of

soybeans varies having the following ranges; 1.0 to 1.7gm/16gm N

(Krober and Cartter, 1966), 1.3 to 1.5gm/16gm N (Aldersr 1949)r and

1.3 to 1.7gm/16gm N (Krober, 1956). The methionine content was

influenced by location grown and planting season (Krober, 1956).

In contrast to being negatively correlated with the protein

content of soybeans, the oil content was positively correlated with

the total sugar, sucrose, and raffinose contents of soybeans (Hymowitz

and Collinsr 1974). The fatty acid composition was relatively

constant with different varieties grown at different locations

(Collins and Sedgwick, 1959) and was influenced by temperature during

the growing season (Howell and Collins, 1957). However the linolenic

and linoleic acid contents of soybeans were negatively correlated

(Howell and Collins, 1957).

A large variation in the raffinose and stachyose contents of

soybeans exists in different varieties (Hymowitz and Collins, 1974).

Likewise, Kennedy et al. (1985) compared five different soybean


varieties and noted varation in sucrose (4.0 to 7.7%), stachyose (3.0

to 4.1%) and raffinose (.7 to .9%) contents. One variety had the

lowest content of each of these oligosaccharides (total of all three

varied between 7.6 to 12.7%). Mwandemele (1985) reported that the

stachyose and raffinose contents of soybeans were positively

correlated. Also the stachyose and raffinose contents were not

correlated with the protein and oil content or yields but the sucrose

content was positively correlated with yield.

Other components of soybeans have been shown to vary, such as

glycinins (31.4 to 38.3% of the total protein; Hughes and Murphys

1983), phytin phosphorus content (.51 to .73%; Averill and King,

1926), available carbohydrates and ash (010ghobo and Fetugas 1984),

and percentage husk and cotyledons (Gandhi et al., 1984). The

minerals with the most varfabilfties in soybeans were frons zinc and

manganese and the lowest variabilities were in calcium and phosphorus

(01oghobo and Fetuga, 1984).

Birk and Waldman (1965) reported that no quantitative differences

were found in UA of three soybean varieties. However, Smith et al.

(1956) found that the UA varied according to variety and location

grown. Portions of the seed also vary in UA. The hull has lower UA

than the hypocotyl, and the colyledons have twofold the UA of the

hypocoty1. Myer and Froseth (1983) compared extruded mixtures of

beans (Phaseolus uulceris) and raw soybeans. The raw soybeans in one

experiment had a UA of 1.87 whereas, the raw soybeans In the second

experiment contained a UA of 1.15. The soybeans with the higher UA

contained 68 units more trypsin inhibitor per mg protein than the


soybeans having the lower UA. A reason for the difference between the

raw soybeans was not Indicated by the researchers.

The amount of trypsin inhibitor activity also varies between

soybean varieties (Kakade et al.r 1969; Hafez 1983). Likewise, Gandhi

et al. (1984) compared the trypsin inhibitor content of eleven

different varieties of soybeans and reported a range of 8.1 to 38.5

trypsin inhibitor units per mg sample. Two varieties having a black

seed coat contained the highest trypsin inhibitor and UA values.

Soybeans have also been found which do not contain the Kunitz trypsin

inhibitor (Leiner and Tomlinson, 1981), with the total trypsin and

chymotrypsin inhibitor activities being approximately one-half and

three-fourths of that of commercial raw soyflour. Soybeans also

contain nonprotein trypsin inhibitors and Hafez and Mohamed (1983)

reported that a large difference occurred due to variety.

Kakade et al. (1972) found a seven-fold variation in lectin

content within 108 different soybean varieties evaluated. Subsequent

research by Pull et al. (1978) noted the absence of lectins in five

varieties of soybeans they assayed. In additions variations in the

isoflavones (compounds having estrogenic properties) and 11poxygenase

(an enzyme that catalyzes the oxidation of lipids) contents have been

reported due to variety and location grown (Eldridge and Kwolaks


Data on differences in feeding value between varieties of

soybeans are limited. Yen et al. (1974) reported depressed

performance of growing pigs fed different varieties of raw soybeans

compared to SBM. However, performance of pigs fed the three varieties

of raw soybeans did not differ. Bajjalieh et al. (1980) fed chicks

two raw soybean varieties (one commercial and one which lacked the

Kunitz inhibitor) compared the responses to chicks fed heated

soybeans. The soybean variety lacking the Kunitz inhibitor permitted

a greater growth response and smaller pancreatic weights of chicks

than the commercial variety of soybeans, although this improvement was

still below the response to chicks fed heated soybeans. Han and

Parsons (1986) confirmed the greater nutritional value of raw soybeans

containing a low trypsin inhibitor (Kunitz) content compared to a raw

commerical variety. These researchers reported that the

availabilities of lysine and methionine were similar in unheated

soybeans having low trypsin inhibitor content and heated debulled

soybeans, and both had higher lysine and methionine availabilities

compared to the raw commerical variety. Yen et al. (1971) fed a

unheated variant soybean which had a greater nutritional value than

other varieties of soybeans they studied and reported that the

performance of rats remained adversely affected compared with rats fed

heated soybeans.

Effect of Storage on Soybeans and Soybean Products

An additional factor which may alter the nutritional value of

soybeans is the method and length of storage. Soybeans have been

stored at 13 to 14% moisture and at 4 to 5 C for six years without

deterioration (Smith and Circle, 1972). Mitchell and Beadles (1949)

reported deterioration in the digestibility and biological value of

soybeans which had been stored for almost three years at 25.5 C.

However pretreatment heating largely prevented this deterioration.

the loss was attributed to an enzymatic factor. Nakayama and Kito

(1981) stored soybeans at 13% moisture for six months and found 45% of

the total phospholipids had decomposed, 72% of the total phospholipids

originally extracted with the oil were lost and phosphatidylcholine

and phosphatidylethanolamine were significantly decreased.

Phosphatidic acid and 1ysophosphatidylcholine concentrations

increased. In additions the oligosaccharide content of soybeans was

decreased in storage but this reduction can be stopped by heating the

soybeans prior to storage which indicates changes due to enzymiatic

processes (Mwandemeles 1985). Saio et al. (1980) reported that bean

color darkened and acid values of extracted crude oil and acidity of

beans Increased as deterioration progressed. Storage temperature and

relative humidity are both related to overall changes during storage

but relative humidity seems to be more important. Nitrogen solubility

index decreased rapidly with Migh temperature and relative humidity.

Yao et al. (1983) reported that storage for 6 months did not affect

trypsin inhibitor activity.

Whole beans were mrore resistant to deterioration during storage

than soybean meals (Saio et al.r 1982). Also full-fat soybean meals

deteriorated more rapidly then defatted meals.

Effect of Ma-iu~rity on Complosition of Soybeans

Maturity of the soybean seed can also influence soybean

composition. Febr et al. (1971) reported that fatty acids were

synthesized by the soybean plant at different rates during pod

filling. Linolenic acid percentage in the seeds decreased rapidly

during the first 30 days and then remained constant throughout the


period of oil deposition. Immature seeds contained only 3% of the UA

(Birk and Waldman, 1965) and had more available zinc (Welch and House,

1983) than mature seeds. Lipoxygenase activity, phytate content and

the ratio of 7S and 11S protein were lower in immature seeds compared

to mature seeds (Yao et al., 1983). Fat content was found to be

higher in earlier maturing varieties than later maturing varieties

(Krivoruchco et al., 1979). Mwandemele (1985) reported positive

correlation between concentrations of various oligosaccharides and day

to maturity. Also, maturation was not related to the oil and protein

contents (Yao et al., 1983).

Research data on the response of soybean maturity on the trypsin

inhibitor activity have been Inconsistent. Collins and Sanders (1976)

reported that the amount of trypsin inhibitor activity increased as

the soybeans matured. However, Yao et al. (1983) found that the

trypsin inhibitor activity and soybean maturity were not related.

Influence of Growing ~Conditions ~on Soybean Complositi'on

As previously stated location where soybeans are grown can

affect their nutrient composition. The use of fertilizer and the

various components of fertilizer can influence the resulting nutrient

content of the harvested soybeans. Gaydou and Arrivets (1983) found

that fertilizer containing a high level of phosphorus increased both

the oil and protein contents of soybeans, while fertilizer comprised

of a high level of potassium increased the oil content but decreased

content of protein. Dolomite fertilizer which is high in both calcium

and magnesium increased the oil content but did not affect protein

content of soybeans. Nitrogen fertilization did not Influence either


oil or protein content of the soybeans. In contrast, Krober and

Cartter (1966) had reported that soils low In nitrogen produced

soybeans with reduced protein content.

Wolf et al. (1982) grew soybeans at temperatures during the

day:night of 18:13, 24:19, 27:22r 30:25 and 33:28 C and obtained a

decrease in linolenic and linoleic acids sucrose (greatly) and

stachyose (slightly) with increased temperature during the growing

period. Increased growing temperature also increased oleic acid, oil

and protein contents. Palmitic and stearic acid, glucose fructose

and raffinose were not affected by different growing temperatures.

Amino acids were generally stable except higher temperature increased

methionine content. Howell and Collins (1957) had previously reported

that the temperature in which the soybeans were grown affected the

linolenic and linoleic acid contents of the soybean oil more than the

variety. Krober and Collins (1948) reported that weather-damaged

soybeans often had higher NPN content than undamaged soybeans.



Whole "full fat" soybeans are an excellent source of protein and

energy for swine diets. Also, soybeans contain other important

components, such as trypsin inhibitors (Liener, 1981) which are

considered the principle heat labile anti-nutritional factors and the

enzyme urease which is monitored in soybean meal processing plants as

an indication of optimum heat processing (Caskey and Knapp, 1944).

Subjecting soybeans to heat processing concurrently improves their

nutritional value and reduces the activities of both trypsin inhibitor

and urease.

Previous research has been inconclusive on the feeding value of

roasted soybeans as a protein supplement for weanling swine (Noland et

al.r 1976; Rust et al., 1972; Crenshaw and Danielson, 1984a). This

could be due in part to the temperature and duration of heat

processing and moisture content of soybeans prior to heating; all are

important factors affecting the nutritional quality of soybeans

(McNaughton and Reece, 1980). Likewise previous research at the

University of Florida (Campbell et al., 1984) indicated that roasting

soybeans at 125 C permitted superior growth of finishing pigs compared

to pigs fed soybeans roasted at 110 C. Therefore, the objectives of

this study were to determine if increasing the moisture content (10%

added water) of Bragg soybeans prior to roasting at 110 or 125 C would

enhance the inactivation of trypsin inhibitors, decrease grease

activity and increase performance when included in diets of weanling


MFaterials and Methods

One-hundred eight crossbred pigs with an average initial weight

of 5 kg were assigned to pens containing six pigs each by initial

weight, sex and litter origin. Each treatment consisted of three

replicate pens. The six dietary treatments (Table 1) were corn-based

diets containing the following protein supplements: (1) soybean meal;

(2) raw soybeans; (3) soybeans roasted at 110 C with no water added

prior to roasting; (4) soybeans roasted at 110 C with 10% water added;

(5) soybeans roasted at 125 C with no water added and (6) soybeans

roasted at 125 C with 10% water added. All diets were formulated to

be isonitrogenous and isocaloric. The whole soybeans used in the

treatments requiring added moisture were placed into plastic

containers prior to roasting and 10% (w/w) water was added. These

soybeans were then mixed periodically with the water by dumping the

soybeans into other containers and then stored overnight. During the

following morning, the soybeans were roasted by passing them through a

Roast-A-Tron 1 gas-fired roaster. Afterwards, the soybeans were

ground in a hammermill before inclusion into their respective diets.

Ether extract of the raw soybeans was determined according to the

procedure outlined by the A0AC (1980). Samples of raw soybeans were

Mix-Mill, Inc., Blufftonr IN.


Ingredients Basal Soybeans
(Diet 1) (Diets 2-6)

Ground yellow corn 68.40 56.90
Soybean meal (48%) 25.40---
Whole soybeans ----- 39.90
Corn oil 3.00---
Dynafos 1.70 1.70
Limestone .80 .80
Salt .25 .25
Trace minerals (CCC a.10 .10
Vitamin prgmix (UF) .10 .10
Antibiotic ,25 ,25
100.00 100.00
Calculated analyses:
Protein, % 18.00 18.00
Metabolizable energy
Kcal/kg 3310 3304

SProvided by Calcium Carbonate Company, Quincy, IL. Contained 200
mg zinc, 100 mg iron, 55 mg manganese 11 mg copper, 1.5 mg
iodine, 1.0 mg cobalt and 20 mg calcium per kg complete diet.

SSupplied 13.2 mg riboflavin; 44.0 mg niacin; 26.4 mg pantothenic
acid; 176.0 mg choline chloride; 22.0 mg vitamin B12; 5,500 IU
vitamin A; 880 ICU vitamin D3 and 22 IU vitamin E per kg of
complete diet.

SProvided 44 m9 chlortetracycline, 44 mg sulfamethazine and 22 mg
penicillin per kg of complete diet.

obtained for nitrogen analysis prior to roasting. These raw soybeans

were digested using a modification of the aluminum block digestion

procedure of Gallaher et al. (1975). Ammonia in the digestate was

measured by semiautomated colorimetry (Hambleton, 1977).

Additionally, trypsin inhibitor activity (TIs Hamerstrand et al.,

1981)r urease activity (UA, Caskey and Knapps 1944) and dry matter

content (A0AC, 1980) were determined on both the raw and roasted

soybeans and SBM.

All pigs were housed during the 35-day trial in an enclosed

nursery building equipped with elevated pens having expanded metal

floors and wire mesh sides. Feed and water were supplied ad libitum.

Pigs were individually weighed and feed consumption for each pen was

determined bi-weekly. The data were subjected to analysis of variance

for a randomized complete block design, with blocks representing

replications. Then the basal diet (Treatment 1) and the raw soybean

diet (Treatment 2) were omitted thus enabling Treatments 3 through 6

to be analyzed as a 2 x 2 factorial arrangement of treatments (added

moisture x roasting temperature). The GLM procedure developed by SAS

(1979) was used for compilation of analyses of variance. Duncans'

multiple range test was used to interpret significant differences

(SAS, 1979).

Results and Discussi~o

Average daily gain of pigs fed the soybean meal (SBM) basal diet

was superior (P<.05) to that of the other dietary treatment groups

(Table 2). Comparing the growth rate among pigs fed the diets

containing full-fat soybeans, pigs fed diets supplemented with raw

O ~110004N



~ mm~~rm 9
v, oorlmm *, o
NOc *** .,


PH~aa Ho ***

2> Lno m cc E
O, v
DU r
C0 Om TT

OL C\O ***.. O

gv, a

q-UO C

C30 ***.n

v) rO O,

Z ~O

UZ Oimn >

z~~*r LnD r $

<3 09 >>>> a,
H< << << <

(unheated) soybeans or soybeans roasted at 110 C with no water added

prior to roasting were adversely (P<.05) affected. Weight gains of

pigs fed the diet containing soybeans which had 10% water added prior

to roasting at 110 C were augmented (P<.05) compared to pigs fed a

diet with soybeans roasted at 110 C without added water and equal

(P>.05) to pigs fed soybeans roasted at 125 C. Likewise addition of

10% water prior to roasting soybeans at 125 C permitted an additional

Increase (P<.05) in pig growth. The decreased performance of pigs fed

raw soybeans was expected and these data agree with previous research

evaluating raw soybeans fed to starting growing and finishing pigs

(Jimenez et al., 1963; Combs et al.r 1967; Hanke et al.r 1972; Yen et

al., 1974). Also, the depressed growth of weanling pigs fed roasted

soybeans is consistent with the findings of Rust et al. (1972) and

Crenshaw and Danielson (1985a) but is in contrast with the work of

Noland et al. (1976).

Feed consumption by pigs fed the diets containing soybean meal or

soybeans having 10% water added prior to being roasted at 125 C were

equal (P>.05). However feed consumption by pigs fed these two diets

was greater (P<.05) when compared to the feed consumed by pigs fed the

other treatment groups. Roasting soybeans at 110 C with or without

added water and at 125 C without inclusion of water did not (P>.05)

increase the quantity consumed when compared to pigs fed the raw

soybean diet. Similarly, feed-to-gain ratios of pigs fed the diets

containing soybean meal or soybeans roasted at 125 C with or without

additional water exceeded (P<.05) the efficiency of pigs fed diets

containing raw soybeans or soybeans roasted at 110 C without water


being added. Rust et al. (1972) reported that the growth of pigs fed

roasted soybeans were depressed but the efficiency of growth was

similar to pigs fed soybean meal. Noland et al. (1976) also reported

similar feed efficiency when weanling pigs were fed either roasted

soybeans or SBM. In contrast, Crenshaw and Danielson (1985a) obtained

depressed feed efficiency when weanling pigs were fed diets containing

roasted soybeans compared to SBM.

The main effects of roasting temperature and water addition are

shown in Table 3. Pigs fed diets containing Bragg soybeans roasted at

125 C consumed more (P<.05) feeds grew faster (P<.05) and were more

(P<.10) efficient in utilization of feed than pigs fed diets

containing soybeans roasted at 110 C. Similarly, adding 10% water

prior to roasting the soybeans improved (P<.05) feed intake and

average daily growth of pigs compared to pigs fed roasted soybeans

without added water prior to roasting. Although the feed-to-gain

ratios between treatments containing 0 or 10% added water had a large

numerical difference, these ratios were not different (P>.05).

The urease (UA) and trypsin inhibitor (TI) activities indices

used to assess soybean quality, of the different soybean products are

presented in Table 4. None of the heat treatments was effective in

lowering the UA of the full-fat soybeans to less than .2. Soybean

meal with a UA range between .05 to .2 is considered to have received

optimum heat processing by commercial soybean processing plants

(Smith, 1977). Roasting soybeans at 110 C without added water reduced

the TI and UA by 25.9 and 7.6%,s respectively. Adding 10% water prior

to roasting at 110 C or roasting at 125 C without added water resulted

Temperature Added Moisture SE#a
110 C 125 C 0% 10%

Avg. initial weight, kg 5.02 5.02 5.02 5.02 ---
Avg. final weight, kg 8.69 12.41 9.35 11.78 -
Avg. daily gain, kg .10b21 .12b 1c 0
Avg. daily feed, kg .37b .47c .35b .48c .03
Avg. feed/gain 4.16d 2.27e 3.78 2.65 .79




SStandard error of the mean.


Column means within main effects with different superscripts are
different (P<.05).

Column means within main effects
different (P<.10).

with different superscripts are

Product Trypsin Inhibitor Urease Activity
mg/gm defatted sample pH change

Unheated soybeans 53.95a 1.97
Roasted 110 C; 0% water 40.00 1.82
Roasted 110 C; 10% water 24.00 1.30
Roasted 125 C; 0% water 22.88 0.55
Roasted 125 C; 10% water 9.18 0.25
Soybean meal (48%) 3.11 0.10



Larger values are indicative of less heat processing.

In similar reductions in TI. The Inclusion of 10% water prior to

roasting at 125 C resulted in the largest reductions in TI and UA

(83.0 and 87.3%, respectively) yet the product still contained between

two- and three-fold the TI and UA of soybean meal. Pig performance

data reflect the difference in heat processing and indicate that

soybean products having a UA of .25 or higher are not adequate protein

supplements for weanling pigs. These results concur with the work of

Albrecht et al. (1966), McNaughton and Reece (1980), and Waldroup

(1985) which suggest that increasing the moisture content of soybeans

prior to heating increased the inactivation of TI and UA. These data

also agree with the findings of Campbell et al. (1984) in that

soybeans roasted at 125 C had a higher nutritional value when compared

to soybeans roasted at 110 C. In addition, the depressed performance

of weanling pigs fed soybeans having a UA higher than .2 is in

agreement with the optimum range of UA used by the American Feed

Manufacturer Association to indicate soybean products that have

received adequate heat processing (Smithr 1977).

The Bragg variety of soybeans roasted in this study had remained

in the field for a longer period of time (1-2 months) than normal

before being harvested. The moisture content of these soybeans was

10.33% and with the addition of 10% water was analyzed to be 20.45%

prior to roasting. Roasting the soybeans at 110 C with and without

added water reduced the moisture content to 13.1 and 9.0%,

respectively. Further evaluation of the soybeans noted lower protein

(32.7%) and higher fat (23.9%) contents than that published by the NRC

(1979; 37.0 and 18.0%, respectively). However, the actual values and

not the NRC (1979) values were used to calculate the experimental

diets. Therefore, the protein content of the experimental diets was




Full-fat soybeans can be utilized as an excellent source of both

energy and protein in swine diets. However, soybeans also contain

several anti-nutritional factors which require inactivation before

yielding a product having the highest nutritional value. Trypsin

inhibitors are considered the most important anti-nutritional factor

in soybeans and subjecting soybeans to heat processing has been used

satisfactorily to reduce the trypsin inhibitor activity (Liener,

1981). Heat processing has been shown to be required when soybeans

are added to diets fed to young pigs and becomes less critical when

soybeans are fed to finishing pigs (Combs and Wallace, 1969).

Although previous research has been inconclusive on the feeding value

of raw soybeans for growing-finishing swine (Combs and Wallacer 1969;

Jensen et al., 1970; Crenshaw and Danielson, 1985c; and Pontif et al.,

1986), raw soybeans have been found to be adequate protein supplements

for gilts and sows (Crenshaw and Danielson, 1984d). Another technique

that can be used to reduce the trypsin inhibitor activity of soybeans

is germination (Bates et al., 1977; Suberbie et al., 1981). Collins

and Sanders (1976) reported that soybeans which had been rinsed twice

daily and allowed to germinate for three days lost up to 13.27. of

their trypsin inhibitor activity.

Numerous new varieties of soybeans are released annually which

have improved agronomic advantages over existing varieties. However

the nutritional value of these new soybean varieties has often been a

minor selection criterion. Significant variations in both oil and

protein contents and urease activity (UA) of soybeans have been

reported to be related to both variety and location grown (Smith et

al., 1956; Caviness, 1973; 01oghobo and Fetugar 1984; Gandhi et al.r

1984). Likewise the trypsin inhibitor activity (TI) also varies

among different soybean varieties (Kakade et al., 1969; Hafez, 1983).

Besides the varietal differences In the constituents of soybeans,

relationships among the various components do exist. Several

researchers (Hymowitz and Collinsr 1974; Krivoruchco et al.r 1979;

Hartwig, 1979; Hafezr 1983) have reported that the protein and fat

components of soybeans were negatively correlated.

These varietal differences would indicate that the heat

processing required to destroy the heat labile anti-nutritional

factors could vary between the different varieties of soybeans. Data

on differences in feeding value among different varieties of soybeans

are limited. Yen et al. (1974) reported no difference in performance

of growing pigs fed either of three different varieties of raw

soybeans. However, the growth of pigs fed a commercial soybean meal

was superior to the pigs fed the diets containing different varieties

of raw soybeans.


The objectives of these studies were: (1) to determine the

relationships of various nutritional components and evaluate their

variability in different varieties of soybeans (2) to determine

variability of similar varieties of soybeans grown at different

locations, (3) to evaluate the trypsin inhibitor and urease activities

of soybeans which were germinated or heated in an autoclave at 110 C

for 7.5 or 15.0 minutes and (4) to evaluate the performance of

growing-finishing pigs fed diets containing either Bragg (unheated and

heated at 110 or 125 C) or Davis (unheated and heated at 110 C)

variety of soybeans.

Materials and Mlethods

Trial 1

Fifteen commerical varieties and three experimental strains of

soybeans were obtained from the University of Florida Agronomy

Department and the protein and fat contents, urease activity (UA) and

trypsin inhibitor activity (TI) were determined using procedures

described below. The effect of heat processing was assessed by

placing approximately 25 g of whole soybeans in aluminum drying pans

having a 6 cm diameter. Three replicate drying pans per soybean

variety were randomly placed on a 30 X 60 cm stainless steel tray and

autoclaved for 7.5 or 15.0 minutes at 110 C and 422 gm/cm2

pressure. The soybeans from the three replicate drying pans were

combined after cooling and ground prior to determination of the TI and

UA. The effect of germination on TI and UA was determined by

germinating the different varieties of soybeans for 7 days and then

only the soybeans which had sprouted were removed, dried and ground.

Trial 2

Four commercial varieties and thirty-four experimental strains of

soybeans were obtained from the same source as soybeans used in Trial

1 and analyzed for protein and fat contents and TI and UA. These

soybeans were grown during the growing season subsequent to those

soybeans used in Trial 1. There is a large variation in the size of

soybean seeds being marketed and the correlation of this variable

with the measured parameters was determined. Soybean size was

quantitated with micrometer calipers from measurements taken on the

long axis directly between the cotyledons of the soybeans.

Trial 3

To evaluate the effect of the varietal differences in soybeans on

their feeding value, the following trial was conducted. One-hundred

and eight crossbred pigs with an average weight of 40 kg were allotted

on the basis of initial weight, sex and litter origin to six dietary

treatment groups. Six pigs were assigned to each pen with three

replicate pens per treatment. The dietary treatments (Table 5)

consisted of the following protein supplements: (1) commercial

soybean meal; (2) unheated Bragg soybeans; (3) Bragg soybeans roasted

at 110 C; (4) Bragg soybeans roasted at 125 C; (5) unheated Davis

soybeans and (6) Davis soybeans roasted at 110 C. Roasting consisted

of passing whole soybeans through a Roast-A-Tron1 gas-fired

roaster. Afterwards, the roasted and unheated soybeans were ground in

a hammermill before inclusion into their respective diets. The two

varieties of soybeans were obtained locally from different suppliers.

Mix-Mills Inc., Bluffton, IN.

Grower Diets Finisher Diets
Ingredients Control Soybeans Control Soybeans

Ground yellow corn 76.80 70.30 82.65 77.35
Soybean meal 20.00 ----- 14.40 --
Whole soybeans ----- 26.50 ----- 19.70
Dicalcium phosphate 1.70 1.70 1.70 1.70
Limestone 0.80 0.80 0.80 0.80
lodized salt 0.25 0.25 0.25 0.25
Trace minerals (CCC a 0.10 0.10 0.10 0.10
Vitamin premix (UF) 0.10 0.10 0.10 0.10
Antibiotic premix' 0,25 0.25 ----
100.00 100.00 100.00 100.00

Calculated analyses:
Protein, % 16.00 16.00 14.00 14.00
Metabolizable energy
Kcal/kg 3250 3250 3250 3250

SProvided by Calcium Carbonate Company, Quincy, IL. Contained 200
mg zincs 100 mg iron, 55 mg manganese 11 mg copper, 1.5 mg
iodine 1.0 mg cobalt and 20 mg calcium per kg complete diet.

SSupplied 13.2 mg riboflavin; 44.0 mg niacin; 26.4 mg pantothenic
acid, 176 mg choline chloride; 22 mg vitamin B12; 5,500 IU
vitamin A; 880 ICU vitamin D3 and 22 IU vitamin E per kg of
complete diet.

c Provided 110 mg chlortetracycline, 110 mg sulfamethazine and 55 mg
penicillin per gm complete grower diet.


The Bragg soybeans were harvested after remaining in the field

approximately one to two months longer than normal. The Davis

soybeans were harvested after a normal period of time. The grower

diets were fed for 28 days and then the pigs were fed their respective

finishing diet.

All pigs were housed in a semi-enclosed concrete barn with

partially slotted floors. Feed and water were supplied ad libitum.

Pig weights and feed consumption were determined bi-weekly.

In all trials the protein and fat contents (A0ACI 1980) and UA

(Caskey and Knapp, 1944) of the soybeans were conducted on air dry

samples. Samples for nitrogen analysis were digested using a

modification of the aluminum block digestion procedure of Gallaher et

al. (1975). Ammonia in the digestate was determined by

semi-automatical colorimetry (Hambleton, 1977). Trypsin inhibitor

activity (Hamerstrand et al., 1981) was determined on defatted soybean

samples. Urease activity and TI of germinated seeds were determined

on samples which had both the fat and moisture removed. All analyses

were conducted in duplicate. In Trials 1 and 2r the individual

variety means were used to calculate the gross correlations among the

various components. Correlations were considered significant if the

probability level was .05 or less. In Trial 3r the growth data were

analyzed by least squares analysis of variance with initial weight as

a covariate. Feed intake and feed efficiency data were subjected to

analysis of variance for a randomized complete block design. Duncan's

multiple range test was used to determine treatment differences. All

statistical analyses were conducted following the procedures developed

by SAS (1979).

0~~00600 00 mH
* * * * 1 * *


mr h.

Nill MM
0il OOC
*Ill ** *~

Nomb 0000c~v

100 D

a cr










Mme ~0~--r
Uer0\m c0


0 zJ

NcO4 r


.. .C

Newtomoma Omm
~~~OWNNM 000

o ;


m N

rl C\

O 04~

**.... Mr



(3\lTCN \Dumo
0000 00 0

000 ~
0 0 ~h


000 ~

0000 O

\Ol \ L
400 ~rr


- Mrr



.. .


** .~


C0 bf
~me -

04 <<~bmu
ururc~nmN m





Q) .




* **3~
o melC

oves l


a ~I~rc
* * . ~~
mo0m30 1

... n
amm \c

cHN r
. M
Mma 3



L L r-*-X
mmy m L w

* * .~-




N 9:
H I cq H
O C,\ cO 0
vMc rC



Results and Discussion

Trial 1

A summary of the proteins fat, and trypsin inhibitor (TI) and

urease activities (UA) of the soybeans is shown in Table 6. The

percentage of protein and fat in the various soybean varieties studied

ranged between 32.74 to 41.50% and 18.80 to 21.13% with an average of

38.17 and 20.16%, respectively. The UA, used in commercial soybean

meal processing plants to indicate adequacy of heat processing,

averaged 1.43 in the unheated soybeans. After heat processing the

soybeans at 110 C for 7.5 and 15.0 minutes, the average UA was reduced

to .14 and .03, respectively. The American Feed Manufacturers

Association considers that soybeans have received adequate heat

processing if the UA is between .05 and .2 (Smith, 1977). Therefore,

heating the different varieties of soybeans for 7.5 minutes at 110 C

was adequate for all varieties except for two of the experimental

strains (F80-6950 and F81-9202). A wide variation among the different

varieties of soybeans (approximately 2.5 fold) was observed in the TI

values ranging from 25.00 to 61.84 mg/gm of sample with a coefficient

of variation of 23.75%. Autoclaving the soybeans for 7.5 or 15.0

minutes reduced the average TI activity to 10.56 and 3.87 mg/gm of

defatted soybean sample, respectively. These data are in agreement

with previous research reporting a wide range in oil and protein

concentrations (Caviness, 1973; 010ghobo and Fetuga, 1984; Ghandhi et

al., 1984), TI (Kakade et al.s1968; Hafez, 1983; Ghandi et al., 1984)

and UA (Smith et al., 1956). The wide range in UA is in contrast with


Waldman (1965). However these researchers only compared a limited

number (3) of soybean varieties.

Germinating the seeds for seven days reduced the average TI

activity to 30.70 mg/gm of defatted soybean sample. The largest

reductions in TI activity were with the soybean varieties initially

having the highest TI activity.

Two varieties (Bragg and Kirby) were grown in different areas

which permitted an evaluation of the effects of location on the

compositional variablity of soybeans. Variation in most of the

constituents was observed in the same variety grown in different

locations. Several factors other than variety such as growing

temperature (Wolf et al., 1982), type of fertilizer applied to the

soil (Gaydor and Arrivets, 1983) and plant maturity (Febr et al.,

1950; Birk and Waldman, 1965; Yao et al., 1983) have been reported to

alter soybean composition.

The experimental strains of soybean used in this study were

classified as being of the vegetable type of soybeans. This type of

soybean is noted for having a bland flavor and large seeds (Hinsonr

1986). All soybeans had a yellow seed coat except for the F80-6717

strain which had a black seed coat. Ghandi et al. (1984) compared the

TI of eleven different varieties of soybeans of which only two

varieties had black seed coats. The soybeans with the black seed

coats had the highest TI contents. The F80-6717 strain (black seed

coat) studied in this trial had a increased TI activity compared to

the average of all varieties but was not the strain with the highest

TI activity. The percentage reduction in TI and UA in the different

varieties of soybeans by heat processing or germination are presented

in Table 7. Heating the soybeans at 110 C for 7.5 or 15.0 minutes

reduced their TI activity by 71.36 and 89.34%r respectively. Whereas,

the UA activity was reduced 90.62 and 97.73% by heat processing the

soybeans for 7.5 and 15.0 minutes, respectively. The increased

inactivation of UA compared to TI concur with data of Borchers et al.

(1947) and McNaughton and Reece (1980). Germinating the soybean seeds

for seven days only lowered the TI content by 15.44%. Previous

research has been inconsistent on the reduction of TI by germinating

soybeans. Desikachar and De (1947) and Collins and Sanders (1978)

reported little change in TI by germinating soybeans. However, Bates

(1977) found a 33% reduction in TI by sprouting soybeans for 4 days.

Similarly, Collins and Sanders (1976) obtained up to 13.2% reduction

in TI after germinating seeds for only 3 days.

Gross correlations associating protein and fat with UA and TI

activities are shown in Table 8. The initial TI activity of the

soybeans (unheated) was positively correlated with the percent

inactivation of TI by heating the soybeans for 7.5 (P<.046) or 15.0

minutes (P<.001). The UA prior to heating the soybeans was also

positively correlated with the percentage TI inactivation by heating

the soybeans for 15.0 minutes (P<.027). However, the initial UA

(prior to heating the soybeans) was not correlated (P>.05) with the

percent reduction of their UA by heating for either 7.5 or 15.0

minutes. Furthermore the protein and fat concentrations of the

soybeans were not correlated (P>.05) with the percent reduction of TI

or UA.


~Trypsin inhibitor ~Urease activity Germination
Heat Processinq, minutes at 110 C
Yarietya (7,5) (15,0) (7,5) (15.0)
Commercial varieties

Asgrow 7372 80.9 90.1 93.4 98.6 21.0
Braxton 73.6 88.9 93.4 99.3 12.0
Coker 237 67.7 90.7 85.1 100.0 10.6
Coker 368 70.2 87.5 96.6 100.0 15.8
Coker 488 65.0 86.1 90.9 98.5 15.6
Cobb 71.0 88.0 88.9 100.0 12.8
Centennial 71.9 89.8 92.5 98.4 7.6
Foster 75.8 90.0 94.8 97.8 32.0
GaSoy-17 73.8 89.2 93.3 96.9 15.7
Hutton 70.8 90.6 91.7 98.3--
RA 604 68.8 88.6 96.8 98.7 5.0
RA 801 73.3 87.8 89.5 97.7 12.4
Wright 60.7 84.3 92.0 94.0 9.0
Grown at different 1cca~tions

Bragg No. 1 64.8 86.9 90.8 93.8 17.1
Bragg No. 2 78.1 94.2 92.3 97.2 -
Kirby No. 1 67.8 92.5 90.2 97.6--
Kirby No. 2 67.0 88.5 91.9 98.5 -
Experismental strains

F80-6717 79.1 89.4 90.1 96.8 13.4
F80-6950 74.2 92.1 79.0 97.2 27.0
F80-9202 72.7 92.0 79.7 95.3 20.0

Mean 71.4 89.3 90.6 97.7 15.4

SRA-Ring around; F-Florida.

Trypsin inhibitora

Heat Processing, minutes

, ,,, ,
,,- 5,

SUrease activity (UAa npH); Trypsin inhibitor activity (TI, mg/gm).


Urease activity

nietorP % 0 0

51; 0

0 0

1C 0

Fat, %


.475 .295
P<.034 P<.207


.438 .135 -.181
P<.054 P<.570 P<.444

-.328 -.132 -.176
P<.157 P<.578 P<.458

.446 .186 .186
P<.0003 P<.049 P<.432

Proteins %

UAs 0.0 min

UA, 7.5 min

UAs 15.0 min

TI, 0.0 min

TIs 7.5 min

-.441 -.264 -.295
P<.051 P<.261 P<.160

.461 .724
P<.006 P<.041

.440 .615
P<.0522 P<.004

.547 .143
P<.013 P<.546

.738 .669
P<.0002 P<.001

.667 .320
P<.001 P<.169



The protein concentration was negatively correlated with the fat

concentration (P<.0001) and UA of the unheated soybeans (P<.052) but

was not correlated (P>.05) with the UA after heat processing or the TI

prior or after heat processing. The negative correlation in the

protein and fat concentrations of soybeans has previously been well

documented (Hymowitz and Collins, 1974; Krivoruchco et al., 1979;

Hartwigs 1979; Hafez, 1983). The fat concentration was positively

correlated with the UA (P<.034) and TI (P<.054) activity prior to heat

processing; whereas after heat processing there was no correlation

(P>.05) between these components. The TI was positively correlated

(P<.0003) with the UA of the unheated soybeans. Both the TI and UA of

the unheated soybeans were positively correlated (P<.05) with the UA

of the soybeans which had received heat processing for 7.5 or 15.0

minutes and the TI activity of soybeans heated for 7.5 minutes. The

TI and UA of the unheated soybeans were not correlated (P>.05) with

the TI activity of the soybeans heated for 15.0 miinutes.

Several researchers (Caskey and Knapps 1944; Albrecht et al.,

1966; McNaughton and Reece, 1980) have reported that the UA is a good

indicator of adequately heat processed soybeans. The positive

correlation in the present study between TI and UA would confirm these

data. The UA was correlated with the TI of the unheated soybeans or

soybeans heated for 7.5 minutes but was not correlated to the TI of

soybeans heated for 15.0 miinutes. A reason for the lack of

correlation of UA with TI of soybeans heated for 15.0 minutes may be

because these soybeans would be considered overheated (UA less than

.05) by the American Feed Manufacturer Association (Smith, 1977).

Balloun et al. (1953) reported that the UA is of little value for

detecting overheated soybean meal.

Trial 2

The seed size protein, fat, UA and TI analyses are presented in

Table 9. The average size of the soybean seeds studied was 7.15 mm

with a range between 5.68 and 10.32 mm. The percentages of protein

and fat in the various soybean varieties ranged between 37.09 to

43.62% and 17.19 to 22.81% with averages of 39.39 and 20.46%,

respectively. The averages of the UA and TI activity (1.52 and 37.24

mg/gmr respectively) are similar to the values obtained in Trial 1.

However, the range in UA (.56-2.05) was greater whereas the range in

TI (23.06-60.05 mg/gm) was similar to what was observed in Trial 1.

The majority of the soybean varieties utilized in this study were

used to confirmi trends noticed in Trial 1. The Late Giant variety is

the parent stock for some vegetable types of soybeans. The F83

experimental strains of -7895, -7923r -7959 and -7962 are closely

related to the F81-9202 of Trial 1 which had an extremely high

activity of TI (61.84 mg/gm sample). Likewise these strains also

contained a high activity of TI with one strain (F83-7923) containing

60.05 mg TI/gm. The isolines are strains which have similar genetic

material except for only one trait and in mrost cases the composition

was simlilaar. In addition, there are several different trypsin

inhibitors present in soybeans. The first trypsin inhibitor isolated

from soybeans has been referred to as the Kunitz inhibitor. The two

L81 strains are from genetic lines noted for the absence of the Kunitz

inhibitor but their overall TI activity was not reduced compared to

Varietya Size, mm
Commercial varieties
Bragg 7.40
Kanrich 7.85
Kirby 6.93
Late Giant 10.00
Experimental strains


EaLE% UA aDAH TIa, mna/a

Seed Coat















D82-3332 (IR)
F85-ll346 (VT)
F85-11349 (VT)
F80-6692 (VT)
F83-7895 (VT)
F83-7923 (VT)
F83-7959 (VT)
F83-8177 (VT)
F85-494 (IL-1)
F85-495 (IL-1)
F85-606 (IL-2)
F85-604 (IL-2)
F85-2297 (LM)
F85-2773 (IR)
F85-2757 (IR)
F85-2853 (IR)
F85-2892 (IR)
F85-2927 (IR)
F85-2983 (IR)
F85-3093 (IR)
F85-3182 (IR)
F85-3208 (IR)
F85-3229 (IR)
F85-3981 (HS)
F85-7356 (HS)
F85-7433 (HS)
L81-4387 (AK)
L81-4590 (AK)
UFV-1 (LM)
Grown at differ
F85-994 (IL-G)
F85-994 (IL-PR
F85-998 (IL-G)
F85-998 (IL-PR
C. V.

6.96 Y
6.82 Y
10.17 Y
10.32 B
7.85 Y
7.58 Y
8.02 Y
8.18 Y
9.61 B
6.01 Y
6.29 Y
6.54 Y
6.18 Y
6.69 Y
6.29 Y
6.37 Y
5.68 Y
5.92 Y
5.68 Y
6.95 Y
6.66 Y
5.92 Y
6.18 Y
6.43 Y
6.08 Y
7.29 Y
6.64 Y
6.23 Y
7.09 Y
6.53 Y
rent locations
7.73 Y
) 7.08 Y
7.04 Y
) 8,09 Y

a IR-insect resistant; VT-vegetable type; LM-late maturing;
IL-isolines; HS-hard seed coat; AK-absence of the Kunitz inhibitor;
B-grown in Gainesviller Florida; PR-grown in Puerto Rico; Y-yellow;
B-black; UA-urease activity; TI-trypsin inhibitor activity.

the average TI activity. These results are in contrast with those of

Leiner and Tomlinson (1981) who reported a 50% reduction in the TI

activity of soybeans in which the Kunitz TI was absent.

Coefficients and probability levels of gross correlations are

given in Table 10. Size of the soybeans was not significantly

correlated with the protein (P>.336) and fat (P>.119) concentrations

but was positively correlated with the UA (P<.0002) and TI (P<.0005).

The lack of a significant correlation between the size of the seeds

and their protein concentration is in agreement with the findings of

Krober and Cartter (1962). Similar to Trial 1, the fat concentration

was negatively correlated with the protein content (P<.014) and

positively correlated with the UA (P<.054) of the unheated soybeans.

Likewise the TI activity was not correlated with the protein

concentration (P<.232) but was positively correlated with the UA

(P<.015). However, inconsistent with Trial 1, the protein

concentration was not correlated with the UA (P<.372) and the fat

concentration was not correlated with the TI activity (P<.130).

~Trial 3

Grower period. Average daily gain and feed efficiency of pigs

fed unheated Bragg soybeans were depressed (P<.05) and growth of pigs

fed soybean meal was superior (P<.05) when compared to pigs fed the

other dietary treatments (Table 11). The Bragg soybeans required

roasting at 110 C to permit growth and feed efficiency similar (P>.05)

to that of pigs which received the unheated Davis soybeans.

Similarly, to equal the growth of pigs fed Davis soybeans roasted at

110 C, the Bragg soybeans had to be roasted at 125 C. Roasting either

variety of soybeans at 110 C improved (P<.05) pig growth and feed

, p ma gm

Urease activity, a pH .383

SUrease activity (UA, npH); Trypsin inhibitor activity (TI, mg/gm).



UAa A H TTa /

Protein N

aF t 5

Sizer mm










Proteins %

Fat, %

I **




oONre e u
rr) N






0 m

a Or

a mr~nl
o mm(3


mon am



~a ~

mN a



1 HQ O

N~ u

O mc

N hl

O mN





C 40

sees a,
mam O Y
<<<< > a


rl mN



ag m

.... )
0 ~ ch

efficiency when compared to pigs fed the unheated soybeans.

Increasing the roasting temperature of the Bragg soybeans from 110 C

to 125 C permitted an additional increase (P<.05) in pig growth but

feed-to-gain ratio was not improved (P>.05).

Feed intake of the diets containing Bragg soybeans (unheated or

roasted at 110 C) was depressed (P<.05) compared to the consumption of

the diets containing either soybean meals Bragg soybeans roasted at

125 CI or Davis soybeans roasted at 110 C. Daily Intake of the diets

containing the unheated soybeans did not differ (P>.05).

_Finisher period. The average daily gain data were adjusted for

the large difference in initial weight due to simultaneously switching

each treatment group to their respective finisher diet. Pigs fed the

diets containing unheated Bragg soybeans continued to have inadequate

growth and feed efficiency. However, growth and feed-to-gain ratio of

pigs fed the diet containing unheated Davis soybeans were equal

(P>.05) to pigs fed diets containing soybean meal or roasted soybeans

of either variety. Roasting the Brag9 soybeans at 110 C improved

(P<.05) their utilization but did not provide (P>.05) an additional

Increase in growth compared to pigs fed the unheated Bragg soybeans.

Increasing the roasting temperature of Bragg soybeans from 110 to 125

C did not improve (P>.05) their feeding value. There were no

differences (P>.05) in feed intake between the dietary treatments

during the finisher period.

0yerall. Pigs fed unheated Bragg soybeans during both the grower

and finisher periods had poorer gain (P<.05) and were less (P<.05)

efficient than any other treatment group. Average daily gain of pigs

was improved (P<.05) by feeding either variety of soybeans roasted at

110 C when compared to pigs fed the unheated soybeans. Feed-to-gain

ratio was Improved (P<.05) by roasting the Bragg soybeans at 110 C but

feed efficiency of pigs fed Davis soybeans roasted at 110 C did not

improve (P>.05) compared to pigs fed unheated soybeans. Roasting the

Bragg soybeans at 125 C permitted an additional increase (P<.05) in

pig growth but feed efficiency (F/G) was similar to pigs fed Bragg

soybeans roasted at 110 C. Feed intake during the entire

growing-finishing period did not differ (P>.05) between the different

treatment groups.

The TI and UA of the different soybean products are presented in

Table 12. None of the heat treatments was sufficient to lower the UA

of either variety of soybeans to less than .2 which is considered by

commercial soybean processing plants as indicative of optimum heat

processing (Smithr 1977). The UA is an indication of the

anti-nutritional factors contained in soybean products (Caskey and

Knapp, 1944; Albrecht et al., 1966; McNaughton and Reece, 1980). The

unheated Davis soybeans had lower UA and TI activities compared to the

Bragg soybeans (unheated or roasted at 110 C). Similarly the Davis

soybeans roasted at 110 C contained lower UA and TI activities then

the Bragg soybeans roasted 110 C. The Davis soybeans roasted at 110 C

also had a lower TI activity compared to the Bragg soybeans roasted at

125 C. Pig performance data reflect this varietal difference in UA

and indicate that soybean products having a UA higher than .2 can be

adequate protein supplements for growing-finishing swine diets. These

results agree with Vandergrift et al. (1983), Rudolph et al. (1983)


Soybean product Urease indexa Trypsin inhibitors
a pH mg/gm sample

Bragg soybeans, unheated 1.97 53.95
Bragg soybeans roasted at 110 C 1.80 44.08
Bragg soybeans roasted at 125 C .52 15.10
Davis soybeans, unheated 1.47 36.97
Davis soybeans roasted at 110 C .84 11.42

SIndex used by commercial soybean processing plants to monitor
quality of heat processing of soybeans. An index less than .20 is
associated with soybeans that have received adequate heat

and Seerley et al. (1974) that soybean products having UA higher than

.2 can still be satisfactory protein supplements for growing-finishing


These data indicated that the extent of heat treatment required

to destroy the heat labile anti-nutritional factors found in raw

soybeans can differ among varieties of soybeans. These varietal

differences could account for contrasting data reported for the

feeding value of unheated or roasted soybeans obtained in previous

studies. Similar performance of finishing pigs fed unheated Davis

soybeans or soybean meal is in agreement with data of Combs and

Wallace (1969) who reported that finishing pigs could efficiently

utilize raw soybeans; whereas the reduced performance of pigs fed

unheated Bragg soybeans in the present study is consistent with data

indicating that feeding raw soybeans would reduce pig performance

regardless of initial age or weight of pigs (Crenshaw and Danielson,

1985c; Pontif et al., 1986). However, Combs and Wallace (1969) along

with Jimenez et al. (1963), Hanke et al. (1972) and Yen et al. (1974)

have noted that raw soybeans are not a satisfactory protein supplement

for growing pigs. The results of the present study are in agreement

with these findings.

As previously stated the Bragg soybeans remained in the field

for one to two months longer than the Davis soybeans before being

harvested. The influence of this additional variable on the results

of this study cannot be determined but requires further assessment.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs