Soyasapogenol A and B Distribution in Soybean (Glycine max
L. Merr.) in Relation to Seed Physiology, Genetic Variability,
and Growing Location
H. P. VASANTHA RUPASINGHE,*,†CHUNG-JA C. JACKSON,†VAINO POYSA,‡
CHRISTINA DI BERARDO,§J. DEREK BEWLEY,§AND JONATHAN JENKINSON|
Guelph Center for Functional Foods, Laboratory Services, University of Guelph, Guelph, Ontario,
N1H 8J7, Canada, Greenhouse and Processing Crops Research Center, Agriculture and Agri-Food
Canada, Harrow, Ontario, N0R 1G0, Canada, Department of Botany, University of Guelph, Guelph,
Ontario, N1G 2W1, Canada, and First Line Seeds Ltd., Guelph, Ontario, N1H 6H8, Canada
An efficient analytical method utilizing high-performance liquid chromatography (HPLC)/evaporative
light scattering detector (ELSD) was developed to isolate and quantify the two major soyasaponin
aglycones or precursors in soybeans, triterpene soyasapogenol A and B. Soaking of seeds in water
up to 15 h did not change the content of soyasapogenols. Seed germination had no influence on
soyasapogenol A content but increased the accumulation of soyasapogenol B. Soyasapogenols were
mainly concentrated in the axis of the seeds as compared with the cotyledons and seed coat. In the
seedling, the root (radicle) contained the highest concentration of soyasapogenol A, while the plumule
had the greatest amounts of soyasapogenol B. In 10 advanced food-grade soybean cultivars grown
in four locations in Ontario, total soyasapogenol content in soybeans was 2 ( 0.3 mg/g. Soyasapogenol
B content (1.5 ( 0.27 mg/g) was 2.5-4.5-fold higher than soyasapogenol A content (0.49 ( 0.1
mg/g). A significant variation in soyasapogenol content was observed among cultivars and growing
locations. There was no significant correlation between the content of soyasapogenols and the total
KEYWORDS: Soybean; Glycine max; soyasaponin; soyasapogenol; triterpene aglycones; HPLC/ELSD;
isoflavones; germination; soaking
The presence of saponins in soybean has attracted consider-
able interest because of both their health benefits and adverse
sensory characteristics. Soybean saponins are triterpenoid gly-
cosides and comprise a hydrophobic aglycone (triterpenoid
soyasapogenol) linked to one or more hydrophilic mono- or
oligosaccharide moieties (1). Soyasaponins are classified into
two major groups, soyasaponin A and B (Figure 1) (2). Group
A acetylated saponins, present in soybean, are implicated as
the phytochemicals mostly responsible for undesirable bitter and
astringent taste (3, 4). However, group B saponins, including
conjugated saponins, possess several health benefits (5, 6). They
appear to possess inhibitory activity against infection by human
immunodeficiency virus (HIV) (7) and the activation of Ep-
stein-Barr virus early antigen (8). Soyasaponin B1in particular
is a potent inhibitor of HIV infection in vitro and offers great
potential in the treatment of retroviral infections (7). Recent in
vitro studies suggest that group B saponins also possess
hypocholesterolemic, immunostimulatory, anticarcinogenic, an-
tioxidative, antitumor, antivirus, antihepatitic, antidiabetic, and
hepatoprotective properties (9). Dietary saponins of soybean are
beneficial in preventing hypercholesterolemia and aortic ath-
erosclerosis in rats (10).
Group A saponins appear to be a naturally occurring form,
and Shiraiwa et al. (11) identified six different group A saponins,
designated as Aa, Ab, Ac, Ad, Ae, and Af, according to their
* To whom correspondence should be addressed. Tel: +1 519-823-1268.
Fax: +1 519-767-6240. E-mail: firstname.lastname@example.org.
†Guelph Center for Functional Foods, University of Guelph.
‡Agriculture and Agri-Food Canada.
§Department of Botany, University of Guelph.
|First Line Seeds Ltd..
Figure1. Chem icalstructures ofsoyasaponinprecursors soyasapogenol
A and B.
5888 J. Agric. Food Chem . 2003, 51, 5888−5894
10.1021/jf0343736 CCC: $25.00 ©2003 Am erican Chem ical Society
Published on Web 08/30/2003
elution order from high-performance liquid chromatography
(HPLC). However, acid hydrolysis of all six saponin A
compounds yielded the common aglycone, soyasapogenol A.
Characteristically, in all of these saponins, the terminal sugar
of the oligosaccharide chain, attached to the C-22 position of
the soyasapogenol A, is acetylated (11). Ireland and Dziedzic
(12) proposed that soyasapogenols C, D, and E are formed as
artifacts during acid hydrolysis. In contrast, Tsukamoto et al.
(13) recognized soyasapogenol E as the aglycone of the third
type of saponin, saponins Bd and Be. However, Bd and Be
saponins are heat labile and thus are presumed to be transformed
into soyasapogenol B during acid hydrolysis (5). The other
soyasaponins, which have been isolated and designated as
soyasaponins I, II, III, and IV, contain soyasapogenol B as the
common aglycone (14). Structurally, soyasapogenol A possesses
only an additional hydroxyl group at the C-21 position as
compared to soyasapogenol B (Figure 1).
The chemical structures of more than 20 saponins from
soybeans and various soy products have been determined (15).
However, it is now apparent that many of these reported
soyasaponins are artifacts formed due to alteration of the
chemical structures of the naturally occurring saponins during
extraction and analysis. Despite a great deal of analytical
research on soyasaponins, distinguishing and quantifying the
two major groups of soyasaponins (group A and B saponins)
has been hampered by the complex nature of the analytical
procedures due to the lack of intact glucoside forms of
soyasaponins (especially group A saponins) for use as standards
for quantitative determination. Therefore, in this study, we have
developed a rapid analytical technique to quantify the precursors
of the group A and B saponins, soyasapogenol A and B,
respectively. The total soyasaponin content is approximately
twice the total soyasapogenol content (16).
Seeds of soybean contain about 0.5% of soyasaponins (dry
weight basis) depending on the variety, cultivation year, location
grown, and degree of maturity (11, 17). Germination has been
suggested to enhance the overall nutritional value of the seeds
and the contents of health-promoting phytochemicals, including
saponins (18), ascorbic acid, and riboflavin (19). To further
characterize the major groups of soyasaponins in soybeans, we
developed a rapid analytical technique to quantitate soyasapo-
genols A and B and determined their distribution (i) in seeds
and seedlings, (ii) during seed soaking and germination, and
(iii) among advanced food-grade soybean cultivars.
MATERIALS AND METHODS
Plant Materials. For the histological and germination studies, three
advanced food-grade soybean cultivars of the 2001 harvest were selected
from a commercial soybean breeder (First Line Seeds Ltd., Guelph,
Ontario, Canada). Cultivar 41102-A has a higher sugar content and
the ability to produce better tasting soyfoods such as soymilk. Cultivar
41102-B has large white seeds of excellent quality, high protein, and
ability to produce firm, smooth, white tofu. Cultivar 41102-C has a
very high protein content (approximately 49% on dry weight basis).
For the comparison of soybean cultivars, 10 advanced food-grade
ones (S20-F8, CL970321, 7025308, 9910, 2004, AC X790P, Harovin-
ton, AC Vin-Pro, AC Hime, and 9305) were selected. They were grown
at four different geographical locations (Chatham, Malden, Tilbury,
and Woodslee) in Ontario, Canada, during the 1999 season. Standard
three replicate randomized complete block yield trails were grown at
each location following normal agronomic practices.
Chemicals. Concentrated hydrochloric acid, glacial acetic acid,
HPLC-grade acetonitrile, and methanol were purchased from Caledon
(Mississauga, Ontario, Canada). 1-Propanol was a product of Fisher
Scientific, and 95% ethanol was purchased from Commercial Alcohol
Inc., Toronto, Ontario, Canada. Deionized water was generated from a
Milli-Q analytical deionization system.
Seed Soaking and Germination. Seed soaking and germination
studies were conducted in triplicate, each replicate containing ap-
proximately 3 g of seeds. Each replicate was soaked separately in a
beaker at room temperature with 200 mL of distilled water, with
aeration. For the germination studies, seeds were soaked for 2 h before
being transferred into glass Petri dishes (15 cm diameter) with
prewetted, sterilized cotton. Seeds were spread evenly, and distilled
water was added periodically and incubated at 25 ( 1 °C.
After the designated soaking or germination period, excess water
was drained, and the seeds were immersed in liquid nitrogen and then
stored at -70 °C until used for freeze-drying. Samples were freeze-
dried for 48 h using a Labconco freeze-dry system (10 × 10-3mbar,
-40 °C) (Caltec Scientific Ltd., Toronto, Ontario, Canada) and ground
immediately into a fine powder with a mortar and pestle. The ground
powder was stored in a desiccator at -70 °C until used for the HPLC
To determine the distribution of soyasapogenols within seeds and
seedlings, approximately 15 g lots of seeds (in triplicate) were soaked
in distilled water for 10 h with aeration before being separated into
axis, seed coat, and cotyledons. For seedling parts, seeds were germ-
inated and grown for 120 h before dissection. The seeds and seedling
parts were freeze-dried and ground into a fine powder as described
Extraction of Saponins and Analysis of Soyasapogenols A and
B. The isolation of saponins from soybean was accomplished by a
modification of the method described by Daveby et al. (20). Finely
ground soybean powder (0.2 g) was dissolved in 30 mL of 80% (v/v)
aqueous ethanol in a round-bottom flask (125 mL) with consistent
mixing in a horizontal water bath/shaker (SW22 model, Julabo Inc.,
Allentown, PA) at 50 °C for 2 h. The residue was removed by cen-
trifuging the extract at 3000 rpm for 10 min using a bench centrifuge
at room temperature and decanting the clear supernatant. Fifteen
milliliters of the supernatant was dried under reduced pressure using a
rotary vacuum evaporator system (Buchi B-481 model, Brinkmann
Instruments, Palo Alto, CA), and the remaining residue was redissolved
in 8 mL of 1 N HCl in anhydrous methanol. The resuspension was
transferred to a screw-capped glass vial and was subjected to acid
hydrolysis at 75 °C for 2.5 h in a horizontal water bath/shaker to release
the aglycones (soyasapogenols) from soyasaponins. The soyasapogenols
produced were isolated by solid phase extraction using a Vac 6CC 500
mg C-18 Sep-Pak cartridge (Waters, Marlborough, MA) by eluting the
8 mL of solution through the cartridge, washing the cartridge with 3
mL of water, and eluting the soyasapogenols with 100% methanol.
The eluent was filtered through a 0.2 µm nylon filter (Waters Chrom-
atography, Mississauga, ON) and a 25 µL aliquot separated by reverse-
phase (RP) HPLC using a Shimadzu 10AD HPLC system consisting a
SCL-10A system controller, SIL-10A autoinjector (Shimadzu Scientific
Instruments, Columbia, MD) and an evaporative light scattering detector
(ELSD) (Alltech ELSD 2000 model, Alltech Assoc., Deerfield, IL).
Soyasapogenols A and B were well-resolved using a 250 mm long
× 4.6 mm i.d. ODS C18column with a C18guard column (Phenomenex,
Torrance, CA) at a flow rate of 0.9 mL/min pumped isocratically with
a mobile phase consisting of acetonitrile:1-propanol:water:0.1% acetic
acid (80:6:13.9:0.1). The detection of soyasapogenols was performed
using the ELSD, which was set to the drift tube temperature of 70 °C,
and the nebulizer nitrogen gas flow was adjusted to 2 mL/min.
Soyasapogenol A and B standards (ChromaDex, Santa Ana, CA) were
initially dissolved in 100% methanol (1000 µg/mL) and stored at -70
°C in a desiccator. Standard curves were obtained by plotting standard
concentrations (soyasapogenol A, 12.5, 25, 50, 75, 100, and 150 µg/
mL; soyasapogenol B, 37.5, 75, 150, 225, 300, and 450 µg/mL) as a
function of peak area in HPLC chromatograms. Quadratic calibration
curves (r2> 0.98) were generated by the Shimadzu Class VP software.
The quantification of soyasapogenol A (Rt) 5.9 min) and soyasapo-
genol B (Rt) 9.9 min) was performed on the basis of the peak area of
chromatograms created by ELSD in comparison to the standard curves
of authentic external standards of soyasapogenols A and B.
HPLC Analysis of Three Isoflavone Aglycones. Analysis of
daidzein, glycitein, and genistein was performed by a modification of
Soyasapogenol A and B Distribution in SoybeanJ. Agric. Food Chem ., Vol. 51, No. 20, 20035889
the method described by Franke et al. (21). Approximately 0.5 g of
finely ground soybean powder was added into 50 mL glass vials.
Simultaneous extraction of isoflavones and acid hydrolysis to their
respective aglycones was carried out by mixing with 12 mL of 2 N
HCl in 100% ethanol and incubating at 125 °C for 2 h. After acid
hydrolysis, the samples were centrifuged at 3000 rpm for 10 min in a
bench centrifuge at room temperature. An aliquot of the clear
supernatant was filtered through a 0.45 mm nylon filter before being
introduced into the HPLC consisting of a Waters 600E multisolvent
delivery system, 717 plus auto sampler, 996 photodiode array detector
monitoring at 200-350 nm or a Waters 486 tunable absorbance UV
detector set at 254 nm and Millennium (version 2.10) chromatography
software. The column used was a 250 mm long × 3.9 mm i.d. Waters
Nova Pak C-18 with a C-18 guard column. All HPLC analyses were
performed at ambient temperature. A Hewlett-Packard 8452A diode
array spectrophotometer was used to verify the concentrations of stock
standards (daidzein, glycitein, and genistein) (Indofine Chemical,
Somerville, NJ). The mobile phases for HPLC consisted of solvent
(A) 4.0% acetic acid in filtered MilliQ water and (B) 100% methanol.
The solvent gradient was as follows: solvent B was increased from 40
to 65% over 10 min, then run constantly for the next 11 min, and finally
decreased back to 40% within 9 min and held for another 5 min before
the next injection. The flow rate was 1 mL/min. The minimum
detectable concentrations for genistein and daidzein were 100 and 185
ng/mL, respectively. UV spectra were recorded, and area responses
were integrated by Waters software.
Statistical Analysis. Statistical analysis of variance in a complete
randomized design was performed using PROC GLM procedure of the
SAS System version 8e for Windows. Mean separations were examined
using Tukey’s Studentized range test (t-test).
RESULTS AND DISCUSSION
Isolation and Determination of Soyasaponins. A primary
goal of the research was to develop an analytical technique to
determine the two major saponin components (soyasapogenols
A and B) in soybeans. In the literature, data on the quantification
of saponin precursors (soyasapogenols) are contradictory be-
cause of the formation of artifacts (soyasapogenols C, D, E,
and F) due to acid hydrolysis during their preparation. In
previous studies, UV detection at the wavelength of 204 nm
was commonly used, but this is inefficient due to the lack of
prominent chromophores for soyasapogenols. Quantification of
the two major groups of soyasaponins (group A and B saponins)
has been hampered also by the complex nature of the analytical
procedures and the lack of intact glucoside forms of soyasa-
ponins (especially group A saponins) to use as standards for
quantitave determination. Recently, Gurfinkel and Rao (2)
proposed a rapid analytical method to quantify total soyasa-
ponins based on direct densitometry, but this method is not able
to distinguish the two major groups of saponins.
In the present study, the alternative use of ELSD was
successfully employed to characterize and quantitate soyasa-
pogenols A and B with very high resolution and high sensitivity.
ELSD is based on the scattering of light by the nebulized solutes
without absorption by the analytes. This technique has been used
successfully in the determination of other saponins (22).
On the basis of previously reported extraction methods for
saponins, five extraction protocols were compared for the
extraction recovery of soyasapogenols (Table 1). Extraction in
80% aqueous ethanol produced the highest recovery of soyasa-
pogenols when compared with 100% methanol (Table 1). The
amount of soyasapogenols obtained from soybean extracts (200
mg of soybean in 30 mL of 80% ethanol) was optimized by
incubating them for 2 h at 50 °C. Multiple extractions were
performed to determine the extraction efficiency for soybeans,
and the third successive reextraction yielded undetectable
amounts of soyasapogenols A and B. A single extraction was
able to extract over 96% of total soyasapogenols from soybean
powder. Before acid hydrolysis of soyasaponins to remove the
sugar moieties, the ethanolic extract was evaporated under
reduced pressure at 40 °C, and the residue was dissolved in 1
N HCl in anhydrous methanol before hydrolysis at 75 °C for
2.5 h. This procedure yielded the highest recovery of soyasa-
pogenols A and B without forming artifacts. Performing acid
hydrolysis with the same strength of H2SO4 produced less
soyasapogenols. HCl-dioxane has been used successfully for
acid hydrolysis of soybeans (23) and other legume (Pueraria
lobata L.) triterpene saponins (24), but it was less efficient for
hydrolysis of soyasaponins than HCl-methanol.
Similarly to the present finding, Daveby et al. (20) observed
higher and more stable extraction of DDMP-conjugated so-
yasaponin I from dehulled peas (Pisum satiVum L.) using 80%
ethanol, as compared to 100% methanol. Most of the previously
reported methods for extractions of saponins included defatting
of soybean powder for at least 36 h in methanol using a Soxhlet
apparatus (3, 14). However, we found that defatting soybean
powder using a Soxhlet extraction (140 °C in hexane for 1 h)
prior to saponin extraction reduced the recovery of soyasapo-
genols by about 25% as compared to nondefatted samples.
Tsukamoto et al. (13) classified soyasapogenol E as a third
aglycone moiety of soyasaponins, Bd and Be. In the present
study, while ELSD could detect the authentic soyasapogenol
E, neither this nor any other related compound was detected in
the extracts of soybeans. Soyasaponins Bd and Be are heat labile
and could have been transformed into soyasapogenol B during
acid hydrolysis (5). However, when soyasapogenol E was
subjected to acid hydrolysis, it was not converted to soyasapo-
genol B or degraded. Therefore, under the present extraction
and hydrolysis conditions, intact soyasaponin E could be
converted to soyasaponin B and then hydrolyzed into soyasa-
pogenol B. Alternatively, under the previously reported rigorous
defatting, extraction, and acid hydrolysis conditions (3, 14),
soyasapogenol E may have formed due to the pinacol rear-
Table 1. Efficiency of Different ExtractionMethods onthe Recovery of Soyasapogenol A and B
71± 0.3240± 130
80%ethanol65± 0.4227± 110
a ,bAll of the extractions (50 °C for 2 h) and acid hydrolysis (75 °C for 2.5 h) were as described in the Materials and Methods.
5890 J. Agric. Food Chem ., Vol. 51, No. 20, 2003 Rupasinghe et al.
rangement of soyasapogenol A (Robert Lange, University of
Guelph, personal communication). The present results supported
the notion that soyasapogenols C, D, E, and F were artifacts of
the extraction and hydrolysis procedure employed (12).
The two soyasapogenols (A and B) were well-resolved with
the retention time difference of 4 min (Figure 2). The retention
time and relative standard deviation (RSD) values for measured
soyasapogenols A and B were 5.9 (RSD ) 1.2%, n ) 18) and
9.9 min (RSD ) 1.8%, n ) 18), respectively, reflecting an
acceptable precision. The within day RSD values for detection
of soyasapogenols A and B were 5.6 (n ) 6) and 7.4 (n ) 6),
respectively. The peak identity was determined by spiking
soybean extracts with known amounts of authentic analytes and
observing the peak overlap of the resultant chromatograms. For
validating analytical accuracy, 200 mg portions of a reference
sample (“control” soybean powder) were spiked with three
different concentrations of authentic soyasapogenol A and
soyasapogenol B for the calculation of percent recovery. The
overall mean recoveries of soyasapogenols A and B were 104.7
and 102.7%, respectively. The limits of quantitation for soyasa-
pogenols A and B were 0.06 and 0.08 mg/g, respectively. Thus,
the method described here can be used as a rapid analytical
tool to determine soyasapogenols A and B content and to
estimate approximate total saponin content in commercial
soybean food products such as soya protein extracts and soybean
Effect of Soaking and Germination on Soyasapogenol
Content. The total soyasapogenol content was significantly
different between the seeds of the three soybean cultivars
employed. The seed size of cultivar 41102-B was larger than
the other two, but it contained the lowest concentration of total
soyasapogenols on a whole seed basis.
Soyasapogenols A and B content was not affected by soaking
seeds in water for up to 15 h in the three cultivars tested (Table
2). Thus, hydration of seeds does not result in any net
biosynthesis of soyasapogenols nor a net leaching of soyasa-
pogenols from seeds. Similarly, Ruiz et al. (25) reported no
influence of soaking on soyasaponin VI content of two other
legumes, chickpeas (Cicer arietinum L.) and lentils (Lens
Seed germination and subsequent seedling growth have a
marked effect on the chemical composition, biochemical
constituents, antinutritional factors, and functional properties of
soybean (26). Development of food products from germinated
seed has been suggested as a way to increase the versatility
and utility of soybean through the in situ modification of certain
specific biologically active compounds including saponins,
phytoestrogens, lecithin, and phytosterols (18). Our findings
indicate that during seedling growth there is a differential effect
on soyasapogenols A and B content. Soyasapogenol A content
was not affected for up to 60 h from the start of imbibition, but
soyasapogenol B content increased after 60 h by 32% (Table
3). Germination was completed relatively faster in cultivar
41102-C (after approximately 20 h) as compared to 36 h in the
other cultivars, but there was no statistically significant interac-
tion of soyasapogenol content between cultivar and germination
Shimoyamada and Okubo (27) found that germination and
growth of soybean for 8 days increased seed soyasaponin content
8-fold. Amaranthus (Amaranthus cruentus L.) seeds contain
0.09-0.1% (dry matter basis) triterpene saponins, and an
increase to 0.18% was observed after 4 days of seedling growth
(28). In contrast, Duhan et al. (29) found that soaking for up to
18 h and germination/seedling growth for up to 48 h reduced
the saponin content in four cultivars of pigeon pea (Cajanus
cajan L.), another grain legume. Loss of saponin during soaking
and germination of this legume species was due to leaching
out of saponin into the soaking water solution through diffusion
Soyasapogenol Distribution in Seed and Seedling Parts.
The ratio of soyasapogenol A in axis:cotyledon was 40:1,
whereas that of soyasapogenol B was 9:1 (Table 4). The total
soyasapogenol concentration was 14-fold greater in the axis
(15.6 mg/g dw) as compared to that in cotyledons (1.1 mg/g
Figure2. RepresentativeHPLC/ELSDchrom atogramofextractfromthe
soybean. Peaks: A, soyasapogenol A; B, soyasapogenol B.
Table 2. Effect of Soaking SoybeanSeeds onTheir Soyasapogenol A
and B Content
Table 3. Effect of Early Seedling Growthonthe Concentrationof
Soyasapogenol A and B
im bibition(h)41102-A 41102-C
0 0.21± 0.02
0 0.78± 0.04
Soyasapogenol A and B Distribution in Soybean J. Agric. Food Chem ., Vol. 51, No. 20, 20035891
dw) (Table 4). However, once the dry weight ratios of seed
part to total seed weight were taken into consideration, the
percentage distribution of total soyasapogenols among axis, seed
coat, and cotyledons was 62, 0.8, and 37.2%, respectively.
Similar to the present finding, Shimoyamada et al. (30)
reported that two metabolites of soyasapogenol A, acetyl-
soyasaponins A1and A4, occur only in immature hypocotyls of
soybean seed. Using HPLC of fluorescent derivatives of the
saponins, Tani et al. (31) reported that soyasaponins, except
soyasaponin II, are located predominantly in the axis of seed
(plumule, hypocotyl, and radicle), with the content of total
soyasaponins in the axis being about six times higher than that
in the cotyledons. In the above study, soyasaponins were not
detected in the seed coat but the axis accounted for about 90%
of the whole grain weight (31). It is evident that soyasapogenols
are synthesized in the physiologically active axis of the seed
and perhaps some of them translocate to the cotyledons. Kudou
et al. (23) reported that soyasaponin Rg is present only in
hypocotyls, soyasaponin ?a in the cotyledons, and soyasaponin
?g in both parts. In addition, ginsenoside, the predominant
saponin compound in ginseng (Panax ginseng L.), is also
localized in specific tissues in roots (31). Unlike saponins, 80-
90% of total seed isoflavones are located in the cotyledons, with
the reminder in the hypocotyls (32).
Dehulling of soaked soybean seeds is a common practice
during soybean processing for soymilk and other soy food
manufacturing. However, the present results suggest that such
a practice could affect the nutritional value of soy foods since
important health-promoting phytochemicals including saponins
are located in the axis, which is removed through dehulling.
Distribution of soyasapogenols A and B was also determined
in the radicle, plumule, and cotyledons in 120 h old seedlings
(Table 5). Soyasapogenol concentration in cotyledons of
seedlings was greater as compared to that in seeds due to the
reduced dry matter content, which resulted from mobilization
and utilization of stored carbohydrates, proteins, and oils from
cotyledons by the embryonic axis during the seedling develop-
ment (19). A distinctly different distribution of soyasapogenols
A and B in the radicle and plumule was noted. Soyasapogenol
A concentration was about 9-fold higher in the radicle as
compared to that in the plumule. In contrast, soyasapogenol B
concentration was about 3-fold higher in the plumule than the
radicle. With regard to the total soyasapogenol distribution
within the seedling, they were more abundant (approximately
2-fold) in the radicle than in the plumule (Table 5).
Distribution of Soyasapogenol among Food-Grade Soy-
bean Cultivars. The concentrations of soyasapogenols A and
B were determined in 10 selected food-grade soybean cultivars,
which were grown in four locations (Table 6). The total
soyasapogenol content in food-grade soybeans ranged from 1.43
to 2.65 mg/g, or an average of 0.2% “as is” weight basis. The
concentration of soyasapogenol B (1.5 ( 0.27 mg/g) was 2.5-
4.5-fold higher than the concentration of soyasapogenol A (0.49
( 0.1 mg/g) in all of the cultivars tested. The analysis of
variance compared the 10 cultivars in four locations. The
concentrations of soyasapogenols A and B were influenced by
both cultivars (p < 0.001) and by location (p < 0.001). Mean
soyasapogenol A content among cultivars ranged from 0.34 to
0.60 mg/g, while mean soyasapogenol B content ranged from
1.19 to 1.88 mg/g (Table 6). Hu et al. (33) found that the total
Table 4. Distributionof Soyasapogenol A and B inthe Seed Parts of Three SoybeanCultivars
concentrationm ean± SD(m g/gdryweight)
totalam ount(m g)
Table 5. Distributionof Soyasapogenol A and B inthe Radicle, Plum ule, and Cotyledons of 120 hold Seedlings inThree SoybeanCultivars
concentrationm ean± SD(m g/gdryweight)
totalam ount(m g)
5892J. Agric. Food Chem ., Vol. 51, No. 20, 2003Rupasinghe et al.
content of three major group B saponins and their non-DDMP
counterparts ranged from 2.5 to 5.85 µmol/g among 46 cultivars
of soybean. Genetically modified “Round-up ready” soybeans
have significantly (p < 0.05) lower amounts of group B
soyasaponin than those of conventional cultivars (33), perhaps
because these soybeans suffer less environmental stress when
other plants are removed by the herbicide (34). Similarly,
Tsukamoto et al. (35) found that soyasaponin concentrations in
cultivated soybeans are 2-fold less than in wild soybeans.
Relationship between Soyasapogenols and Isoflavones.
Isoflavones in soybean have been credited with performing
several health-promoting functions including prevention of
cardiovascular diseases, hormone-related cancers such as breast
and prostate cancers, and menopausal symptoms (36-38).
Therefore, attempts have been made to enhance the isoflavone
content of soybean cultivars and soy foods (39). We determined
the content of three isoflavone aglycones, genistein, daidzein,
and glycitein, in the 10 soybean cultivars grown in four locations
and compared them with the content of soyasaponin aglycones,
soyasapogenols A and B.
The concentration of total isoflavone aglycones ranged from
0.91 to 2.45 mg/g. The ratio of genistein, daidzein, and glycitein
in the 10 cultivars was approximately 49:46:5. Total soyasa-
pogenols A and B content (Table 6) and total isoflavone
aglycone content (data not presented) were the highest in
soybean cultivars grown in Woodslee. There was no apparent
relationship (r2) 0.057) between the distribution of isoflavone
and soyasapogenols in the 10 cultivars tested (Figure 3).
Similarly, Hu et al. (33) found no statistically significant
correlation between total isoflavone concentrations and six group
B saponins in 46 cultivars of soybeans. This relationship needs
to be further analyzed using a larger number of more genetically
diverse soybean cultivars. It is also important to study the effect
of growing conditions such as total heat units and soil moisture
and nutrient factors on the accumulation of these health-
promoting secondary metabolites in soybean. Among the 10
cultivars, the ratio of total isoflavone aglycones to total
soyasapogenols ranged from 0.5 to 1.37 on mg/g basis.
Biosynthesis of sapogenols and isoflavones occurs through two
distinct metabolic pathways, the isoprenoid pathway and the
phenylpropanoid pathway, respectively, and there is not any
close metabolic coordination between the biosynthetic pathways
for these two groups of secondary metabolites.
Identification and development of food-grade cultivars with
enhanced health-promoting phytochemicals are important for
the value-added food industry. Enhancement of beneficial group
B saponins and isoflavones, together with the elimination of
group A saponins, will contribute to improvement of the quality
of soybean and their food products and better public acceptance.
We acknowledge the technical assistance provided by Cameron
Lyttle and Sara Little.
(1) Lasztity, R.; Hidvegi, M.; Bata, A. Saponins in food. Food ReV.
Int. 1998, 14, 371-390.
(2) Gurfinkel, D. M.; Rao, A. V. Determination of saponins in
legumes by direct densitometry. J. Agric. Food Chem. 2002, 50,
(3) Kitagawa, I.; Taniyama, T.; Nagahama, Y.; Okubo, K.; Yamau-
chi, F.; Yoshikawa, M. Saponin and sapogenol. XLII. Structures
of acetyl-soyasaponins A1, A2, and A3, astringent partially
acetylated bisdesmosides of soyasapogenol A, from American
soybean, the seeds of Glycine max Merrill. Chem. Pharm. Bull.
1988, 36, 2819-2828.
(4) Okubo, K.; Iijima, M.; Kobayashi, Y.; Yoshikoshi, M.; Uchida,
T.; Kudou, S. Components responsible for the undesirable taste
of soybean seeds. Biosci., Biotechnol., Biochem. 1992, 56, 99-
(5) Kudou, S.; Tonomura, M.; Tsukamoto, C.; Shimoyamada, M.;
Uchida, T.; Okubo, K. Isolation and structural elucidation of
DMPP-conjugated soyasaponins as genuine saponins from
soybean seeds. Biosci., Biotechnol., Biochem. 1992, 56, 142-
(6) Rao, A. V.; Gurfinkel, D. M. The bioactivity of saponins:
Triterpenoid and steroidal glycosides. Drug Metab. Drug Inter-
act. 2000, 17, 211-236.
(7) Nakashima, H.; Okubo, K.; Honda, Y.; Tamura, T.; Matsuda,
S.; Yamamoto, N. Inhibitory effect of glycosides such as saponin
from soybean on the infectivity of HIV in vitro. AIDS 1989, 3,
(8) Konoshima, T.; Kozuka, M. Constituents of leguminous plants,
XIII. New triterpenoid saponins from Wisteria brachybotrys. J.
Nat. Prod. 1991, 54, 830-836.
Table 6. Distributionof Soyasapogenol A andB inTenFood-Grade
Soybean Cultivars Grown in Four Locations of Ontario, Canadaa
cultivarsChatham Malden TilburyWoodslee m ean± SDa
AC Him e
AC Him e
aDifferentsuperscriptsinthecolum nindicatesignificantdifferencesinthem ean
at R ) 0.05.
Figure 3. Relationshipbetweenthe concentrations of soyasapogenol A
and B and those of three isoflavone aglycones (daidzein, glycitein, and
genistein) in 10 food-grade soybean cultivars grown in four different
locations in Ontario.
Soyasapogenol A and B Distribution in SoybeanJ. Agric. Food Chem ., Vol. 51, No. 20, 2003 5893
(9) Fournier, D. B.; Erdman, J. W., Jr.; Gordon, G. B. Soy, its Download full-text
components, and cancer prevention: a review of in vitro, animal,
and human data. Cancer Epidemiol., Biomarkers PreV. 1998, 7,
(10) Oakenfull, D. G.; Topping, D. L.; Illman, R. J.; Fenwick, D. E.
Prevention of dietary hypercholesterolaemia in the rat by soya
bean and Quillaja saponins. Nutr. Rep. Int. 1984, 29, 1039-
(11) Shiraiwa, M.; Kudo, S.; Shimoyamada, M.; Harada, K.; Okubo,
K. Composition and structure of “group A saponin” in soybean
seed. Agric. Biol. Chem. 1991, 55, 315-322.
(12) Ireland, P. A.; Dziedzic, S. Z. Effect of hydrolysis on sapogenin
release in soya. J. Agric. Food Chem. 1986, 34, 1037-1041.
(13) Tsukamoto, C.; Kikuchi, A.; Kudou, S.; Harada, K.; Kitamura,
K.; Okubo, K. Group A acetyl saponin-deficient mutant from
the wild soybean. Phytochemistry 1993, 31, 4139-4142.
(14) Ireland. P. A.; Dziedzic, S. Z. Saponins and sapogenins of chick
pea, haricot bean and red kidney bean. Food Chem. 1987, 23,
(15) Berhow, M. A.; Cantrell, C. L.; Duval, S. M.; Dobbins, T.;
Maynes, J.; Vaughn, S. F. Analysis and quantitation of group B
saponins in processed soybean products. Phytochem. Anal. 2002,
(16) Gestetner, B.; Birk, Y.; Bondi, Y.; Tencer, Y. Method for the
determination of sapogenin and saponin contents in soybeans.
Phytochemistry 1966, 5, 803-806.
(17) Fenwick, D. E.; Oakenfull, D. Saponin content of food plants
and some prepared foods. J. Sci. Food Agric. 1981, 34, 186-
(18) Bau, H. M.; Villaume, C.; Mejean, L. Effects of soybean (Glycine
max) germination on biologically active components, nutritional
values of seeds, and biological characteristics in rats. Nahrung
2000, 44, 2-6.
(19) Ahmad, S.; Pathak, D. K. Nutritional changes in soybean during
germination. J. Food Sci. Technol. 2000, 37, 665-666.
(20) Daveby, Y. D.; Aman, P.; Betz, J. M.; Musser, S. M. Effect of
storage and extraction on ratio of soyasaponin I to 2,3-dihydro-
2,5-dihydroxy-6-methyl-4-pyrone-conjugated soyasaponin I in
dehulled peas (Pisum satiVum L.). J. Sci. Food Agric. 1998, 78,
(21) Franke, A. A.; Custer, L. J.; Cerna, C. M.; Narala, K. Rapid
HPLC analysis of dietary phytoestrogens from legumes and from
human urine. Proc. Soc. Exp. Biol. Med. 1995, 208, 16-
(22) Ganzera, M.; Bedir, E.; Khan, I. A. Determination of steroidal
saponins in Tribulus terrestris by reversed-phase high-perfor-
mance liquid chromatography and evaporative light scattering
detection. J. Pharm. Sci. 2001, 90, 1752-1758.
(23) Kudou, S.; Tonomura, M.; Tsukamoto, C.; Uchida, T.; Sakabe,
T.; Tamura, N.; Okubo, K. Isolation and structural elucidation
of DMPP-conjugated soyasaponins as genuine saponins from
soybean seeds. Biosci., Biotechnol., Biochem. 1993, 57, 546-
(24) Oh, S.-R.; Kinjo, J.; Shii, Y.; Ikeda, T.; Nohara, T.; Ahn, K. S.;
Kim, J. H.; Lee, H. K. Effects of triterpenoids from Pueraria
lobata on immunohemolysis: ?-D-glucuronic acid plays an active
role in anticomplementary activity in vitro. Planta Med. 2000,
(25) Ruiz, R. G.; Price K. R.; Arthur, A. E.; Rose, M. E.; Rhodes,
M. J. C.; Fenwick, R. G. Effect of soaking and cooking on the
saponin content and composition of chickpeas (Cicer arietinum)
and lentils (Lens culinaris). J. Agric. Food Chem. 1996, 44,
(26) Suberbie, R. D.; Mendizabal, C.; Mendizabal, J. Germination
of soybeans and its modifying effects on the quality of full-fat
soy flour processing method. J. Am. Oil Chem. Soc. 1981, 58,
(27) Shimoyamada, M.; Okubo, K. Variation in saponin contents in
germinating soybean seeds and effect of light irradiation. Agric.
Biol. Chem. 1991, 55, 577-579.
(28) Oleszek, W.; Junkuszew, M.; Stochmal, A. Determination and
toxicity of saponins from Amaranthus cruentus seeds. J. Agric.
Food Chem. 1999, 47, 3685-3687.
(29) Duhan, A.; Khetarpaul, N.; Bishnoi, S. Saponin content and
trypsin inhibitor activity in processed and cooked pigeon pea
cultivars. Int. J. Food Sci. Nutr. 2001, 52, 53-59.
(30) Shimoyamada, M.; Kudo, S.; Okubo, K.; Yamauchi, F.; Harada,
K. Distributions of saponin constituents in some varieties of
soybean plant. Agric. Biol. Chem. 1990, 54, 77-81.
(31) Tani, T.; Katsuki, T.; Kubo, M.; Arichi, S.; Kitagawa, I.
Histochemistry. V. Soyasaponins in soybean (Glycine max
MERRILL) seed. Chem. Pharm. Bull. 1985, 33, 3829-3833.
(32) Tsukamoto, C.; Shimada, S.; Igita, K.; Kudou, S.; Kokubun, M.;
Okubo, K.; Kitamura, K. Factors affecting isoflavone content
in soybean seeds: Changes in isoflavones, saponins, and
composition of fatty acids at different temperatures during seed
development. J. Agric. Food Chem. 1995, 43, 1184-1192.
(33) Hu, J.; Lee, S.-O.; Hendrich, S.; Murphy, P. A. Quantification
of the group B soyasaponins by high-performance liquid chro-
matography. J. Agric. Food Chem. 2002, 50, 2587-2594.
(34) Hostettmann, K.; Marston, A. Saponins. In Chemistry and
Pharmacology of Natural Products; Phillipson. J. D., Ayres, D.
C., Baxter, H., Eds.; Cambridge University Press: New York,
1995; pp 298-309.
(35) Tsukamoto, C.; Kikuchi, A.; Kudou, S.; Harada, K.; Iwasaki,
T.; Okubo, K. Genetic improvement of saponin components in
soybean. In Food Phytochemicals for Cancer PreVention I:
Fruits and Vegetables; Huang, M. T., Osawa, T., Ho, C. T.,
Rosen, R. T., Eds.; American Chemical Society: Washington,
DC, 1991; pp 372-379.
(36) Anthony, M. S.; Clarkson, T. B.; Hughes, C. L., Jr.; Morgan, T.
M.; Burke, G. L. Soybean isoflavones improve cardiovascular
risk factors without affecting the reproductive system of peri-
pubertal rhesus monkeys. J. Nutr. 1996, 126, 43-50.
(37) Jenkins, D. J. A.; Kendall, D. W. C.; Jackson, C.-J. C.; Connelly,
P. W.; Parker, T.; Faulkner, D.; Vidgen, E.; Cunnane, S. C.;
Leiter, L. A.; Josse, R. G. Effects of high- and low-isoflavone
soyfoods on blood lipids, oxidized LDL homocysteine, and blood
pressure in hyperlipidemic men and women. Am. J. Clin. Nutr.
2002, 76, 365-372.
(38) Zava, D.; Duwe, G. Estrogenic and antiproliferative properties
of genistein and other flavonoids in human breast cancer cells
in vitro. Nutr. Cancer 1997, 27, 31-40.
(39) Jackson, C.-J. C.; Rupasinghe, H. P. V. Human dietary sources
of phytoestrogens and methods, Food sources and composition
of phytoestrogens. In Phytoestrogens and Health; Gilani, G. S.,
Anderson, J. J. B., Eds.; AOCS Press: Champaign, IL, 2002; pp
Received for review April 11, 2003. Revised manuscript received July
23, 2003. Accepted August 5, 2003. This research was funded by the
Ontario Soybean Growers (OSG), Ontario Ministry of Agriculture and
Food (OMAF), and Natural Science and Engineering Research Council
of Canada (NSERC).
5894 J. Agric. Food Chem ., Vol. 51, No. 20, 2003Rupasinghe et al.