ISOFLAVONE CONTENTS IN UHT AND FRESH SOYMILK AVAILABLE IN AMPHUR MUANG, CHIANG MAI

Chiang Mai Med Bull 2004;43(4):151-161.
Original article
ISOFLAVONE CONTENTS IN UHT AND FRESH SOYMILK
AVAILABLE IN AMPHUR MUANG, CHIANG MAI
Supanimit Teekachunhatean, M.D. Noppamas Rojanasthien, M.D.,
Chaichan Sangdee, Ph.D.
Department of Pharmacology, Faculty of Medicine, Chiang Mai University
Abstract
Objective To determine the contents of isoflavones in the form of aglycones (daidzein,
genistein) as well as respective ?-glycosides (daidzin, genistin), and their presence in
UHT and fresh soymilk available in Amphur Muang, Chiang Mai, Thailand.
Methods Ten samples of ultra high temparature (UHT) processed soymilk and 20
samples of fresh soymilk were randomly purchased from different zones. Isoflavone
contents in each sample were determined by high performance liquid chromatography.
Results Isoflavone contents in both soy beverages ranged from approximately 10-70
mg/serving, and ?-glycosides dominated. The median concentration of aglycones in
fresh soymilk was significantly greater than that in UHT soymilk, whereas, the median
values of ?-glycoside concentrations, total isoflavone concentrations and the total
isoflavones per serving (mg) in UHT soymilk were significantly greater than those in
fresh soymilk. Nonetheless, the median value of maximal absorbable isoflavones per
serving (in the unit of µmol) did not significantly differ between the two soy beverages.
Conclusion Isoflavone contents in both soy beverages varied substantially. Although the
proportions and concentrations of aglycones and ?-glycosides as well as the total
isoflavones per serving (mg) were significantly different between both soy beverages,
the maximal absorbable isoflavones per serving (µmol) did not significantly differ.
Chiang Mai Med Bull 2004;43(4):151-161.
Keywords: isoflavone contents, isoflavone concentrations, soymilk
Isoflavones, which are abundant in
structure of the potent synthetic estro-
soy, make up the most common form of
gen, diethylstilbesterol. The basic
phytoestrogens. They have a common
structural feature of Isoflavones consists
diphenolic structure that resembles the
of 2 benzene rings (A and B) linked
Address requests for reprints: Supanimit Teekachunhatean, M.D. Department of Pharmacology, Faculty of
Medicine, Chiang Mai University, Chiang Mai 50200, Thailand. E-mail:steekach@mail.med.cmu.ac.th
Received 18 July 2004, and in revised form 4 August 2004.

---

152
Teekachunhatean S, Rojanastein N, Sangdee C.
through a heterocyclic pyrane C ring
alleviate postmenopausal symptoms(6,7)
(Fig. 1).(1) The primary isoflavones in
and prevent osteoporosis.(8-10) Moreover,
soybeans are daidzein (4',7-dihydroxyiso-
unlike traditional hormone replacement
flavone), genistein (4',5,7-trihydroxyiso-
therapy (HRT), isoflavones might asso-
flavone) and their respective ?-glyco-
ciate with decreased risk of endometrial
sides, daidzin and genistin (sugar is
and breast cancer as well as atherosclero
attached at the 7 position of the A ring).
tic diseases.(11) Because of these advan-
Much lower amounts of glycitein (7,4'-
tages, isoflavones might be recommended
dihydroxy-6-methoxyisoflavone) and its
instead of HRT as a safe, effective
glycosides, glycitin, are also present in
dietary supplement for postmenopausal
soybeans.(2) Most of the soy products
women, especially those with increased
have a total isoflavone concentration of
risk to deleterious HRT effects such as
1-3 mg/g.(3) In nonfermented soy food,
heart disease, breast cancer, stroke, blood
the isoflavones appear mostly as the
clots, gallbladder disease, weight gain,
glycoside conjugates, whereas in
bloating, irritability and nausea.(12)
fermented soy products such as miso, the
UHT soymilk, which has been pro-
aglycones dominate.(3)
cessed with a technology called Ultra
Isoflavones appear to have both estro-
High Temperature, and fresh soymilk are
genic and antiestrogenic effects depen-
the most popular soy beverages available
ding on the target tissues.(4,5) Available
in Asian countries including Thailand,
data suggest that isoflavones exert estro-
due to their low cost and high-quality
gen-like effects on the bone and blood
protein. Nonetheless, the isoflavone con-
vessels, but not on breast and endome-
tents in these products may vary sub-
trial tissue. This contrasts with estrogen,
stantially according to the quality and
which has estrogenic effects on all
amount of soybeans used as raw material
tissues. The tissue selective effects of
as well as food-processing steps.(3) In
isoflavones have led many experts to
addition, the quantification of isoflavone
classify isoflavones as natural selective
contents in these products still has not
estrogen receptor modulators (SERMs).
been widely investigated in Thailand.
Several lines of scientific evidence also
Therefore, the purpose of this study was
demonstrate that isoflavones may help to
to determine the contents of isoflavones
Figure 1. Structures of Isoflavones. Left: Aglycones (daidzein, genistein and glycitein). Right:
Respective glycoside conjugates of aglycones (including ?-glycosides, 6"-O-acetylglucosides and 6"-
O-malonylglucosides of aglycones). Different side-chains are attached to R1-R5.(1)

---

Isoflavone contents in soymilk
153
Figure 2. Chromatograms of isoflavones in soy beverage.
(A) Quantification of daidzein (k=6.80 min) and genistein (k=11.11 min) using fluorescein
(k=13.38) as the internal standard.
(B) Quantification of daidzin (k=7.348 min) and genistin (k=13.07 min) using
chloramphenicol (k=17.64) as the internal standard.
in the form of aglycone (daidzein and
Methods
genistein) as well as their respective ?-
Study design
glycosides (daidzin and genistin) in UHT
This was a crossectional study to
and fresh soymilk available in Amphur
determine the contents of isoflavones in
Muang, Chiang Mai, Thailand. The
UHT and fresh soymilk randomly pur-
results from this study would be used as
chased in Amphur Muang, Chiang Mai.
a guideline to determine the daily amount
of soy beverages sufficient to benefit
postmenopausal health.

---

154
Teekachunhatean S, Rojanastein N, Sangdee C.
Samples for quantification of isofla-
mL/min and the analytes were detected
vones
by UV absorption at 259 nm. Chromato-
A total of 30 samples (10 and 20 sam-
grams of isoflavones and their retention
ples of UHT and fresh soymilk, respec-
times are demonstrated in Fig. 2. The
tively) were randomly purchased from
Isoflavone contents of unknown samples
different zones in Amphur Muang, Chiang
were determined using a calibration curve
Mai. The characteristics, including some
of the peak height ratios of isoflavones
of the declared nutritional information of
and IS versus respective isoflavone con-
UHT soymilk used in this study, are pre-
centrations with use of linear regression.
sented in Table 1. Zones and quantity of
All samples were analyzed within the
randomly purchased fresh soymilk sam-
same day, in which intraday assay
ples are shown in Table 2.
validation was performed. Each sample
was analyzed by a single measurement.
Quantification of isoflavones in soy
beverage
Quantification of daidzein and geni-
The sample extraction and determined
stein
concentration of isoflavones in both
The mobile phase consisted of 5 mM
UHT and fresh soymilk were modified
phosphoric acid in methanol/acetonitrile
from the method described by Nakamura
(100:85, v/v). Standard aglycones (daid-
et al.(13) Briefly, 1 mL of soy beverage
zein and genistein) were spiked in serial
was mixed with 9 mL of 80% methanol
dilution to obtain a standard calibration
in water and sonicated for 30 min. Then,
curve of 32.5, 75, 150, 300, 600 and
1 mL of the mixture was centrifuged,
1,200 ng/mL. Five quality control (QC)
and 50 µL of clear supernatant was
samples, spiked at 3 different concentra-
mixed with 30 µL of mobile phase and
tions of daidzein and genistein (100, 500
spiked with 20 µL of internal standard
and 1,000 ng/mL), were analyzed for
(IS, 100,000 ng/mL of fluorescein and
intraday assay validation. The average
50,000 ng/mL of chloramphenicol for
coefficient of intraday assay validation
the quantification of acyclones and ?-
variations for daidzein and genistein was
glycosides, respectively). Five µL of the
0.85% and 1.64%, respectively.
mixture was injected onto a high perfor-
mance liquid chromatography (HPLC)
Quantification of daidzin and genistin
system. The samples were analyzed on a
The mobile phase consisted of 5 mM
C18 column (Inertsil, 150 mm x 4.6 I.D.,
phosphoric acid in methanol/acetonitrile
5 µm, GL Science, Tokyo, Japan) with a
(80:20, v/v). Standard ?-glycosides (daid
C18 guard column (Inertsil ODS-3, 10
zin and genistin) were spiked in serial
mm x 4.0 I.D., 5 µm, GL Science, Tokyo,
dilution to obtain a standard calibration
Japan). Separation was performed isocra-
curve of 125, 250, 500, 1,000, 2,000 and
tically at 50 oC. The flow rate of mobile
4,000 ng/mL. Samples containing daid-
phase (see below) was maintained at 1
zin or genistin in excess of 4,000 ng/mL

---

Isoflavone contents in soymilk
155

---

156
Teekachunhatean S, Rojanastein N, Sangdee C.
beverages varied substantially from ap-
Table 2. Zones and quantity of randomly pur-
chased fresh soymilk samples.
proximately 10-70 mg per serving. The 5
Zones
Quantity of
samples with the highest total isofla-
sample(s)
vones per serving were VSY (67.94 mg),
Chiang Mai Gate Market
3*
VM-J (57.88 mg), fresh soymilk from
Ton Phayom Market
3*
Ton Lamyai Market (48.61 mg), LS-J
Tanin Market
3*
(37.96 mg) and LS-Bot (33.29 mg). On
Chang Phuak Market
3*
the other hand, the 5 samples with the
Warorot Market
2*
lowest total isoflavones per serving were
Ton Lamyai Market
2*
VM (8.73 mg), fresh soymilk from
Muang Mai Market
2*
San Pa Khoi Market
1
Chiang Mai Gate Market (9.58 mg),
In front of Chiang Mai
1
fresh soymilk from San Pa Khoi market
University
(11.02 mg), VM-Bot (11.49 mg) and
Total
20
fresh soymilk from Tanin market (11.67
*From different vendors in the same zone.
mg). Isoflavones in the form of ?-glyco-
sides were present in a greater amount
were analyzed after diluting with concen-
than aglycones in both soy beverages.
trations within the calibration range of
The proportions of aglycones : ?-glyco-
the assay. Five QC samples spiked at 3
sides were 1:14 and 1:2 in UHT and
different concentrations of daidzein and
fresh soymilk, respectively. However,
genistein (375, 1,500 and 3,000 ng/mL)
the median concentration of aglycones
were analyzed for intraday assay valida-
(daidzein and genistein) in fresh soymilk
tion. The average CV of intraday assay
was significantly greater than that in
for daidzin and genistin was 0.58% and
UHT soymilk. On the contrary, the
0.69%, respectively.
median value of ?-glycoside concentra-
tions (daidzin and genistin), the total
Statistical analysis
isoflavone concentrations and the total
The concentrations and contents of
isoflavones per serving (mg) in UHT
isoflavones in UHT and fresh soymilk
soymilk were significantly greater than
were presented as median (range). The
those in fresh soymilk. Nonetheless, the
differences in the median values of these
median values of maximal absorbable
variables between the two soy beverages
isoflavones per serving (in the unit of µ
were analyzed by using Wilcoxon’s rank
mol) did not significantly differ between
sum test.
the two soy beverages.
Result
Discussion
The concentrations and contents of
In this study, we Initially tried to
isoflavones in UHT and fresh soymilk
measure the total isoflavone contents in
are shown in Table 3 and Fig. 3. The
each preparation by using the acid hy-
total isoflavone contents in both soy
drolysis method.(3,14) The principle of

---

Isoflavone contents in soymilk
157
Table 3. Concentrations and contents of isoflavones in randomized samples of UHT and fresh
soymilk.
Variables
UHT soymilk
Fresh soymilk
(n=10)
(n=20)
Concentrations of aglycones (mg/mL)
Daidzein
0.004
0.008**
(0.002-0.007)
(0.003-0.016)
Genistein
0.003
0.011**
(0.001-0.006)
(0.004-0.018)
Summation of daidzein and genistein
0.007
0.019**
(0.003-0.012)
(0.007-0.034)
Concentrations of ?-glycosides (mg/mL)
Daidzin
0.04
0.01**
(0.01-0.12)
(0.004-0.04)
Genistin
0.06
0.03**
(0.02-0.17)
(0.01-0.08)
Summation of daidzin and genistin
0.10
0.04**
(0.03-0.29)
(0.02-0.13)
Concentration of total isoflavones1 (mg/mL)
0.10
0.06*
(0.04-0.30)
(0.03-0.16)
Proportion of aglycones : ?-glycosides
1:14
1:2
Total isoflavones per serving2 (mg)
30.09
18.86*
(8.73-67.94)
(9.58-48.61)
Maximal absorbable isoflavones per serving3 (µmol)
74.51
52.93
(22.07-160.73)
(27.75-127.04)
Data represent median (range). **p<0.01, *0.01<p<0.05 compared to median values of UHT soymilk.
1Summative concentrations of daidzein, genistein, daidzin and genistin, 2Calculation from the
concentration of total isoflavones (mg/ml) x serving size (exact volume for UHT soymilk or 300 ml
for fresh soymilk). 3Summation of each isoflavone content divided by its respective molecular weight.
g

l
o
)
1 8 0
m
g)
(u

(
m
g
1 6 0
r

s
e
r
v
i
n
i
n
rv
1 4 0
se
s
pe
r

pe
1 2 0
es
n
o
1 0 0
v
s
o
f
l
a
v
one
a
fl
i
so
8 0
l
e
i
l
e
b
6 0
r
bab
r
ba
s
o
o
4 0
abs
l

a
b
l
a
a
2 0
m
xi
a
a
x
i
m
M
0
M
UHT soymilk Fresh soymilk
0
1
2
3
(n=10) (n=20)
Figure 3. Maximal absorbable isoflavones per serving in randomized samples of UHT and fresh soymilk.
Horizontal bar represents median value.

---

158
Teekachunhatean S, Rojanastein N, Sangdee C.
this assay is to hydrolyze ?-glycoside
are readily hydrolyzed by human intesti-
conjugates of isoflavones to aglycones,
nal or gut microflora glucosidases to
and detect the total aglycones that reflect
release highly absorbable aglycones.(17)
the total isoflavones. Unfortunately, we
Irrespective of the relative proportions of
failed to measure the total isoflavones as
individual isoflavone conjugates in dif-
aglycones after acid hydrolysis. The
ferent soy products, the aglycones are
recovery of daidzein and genistein was
ultimately released, and it has been
very low (approximately 20-30%) as
reported that isoflavone aglycones, ?-gly
compared to the other published data.
cosides, and glycosides with different
However, since isoflavones are present
side chains are similar in bioavailability.
predominantly as ?-glycoside conjugates
(18) Although this present study revealed
(e.g. daidzin and genistin) in most
that the median total isoflavones per
commercially available soy products(3)
serving (mg) in UHT soymilk was signi-
(with the exception of fermented soy
ficantly greater than that in fresh
products), we used a specific HPLC con-
soymilk, this could not directly interpret
dition for measuring daidzin and genis-
that UHT soymilk would provide more
tin, and another for measuring aglycones
absorbable isoflavones after ingestion.
(daidzein and genistein) without acid
Because the ?-glycosides were present in
hydrolysis. Isoflavones in other forms
greater proportion in UHT soymilk than
(malonyl glycoside and acetyl glycoside
in fresh soymilk, the larger amount
conjugates) were not determined because
might be due in part to the weight of
their commercial standards were not
sugar moiety. Therefore, in order to
available. Glycitein and its derivatives
determine the amount of maximal absorb-
were also not determined, due to far less
able isoflavones in both soy beverages,
being found in soybeans.(2,15,16)
the molecular weights of aglycones and
This study demonstrated that the
their respective ?-glycosides were used
major forms of isoflavones present in
to calculate the total amount of isofla-
both UHT and fresh soymilk were ?-
vones per serving in the unit of µmol. In
glycosides. These data are consistent with
this aspect, the median values of maximal
other previous studies.(3,15,16) However,
absorbable isoflavones (µmol) between
significant differences in the proportion
the two soy beverages did not differ
and concentrations of aglycones and ?-
significantly. However, this insignificance
glycosides, as well as the total isoflavo-
might result from inadequate power of
nes per serving between these two soy
the test, due to a small sample size.
beverages, might be the result of dif-
It has been suggested that a daily
ferent raw materials and conditions used
intake of soy isoflavone extract contain-
in the processing and preparation such as
ing 50/50 mg of genistin and daidzin(6)
heat processing and fermentation.(3)
or 76 mg of isoflavone(7) can significantly
After ingestion, aglycones are readily
decrease hot flushes in the group treated
absorbed, whereas, glycoside conjugates
with soy products over the placebo. A

---

Isoflavone contents in soymilk
159
meta-analysis of 38 clinical trials that
centrations, the daily volume of con-
examined the relationship between soy
sumption would be lessened. Our study
protein intake and serum lipids have
demonstrated that the isoflavone contents
shown that the consumption of soy in
in the samples of both UHT and fresh
men and women is associated with a
soymilk varied substantially, and the
significant decrease in serum cholesterol,
total isoflavone contents in some UHT
LDL and triglyceride levels.(19) In a
soymilk (produced by the same manufac-
randomized, double-blind, placebo-con-
turing company using similar trade
trolled trial that examined the effects of
names) varied widely from approximate-
dietary soy supplements containing 118
ly 10-70 mg/serving. These data indicate
mg of isoflavones on the lipid profiles of
that raw material (soybeans) as well as
men and postmenopausal women with
conditions of food processing and prepa-
relatively normal cholesterol levels, the
ration greatly influence the variation of
LDL/HDL ratio decreased in the isofla-
isoflavone contents in soy beverages (as
vone treatment groups without any
mentioned above). Therefore, quantifica-
change in total cholesterol.(20) In addition,
tion should be repeated further to deter-
isoflavone supplementation (61.8 mg of
mine the consistency of isoflavone con-
isoflavones) for four weeks showed
tents in the soy beverages randomly
potentially beneficial effects on bone
collected at different time-points (or a
metabolism and serum lipids in perime-
different lot, in the case of UHT soy-
nopausal women in the randomized
milk).
controlled trial.(9) Another trial demon-
However, these inexpensive soy be-
strated that a continuous dietary intake of
verages might be considered an appro-
isoflavones (37.3 mg/day) for ten weeks
priate alternative food supplement to the
might inhibit postmenopausal osteoporo-
more expensive soy extract capsules and
sis.(10) Therefore, in our opinion, the total
HRT for Thai postmenopausal women,
dose of isoflavones that benefits meno-
especially those with increased risk to
pausal health is up to approximately 100
the deleterious effects of HRT. However,
mg/day. Since our pilot study revealed
since a different proportion of daidzein,
that the bioavailability of genistein and
genistein and their ?-glycosides in
daidzein was approximately the same as
different soy products might affect the
that for soy extract capsules (unpublished
beneficial effects for postmenopausal
data), the amount of soy beverage that
women, clinical studies in this aspect
provides an isoflavone bioavailability
should be investigated further.
equivalent to 100 mg of soy isoflavones
should be equal to approximately 3
Conclusion
servings/day of UHT soymilk, or 5
Isoflavone contents in both soy be-
servings/day of fresh soymilk. If one
verages varied substantially. Although
consumes other soy food or chooses a
the proportions and the median concen-
soy beverage with higher isoflavone con-
trations of aglycones and ?-glycosides

---

160
Teekachunhatean S, Rojanastein N, Sangdee C.
were significantly different between both
J Clin Nutr 1999;68(suppl):1375S-9S.
soy beverages, the median value of maxi-
9. Uesugi T, Fukui Y, Yamori Y. Beneficial
mal absorbable isoflavones per serving
effects of soybean isoflavone supplementa-
tion on bone metabolism and serum lipids in
(µmol) did not significantly differ.
postmenopausal Japanese women: a four-
week study. J Am Coll Nutr 2002;21:97-102.
Acknowledgements
10. Yamori Y, Moriguchi EH, Teramoto T, et al.
The authors would like to acknowledge Asso-
Soybean isoflavones reduce postmenopausal
ciate Professor Dr. Saipin Pongsatha, Depart-
bone resorption in female Japanese immi-
ment of Obstetrics and Gynecology, Faculty of
grants in Brazil: a ten-week study. J Am
Medicine, Chiang Mai University, for being
Coll Nutr 2002;21:560-7.
consultant of this research.
11. Mackey R, Eden J. Phytoestrogens and the
menopause. Climacteric 1998;1:302-8.
12. Warren MP. A comparative review of the
References
risks and benefits of hormone replacement
1. Miksicek RJ. Commonly occurring plant
therapy regimens. Am J Obstet Gynecol
flavonoids have estrogenic activity. Mol
2004;190:1141-67.
Pharmacol 1993;44:37-43.
13. Nakamura Y, Tsuji S, Tonogai Y. Determina-
2. Wang H, Murphy PA. Isoflavone composi-
tion of the levels of isoflavonoids in soybeans
tion of American and Japanese soybeans in
and soy-derived foods and estimation of
Iowa: effects of variety, crop year, and
isoflavonoids in the Japanese daily intake. J
location. J Agric Food Chem 1994;42:1674-7.
AOAC Int 2000;83:635-50.
3. Wang H, Murphy PA. Isoflavone content in
14. Barnes S, Kirk M, Coward L. Isoflavones
commercial soybean foods. J Agric Food
and their conjugates in soy foods: extraction
Chem 1994;42:1666-73.
conditions and analysis by HPLC-mass
4. Martin PM, Horwitz KB, Ryan DS, McGuire
spectrometry. J Agric Food Chem 1994;42:
WL. Phytoestrogen interaction with estrogen
2466-74.
receptors in human breast cancer cells.
15. Wiseman H, Casey K, Clarke DB, Barnes
Endocrinology 1978;103:1860-7.
KA, Bowey E. Isoflavone aglycon and
5. Kuiper GG, Lemmen JG, Carlsson B, et al.
glucoconjugate content of high- and low-soy
Interaction of estrogenic chemicals and
U.K. foods used in nutritional studies. J
phytoestrogens with estrogen receptor beta.
Agric Food Chem 2002;50:1404-10.
Endocrinology 1998;139:4252-63.
16. Coward L, Smith M, Kirk M, Barnes S.
6. Upmalis DH, Lobo R, Bradley L, Warren M,
Chemical modification of isoflavones in
Cone FL, Lamia CA. Vasomotor symptom
soyfoods during cooking and processing.
relief by soy isoflavone extract tablets on
Am J Clin Nutr 1998;68(suppl):1486S-91S.
postmenopausal women: a multicenter,
17. Xu X, Harris KS, Wang HJ, Murphy PA,
double-blind, randomized, placebo-controlled
Hendrich S. Bioavailability of soybean
study. Menopause 2000;7:236-42.
isoflavones depends upon gut microflora in
7. Albertazzi P, Pansini F, Bonaccorsi G,
women. J Nutr 1995;125:2307-15.
Alakel L. The effect of dietary soy supple-
18. Xu X, Wang HJ, Murphy PA, Hendrich S.
mentation on hot flushes. Obstet Gynecol
Neither background diet nor type of soy
1998;91:6-11.
food affect short-term isoflavone bioavaila-
8. Potter SM, Baum JA, Teng H, Stillman RJ,
bility in women. J Nutr 2000;130:798-801.
Shay NF, Erdman JW Jr. Soy Protein and
19. Anderson J, Johnston B, Cook-Newell M.
isoflavones: their effects on blood lipids and
Meta-analysis of the effects of soy protein
bone density in postmenopausal women. Am
intake on serum lipids. N Engl J Med 1995;

---