You are on page 1of 7

European Journal of Clinical Nutrition (2008) 62, 14191425

& 2008 Macmillan Publishers Limited All rights reserved 0954-3007/08 $32.00
www.nature.com/ejcn

ORIGINAL ARTICLE

Consumption of isoflavone-rich soy protein does not


alter homocysteine or markers of inflammation in
postmenopausal women
KA Greany1, JA Nettleton1, KE Wangen1, W Thomas2 and MS Kurzer1
1
Department of Food Science and Nutrition, University of Minnesota, St Paul, MN, USA and 2Department of Biostatistics, University of
Minnesota, Minneapolis, MN, USA

Background/Objective: To investigate the effect of soy protein containing isoflavones on homocysteine (Hcy), C-reactive
protein (CRP), soluble E-selectin (sE-selectin), soluble vascular adhesion molecule-1 (sVCAM-1) and soluble intercellular adhesion
molecule-1 (sICAM-1).
Subject/Methods: In a randomized crossover design, 34 postmenopausal women consumed soy protein isolate (2675 g
protein containing 4478 mg isoflavones per day) or milk protein isolate (2675 g protein per day) for 6 weeks each. Fasting
blood samples were collected at the end of each diet period and end points analyzed by enzyme-linked immunosorbent assay.
Results: Concentrations of Hcy, CRP, sE-selectin, sVCAM-1 and sICAM-1 were not different between soy and milk diet
treatments. Results did not differ by equol production status or by baseline lipid concentration. Adjustment for intake of folate
and methionine did not alter the Hcy results.
Conclusions: These data suggest that decreasing vascular inflammation and Hcy concentration are not likely mechanisms by
which soy consumption reduces coronary heart disease risk.

European Journal of Clinical Nutrition (2008) 62, 14191425; doi:10.1038/sj.ejcn.1602885; published online 5 September 2007
Keywords: soy; isoflavone; homocysteine; inflammation; postmenopausal women

Introduction
Consumption of soy foods has been associated with a
reduced risk of coronary heart disease (CHD) in women
(Zhang et al., 2003). However, the cardiovascular risk factors
altered by soy consumption and mechanisms involved have
yet to be conclusively established. Historically, the majority
of research has focused on soy and plasma lipids, but as the
paradigm of atherogenesis has shifted from a disease of
simple lipid accumulation to a complex condition of chronic
Correspondence: Dr MS Kurzer, Department of Food Science and Nutrition,
University of Minnesota, 1334 Eckles Avenue, St Paul, MN 55108, USA.
E-mail: mkurzer@umn.edu
Guarantor: MS Kurzer.
Contributors: KAG was the primary author, assisted with subject recruiting and
study coordination, and performed the lipid and homocysteine analyses; JAN
assisted with subject recruiting and study coordination; KEW assisted with
study design; WT assisted with study design and statistical analysis; MSK was
the principal investigator and supervised all aspects of the study. All authors
contributed to the data interpretation and manuscript preparation.
Received 29 January 2007; revised 30 May 2007; accepted 20 July 2007;
published online 5 September 2007

inflammation and vascular dysfunction, additional risk


factors (that is, homocysteine (Hcy) and markers of vascular
inflammation) have emerged as potential targets for soy.
Elevated concentrations of Hcy (Boushey et al., 1995;
Ridker et al., 1999), C-reactive protein (CRP) (Ridker et al.,
1998, 2000) and vascular adhesion molecules (Hwang et al.,
1997; Ridker et al., 2000) such as soluble E-selectin (sEselectin), soluble vascular adhesion molecule-1 (sVCAM-1)
and soluble intercellular adhesion molecule-1 (sICAM-1)
have been associated with increased risk of CHD. Estrogen
replacement therapy and the selective estrogen receptor
modulator (SERM) raloxifene lower plasma Hcy (Walsh et al.,
2000; Yildirir et al., 2002). Estrogen, but not raloxifene,
increases circulating CRP concentration (de Valk-de Roo
et al., 1999; Walsh et al., 2000; Yildirir et al., 2002) by
increasing hepatic production of CRP (Walsh et al., 2001;
Skouby et al., 2002). Both estrogen and raloxifene have been
reported to decrease vascular adhesion molecules (Koh et al.,
1999; Blum et al., 2000) by binding to estrogen receptors on
the vascular endothelium and modulating transcription of
these adhesion molecules (Caulin-Glaser et al., 1996).

Soy, homocysteine and inflammation


KA Greany et al

1420
Soy isoflavones are capable of binding to the estrogen
receptor and have a similar affinity for the b-isoform
(predominant form in vascular tissue) as estradiol (Gruber
et al., 2002). By mimicking the effect of a SERM, soy
isoflavones may favorably alter Hcy, CRP and adhesion
molecules. The isoflavone metabolite equol possesses higher
affinity than its precursor daidzein (Gruber et al., 2002);
therefore, the ability to produce equol (present in one-third
of the population) may result in an augmented effect (Lampe
et al., 1998; Xu et al., 2000; Setchell et al., 2002). In addition
to the isoflavones, soy protein contains other components,
such as phytate and soy peptides, which may also exert an
effect on CHD risk factors.
The hypothesis of this study was that consumption of
isoflavone-containing soy protein isolate would alter these
CHD risk factors in a pattern resembling the actions of a
SERM, that is, a decrease in Hcy, sE-selectin, sVCAM-1 and
sICAM-1 without an increase in CRP, and that these
alterations would be amplified in equol producers.

Subjects and methods


Study design
The present study was part of a larger project investigating
the effects of soy and probiotic consumption on plasma
lipids, hormones and isoflavones in postmenopausal women
with and without a history of breast cancer (Greany et al.,
2004; Nettleton et al., 2004, 2005). The principal study
incorporated four 6-week diet treatments separated by
2-week washout periods in a randomized crossover design;
the length of the washout was considered adequate for
preventing residual effects protein isolate or probiotics on
subsequent diet treatments while maximizing subject retention in the study. Samples representing the diet periods with
probiotic consumption were not included in the present
analysis of inflammation markers and Hcy. Participants were
free-living and consumed their habitual diets supplemented
by soy or milk protein isolate (both by Solae, St Louis, MO,
USA). Soy and milk protein isolates were distributed relative
to body weight and provided 0.38 g protein per kg body
weight per day (2675 g day1). Soy protein isolate contained
1.16 mg isoflavones g1 powder (57% genistein, 34% daidzein and 9% glycitein; analyzed by Dr Pat Murphy, Department of Food Science and Human Nutrition, Iowa State
University, Ames, IA, USA) and provided 0.64 mg total
isoflavones per kg body weight per day (4478 mg day1
isoflavones expressed as aglycone equivalents). Subjects were
instructed to consume their daily allotment of protein
powder divided in two doses; suggested uses for the powder
were mixing with water, juice, milk, hot cocoa or adding to
soft foods such as oatmeal. The importance of substituting
the powder mixture for other foods to avoid weight gain was
emphasized. In addition, subjects were required to exclude
soy food products, flaxseed, alcoholic beverages, herbal or
isoflavone supplements, and vitamin/mineral supplements
European Journal of Clinical Nutrition

containing greater than 100% of the daily value for any


nutrient from their habitual diet and to limit intake of
legumes to a single serving per week. Subjects recorded their
powder consumption and any deviations from study restrictions on individualized calendars.

Subjects
The 34 subjects included in this study were obtained via a
pool of postmenopausal women recruited from the Minneapolis/St Paul metropolitan area by local newspapers and
community advertisement. Respondents were screened by
telephone, in-person interview, and health assessment to
verify they met the following inclusion criteria: absence of
menstrual bleeding for past 12 months, follicle-stimulating
hormone concentration 435 IU l1, no hormonal therapy
(hormone replacement, raloxifene, topical progesterone
cream, steroids, thyroid replacement) or oral antibiotics
within the past 3 months, no current use of hypocholesterolemic agents, body mass index 1836 kg m2, weight
change o6.8 kg during past year and no current tobacco
use. The study was approved by the Institutional Review
Boards of the University of Minnesota and the US Army
Medical Research and Materiel Command. All subjects gave
written informed consent before participating in the study.

Data collection and analysis


Fasting blood samples were drawn between 0600 and 1000
hours on day 1 of the first diet period (baseline) and at the
completion of each diet period. Blood for serum was drawn
into tubes containing gel clot activator, incubated at room
temperature for 1030 min and centrifuged for 10 min at 4 1C
and 2000 g. Blood for plasma Hcy was drawn into tubes
containing EDTA, centrifuged at 15 1C and 2000 g for 10 min,
and aprotinin and sodium azide added to final concentrations of 1 mg l1 and 1 g l1, respectively. Blood for plasma
isoflavones was drawn into tubes containing heparin,
centrifuged at 15 1C and 2000 g for 10 min and sodium azide
and ascorbic acid added to final concentrations of 1 g l1
each. All aliquots were frozen at 70 1C until analysis. Before
the isoflavone assays, designated plasma was thawed,
divided into aliquots and lyophilized for analysis.
Serum samples were analyzed in duplicate in a single run
for sE-selectin, sVCAM-1 and sICAM-1 using enzyme-linked
immunosorbent assays (R&D Systems, Minneapolis, MN,
USA). Intra-assay variability was 5, 6 and 3%, respectively.
Serum samples for CRP were analyzed in duplicate in a single
run using enzyme-linked immunosorbent assay (ALPCO
Diagnostics, Windham, NH, USA). Intra-assay variability
was 5%. All samples from each subject were assayed in
duplicate on the same day for plasma total Hcy using
enzyme-linked immunosorbent assays (Diazyme Laboratories, San Diego, CA, USA). Intra-assay variability was 4%
and interassay variability was 13%.

Soy, homocysteine and inflammation


KA Greany et al

1421
Plasma isoflavones (genistein, daidzein, equol and Odesmethylangolensin) were analyzed by competitive timeresolved fluoroimmunoassay as described previously (Wang
et al., 2000; LHomme et al., 2002; Brouwers et al., 2003). Intraassay variability was 6.3, 7.1 and 8.1% and interassay variability
was 9.5, 9.9 and 15.2% for genistein, daidzein and equol,
respectively. Plasma O-desmethylangolensin was analyzed in a
single batch with intra-assay variability of 15.8%.
Subjects collected three 24-h urine samples on the final 3
days of each diet treatment; volume of each collection was
recorded, sodium azide added to a final concentration of 1 g
l1 and samples frozen at 20 1C. Before urinary equol
analysis, urine was thawed, pooled proportionally for a 3-day
sample, aliquoted and lyophilized for equol quantification.
Urinary equol was analyzed by competitive time-resolved
fluoroimmunoassay and a gas chromatographymass spectrometry conversion factor applied to these values as
described previously (Brouwers et al., 2003).
Subjects completed detailed diet records for the 3 days
before the first diet period (baseline) and for the final 3 days
of each diet period. Food records were analyzed using
Nutritionist V version 2.1 (First DataBank, Inc., Hearst
Corporation, San Bruno, CA, USA).

Statistical analysis
Statistical analysis was performed with SAS version 8.2. (SAS
Institute Inc., Cary, NC, USA) and SPSS version 12.0 (SPSS Inc.,
Chicago, IL, USA). Differences in Hcy and inflammation
markers between soy and milk treatments and the interactions of treatment with equol production status (high/low),
baseline lipid status (hypercholesterolemic/normocholesterolemic) and plasma isoflavone group (above/below median)
were assessed with repeated measures analysis of variance
with group as a between-subjects factor. Paired t-tests were
used to compare energy, nutrient and isoflavone intake
between soy and milk treatments. Pearsons correlation
coefficients were calculated to determine the association
between total plasma isoflavone concentration and soy-milk
difference for all variables. Spearmans correlation coefficients
were used to assess the relationship between urinary and
plasma equol. Data are presented as means7standard deviation (s.d.). P-values less than 0.05 were considered significant.

Results
Subjects
There were no differences in baseline characteristics or
response to diet treatment between women with and without a history of breast cancer; therefore, these groups were
combined for all analyses. Baseline characteristics of the 34
subjects included in this study are shown in Table 1. Twentytwo women (65%) were considered hypercholesterolemic
(total cholesterol, TC45.18 mmol l1). Based on subject
written records, telephone and in-person conversations with

Table 1

Clinical characteristics of subjects at baselinea

Age (years)
Years since menopause
Weight (kg)
BMI (kg m2)
Total cholesterol (mmol l1)
HDL-C (mmol l1)
LDL-C (mmol l1)
Triacylglycerol (mmol l1)
Hcy (mmol l1)
CRP (mg l1)
sE-selectin (ng ml1)
sVCAM-1 (ng ml1)
sICAM-1 (ng ml1)

Value (n 34)

Range

57.776.0
9.276.1
68.4711.6
25.074.3
5.1870.076
1.2670.23
3.4670.64
1.0070.41
10.473.2
2.673.3
37.6719.6
8547176
211740

4769
120
46.495.5
18.637.3
3.406.57
0.711.91
2.384.78
0.422.16
2.217.1
0.111.4
12.4110.8
6021486
153322

Abbreviations: BMI, body mass index; CRP, C-reactive protein; Hcy, homocysteine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; Eselectin, soluble E-selectin; sICAM-1, soluble intercellular adhesion molecule1; sVCAM-1, soluble vascular adhesion molecule-1.
a
Data expressed as mean7s.d.

subjects, and changes in plasma isoflavone concentrations,


adherence to the study protocol was good. The protein
isolates were well tolerated, as assessed by responses to
nonleading questions at study visits and mid-period telephone conversations.

Dietary intake
There were no differences in energy or macronutrient intake
between soy and milk diet treatments based on 3-day food
records completed during the study (Table 2). However, as a
result of the inherent differences in soy and milk protein
isolates, folate intake was higher during the soy treatment
and methionine intake was higher during the milk treatment. Adjusting values of Hcy for these differences did not
alter the results; therefore, unadjusted means are reported for
all variables.

Equol production
Based on visual identification of a clear demarcation in the
urinary and plasma isoflavone data, high-equol producers
were defined as women with plasma equol concentration415 nmol l1 and urinary equol excretion41500 nmol
day1 while consuming the soy treatment. Plasma and
urinary equol concentrations were highly correlated
(r 0.75, Po0.0001) and both methods of classification
established the same subjects as high producers. Six women
(18%) in the present study were considered high-equol
producers and twenty-eight women low-equol producers.

Hcy and inflammatory markers


Hcy, CRP, sE-selectin, sVCAM-1 and sICAM-1 were not
different between soy and milk treatments (Table 3). There
European Journal of Clinical Nutrition

Soy, homocysteine and inflammation


KA Greany et al

1422
Table 2 Daily energy and nutrient intakes during soy and milk treatmentsa
Soy treatment (n 34)

Energy (MJ day1)


Protein (g day1)
Carbohydrate (g day1)
Total fat (g day1)
Saturated fat (g day1)
Cholesterol (mg day1)
Folate (mg day1)
Vitamin B6 (mg day1)
Vitamin B12 (mg day1)
Methionine (mg day1)
Total isoflavones (mg day1)
Genistein
Daidzein
Glycitein

Milk treatment (n 34)

Powder alone

Powder plus diet

Powder alone

Powder plus diet

0.5470.12
26.074.7
5.271.2
1.070.6
0.370.6
0.070.0
78.0711.7
0.070.0
1.270.6
338758
43.777.6
24.874.7
15.172.9
3.870.6

7.7171.63
91.2719.8
243747
60.8721.5
19.678.5
2187131
3667122b
1.670.9
4.071.8
14587436c
43.777.6
24.874.7
15.172.9
3.870.6

0.5470.12
26.074.7
6.571.2
0.370.6
0.270.6
0.070.0
39.075.8
0.070.0
1.270.6
8587151
ND
ND
ND
ND

7.5271.40
90.3716.9
244762
54.6718.5
17.477.7
2077108
3127108
1.670.9
4.671.8
19557460
ND
ND
ND
ND

Abbreviation: ND, not detected.


a
Data expressed as mean7s.d.
Significantly different from milk treatment: bPo0.05, cPo0.01.

Table 3 Concentrations of Hcy, CRP, sE-selectin, sVCAM-1, sICAM-1


and plasma total isoflavones after soy and milk treatmentsa

Hcy (mmol l1)


CRP (mg l1)
sE-selectin (ng ml1)
sVCAM-1 (ng ml1)
sICAM-1 (ng ml1)
Total isoflavones
(nmol l1)

Soy treatment
(n 34)

Milk treatment
(n 34)

Pb

9.5972.23
2.9573.83
35.4714.6
8237152
214738
5207341

9.4672.43
2.3172.21
35.7713.6
8167151
211733
1079

0.810
0.197
0.827
0.528
0.716

Soy diet

Abbreviations: CRP, C-reactive protein; Hcy, homocysteine; sE-selectin,


soluble E-selectin; sICAM-1, soluble intercellular adhesion molecule-1;
sVCAM-1, soluble vascular adhesion molecule-1.
a
Values are mean7s.d.
b
Comparison between soy and milk treatments.

was no interaction between baseline lipid status (hyper vs


normocholesterolemic) and treatment for any variable
(P 0.2850.930). Likewise, there was no effect of equol
production status on response to the intervention (Hcy 9.39
vs 10.54, P 0.622; CRP 2.73 vs 3.98, P 0.162; sE-selectin
34.6 vs 39.3, P 0.408; sVCAM-1 825 vs 815, P 0.900;
sICAM-1 216 vs 206, P 0.634 for low equol producers
compared to high equol producers during soy diet).
In a post hoc analysis, subjects were divided into two
groups based on plasma isoflavone concentration (above or
below the median concentration of 508 nmol l1). There was
a significant interaction (P 0.046) between plasma isoflavone group and diet on Hcy concentration (Table 4). The
differences in Hcy between soy and milk treatments were in
opposite directions for subjects above and below the median
and were correlated with total plasma isoflavone concentraEuropean Journal of Clinical Nutrition

Table 4 Concentration of Hcy after soy and milk treatments in


postmenopausal women grouped according to plasma total isoflavone
concentrationa
Milk diet Soymilk difference

Hcy (mmol l1)


High-isoflavone group (n 17) 9.172.1 10.172.2
Low-isoflavone group (n 17) 10.172.3 8.872.5

0.9773.1b
1.2473.1

Abbreviation: Hcy, homocysteine.


Women divided into two groups: high isoflavone (total plasma isoflavone
concentration above median of 508 nmol l1 after soy diet) and low isoflavone
(total plasma isoflavone concentration below median of 508 nmol l1).
a
Data expressed as mean7s.d.
b
Significantly different than the soy-milk difference for women in the lowisoflavone group (P 0.046).

tions (r 0.348; P 0.044). However, this partitioning of


subjects did not influence the results of diet on CRP,
sE-selectin, sVCAM-1 or sICAM-1. Total plasma isoflavone
concentration was not correlated with the difference between soy and milk treatments for CRP or adhesion
molecules.

Discussion
The results of this investigation suggest that while soy
isoflavones do not produce estrogens proinflammatory
increase in CRP, neither do they exert an anti-inflammatory
effect on vascular adhesion molecules. With the exception of
one study reporting an improvement in CRP (Hall et al.,
2005), a preponderance of recently published studies

Soy, homocysteine and inflammation


KA Greany et al

1423
administering soy protein isolate or isolated isoflavones
concur with the absence of effect on CRP shown in the
present study (Teede et al., 2004; DAnna et al., 2005; Hilpert
et al., 2005; Yildiz et al., 2005; Hanson et al., 2006; RyanBorchers et al., 2006). Regardless of intervention length (1
month to 3 years), vehicle of administration (soy protein
isolate or isoflavone tablets) or isoflavone dose (10129 mg),
significant effects on CRP were not observed in these studies.
Although a lowsaturated fat diet including soy foods, plant
sterols, viscous fiber and nuts demonstrated a significant
decline in CRP (Jenkins et al., 2003), an intervention
incorporating only the soy foods did not result in a
reduction (Jenkins et al., 2002a), suggesting that soy was
not responsible for the decrease in CRP. Collectively, these
studies suggest that soy isoflavones do not mimic the effect
of estrogen on hepatic CRP-synthesis and are neutral to CRP
concentrations.
Although the majority of published investigations have
failed to find a decrease in sVCAM-1 (Blum et al., 2003;
Steinberg et al., 2003; West et al., 2005), sICAM-1 (Blum et al.,
2003; Steinberg et al., 2003; Hall et al., 2005) and sE-selectin
(Blum et al., 2003; Steinberg et al., 2003; Hall et al., 2005)
following isoflavone consumption, several investigations
have reported a change. A 3-year study conducted with
female monkeys receiving 19% of kilocalorie as soy protein
and the equivalent of 129 mg isoflavones found a significant
decrease in sVCAM-1 but no change in sE-selectin (Register
et al., 2005). Eight weeks of an isoflavone-enriched cereal bar
resulted in decreased VCAM-1 in postmenopausal women
with the AA genotype for ERb AluI but not for other
genotypes (Hall et al., 2005). After consuming a 40 mg
isoflavone tablet b.i.d. for 6 weeks, subjects receiving tablets
with high-formononetin (a precursor of daidzein) but not
biochanin (a precursor of genistein) content experienced a
significant decrease in sVCAM-1 compared to placebo (Teede
et al., 2003). Isoflavones in tablet form (57 mg b.i.d. for 3
months) resulted in a decrease in sE-selectin after both
placebo and isoflavone but no difference between placebo
and active treatments (Nikander et al., 2003). However, when
subjects were stratified based on plasma daidzein and
genistein concentrations, women with isoflavone concentrations above the median exhibited a significant decrease
while those below showed no change (Nikander et al., 2003).
The mixed results of these investigations of isoflavones
and adhesion molecules highlight several important issues
that may influence study outcomes: duration, dose and
isoflavone bioavailability/bioactivity. The present study and
the three studies reporting no effects (Blum et al., 2003;
Steinberg et al., 2003; West et al., 2005) were relatively short
(6 weeks); it is possible a longer duration is required for a
decrease to be observed. These studies similarly incorporated
B25 g day1 of isoflavone-rich soy protein; a higher dose
may be needed to achieve the threshold plasma isoflavone
concentration for an effect on adhesion molecules. For
example, the median concentrations of isoflavones (particularly daidzein) in the study with concentration-dependent

results (Nikander et al., 2003) were substantially higher than


in the present study (daidzein 861 vs 74 nmol l1; genistein
363 vs 238 nmol l1). This observation (Nikander et al.,
2003), in conjunction with the effect observed with
formononetin-rich but not biochanin-rich tablets (Teede
et al., 2003) suggests that daidzein may be the primary
isoflavone responsible for modulating adhesion molecules.
Alternatively, ability to convert daidzein to equol could be a
factor in the effectiveness of the formononetin-rich preparation; however, in the present study as well as two others with
results divided by equol production (Steinberg et al., 2003;
West et al., 2005), no differences were found in the responses
of equol producers compared with nonproducers.
Although the present study did not find an effect of soy
consumption on plasma Hcy concentration in all subjects
combined, when subjects were divided based on total plasma
isoflavone concentration, soy consumption tended to decrease Hcy in subjects with isoflavone concentrations above
the median. This suggests that differences in isoflavone
bioavailability among subjects may influence the effect of
soy on Hcy. Results of similar Hcy investigations have not
been published stratified by plasma isoflavone concentration
to refute or support this theory. However, studies reporting a
change in Hcy (Puska et al., 2002; Tonstad et al., 2002)
provided 2- to 3-fold higher isoflavone doses than those
without a significant effect (DAnna et al., 2005; Roughead
et al., 2005; Reimann et al., 2006). While these investigations
suggest a potential role for isoflavones in lowering Hcy, a
recent study manipulating soy protein content of both
isoflavones and phytate contradicts this theory (Hanson
et al., 2006). In this study of postmenopausal women, soy
protein with native phytate levels reduced Hcy concentrations regardless of isoflavone level, suggesting phytate
instead of isoflavones as the responsible component (Hanson
et al., 2006).
Another alternative explanation for the potential Hcylowering effect of soy is the lower methionine content of soy
protein (1.3 g per 100 g protein) compared with milk protein
(3.3 g per 100 g). This may be particularly pertinent for
studies employing high intakes of protein (3050 g) (Puska
et al., 2002; Tonstad et al., 2002; Jenkins et al., 2002b) in
which Hcy concentrations increased during milk protein
consumption but did not change during soy consumption,
resulting in a significant difference between soy and milk
(Puska et al., 2002; Tonstad et al., 2002). This hypothesis
would also account for the lack of effect observed with 54 mg
genistein supplementation (DAnna et al., 2005), 50 mg of
isoflavones isolated from soy (Reimann et al., 2006) and
substitution of 25 g soy protein for meat protein (Roughead
et al., 2005) as these studies did not include a soy vs milk
protein comparison.
Finally, the plasma lipid concentrations of subjects may
explain variability in Hcy response to soy consumption; soy
may be more likely to influence Hcy in subjects with
elevated low-density lipoprotein cholesterol (LDL-C) concentrations. Although the baseline TC and LDL-C concenEuropean Journal of Clinical Nutrition

Soy, homocysteine and inflammation


KA Greany et al

1424
trations of subjects in the present study were not related to
the effect of soy on Hcy, these subjects were normo- or
mildly hypercholesterolemic with a mean LDL-C of
3.46 mmol l1. In contrast, the baseline LDL-C concentrations in studies demonstrating a significant effect of soy on
Hcy were substantially elevated at 4.56, 5.08 and 4.77 mmol
l1 (Puska et al., 2002; Tonstad et al., 2002; Jenkins et al.,
2002b). In hypercholesterolemia, the Hcy content of lipoproteins is proportionally higher than in a normocholesterolemic environment (Olszewski and McCully, 1991). Soy
protein consumption increases LDL receptor expression
(Baum et al., 1998; Lovati et al., 2000) and removal of Hcycontaining lipoproteins from the plasma and may thereby
augment the decrease in Hcy.
The present study examined the response of CHD risk
factors (CRP, sE-selectin, sVCAM-1, sICAM-1 and Hcy) to a
dose of isoflavone-rich soy protein deemed effective for
decreasing plasma lipids by the FDA health claim (FDA,
1999). Although the majority of relevant published studies
are in agreement with the lack of significant effects reported
here, the current study was limited in several ways. The
isoflavone content of the soy treatment was lower than that
achievable through isolated isoflavone supplements and
may not have been sufficient to exert a therapeutic effect.
The subjects were apparently healthy and exhibited concentrations of the plasma/serum markers generally within
the normal ranges. The diet consumed by the subjects was
not overtly atherogenic and in fact approximated the
American Heart Association Step I diet. These details may
have decreased the potential for observing significant
changes. In addition, the number of equol producers was
small, limiting the ability to detect differences between
producers and nonproducers.
This study demonstrated that the incorporation of a
moderate dose of isoflavone-containing soy protein isolate
into a generally healthy diet does not decrease Hcy (except
in subjects exhibiting higher concentrations of plasma
isoflavones), CRP or the vascular adhesion molecules sEselectin, sICAM-1 or sVCAM-1. Several recent studies confirm the lack of effect with this level of soy protein/
isoflavones, although the possibility of a favorable effect at
higher (pharmacologic) intakes does exist. Further research is
needed to support or refute the ability of higher doses to
impact Hcy, CRP and vascular adhesion molecules, and
elucidate the responsible mechanisms. Until additional
research corroborates the isolated findings of decreased
Hcy, CRP or adhesion molecules, health professionals and
health conscious adults should implement other approaches
to modify these CHD risk factors.

Acknowledgements
We acknowledge the staff of the University of Minnesota
General Clinical Research Center for assistance with sample
procurement and processing, the staff of the Cytokine
European Journal of Clinical Nutrition

Reference Laboratory, staff at the University of Minnesota


for assistance with analysis of CRP, sICAM-1, sVCAM-1 and
sE-selectin, and the study participants for their commitment
to the project.
This study was supported by the United States Army
Department of Defense Grant DAMD17-99-1-9297, General
Clinical Research Center Grant M01-RR00400 from the
National Center for Research Resources, the Minnesota
Agricultural Station and the Solae, St Louis, MO, USA.
Protein powders were provided by the Solae.

References
Baum JA, Teng H, Erdman Jr JW, Weigel RM, Klein BP, Persky VW
et al. (1998). Long-term intake of soy protein improves blood lipid
profiles and increases mononuclear cell low-density-lipoprotein
receptor messenger RNA in hypercholesterolemic, postmenopausal women. Am J Clin Nutr 68, 545551.
Blum A, Lang N, Peleg A, Vidger F, Israeli P, Gumanovsky M et al.
(2003). Effects of oral soy protein on markers of inflammation in
postmenopausal women with mild hypercholesterolemia. Am
Heart J 145, e7.
Blum A, Schenke WH, Hathaway L, Mincemoyer R, Csako G,
Waclawiw MA et al. (2000). Effects of estrogen and the selective
estrogen receptor modulator raloxifene on markers of inflammation in postmenopausal women. Am J Cardiol 86, 892895.
Boushey CJ, Beresford SA, Omenn GS, Motulsky AG (1995). A
quantitative assessment of plasma homocysteine as a risk factor
for vascular disease: probable benefits of increasing folic acid
intakes. JAMA 274, 10491057.
Brouwers E, LHomme R, Al-Maharik N, Lapcik O, Hampl R, Wahala K
et al. (2003). Time-resolved fluoroimmunoassay for equol in
plasma and urine. J Steroid Biochem Mol Biol 84, 577588.
Caulin-Glaser T, Watson CA, Pardi R, Bender JR (1996). Effects of 17
beta-estradiol on cytokine-induced endothelial cell adhesion
molecule expression. J Clin Invest 98, 3642.
DAnna R, Baviera G, Corrado F, Cancellieri F, Crisafulli A, Squadrito
F (2005). The effect of the phytoestrogen genistein and hormone
replacement therapy on homocysteine and C-reactive protein
level in postmenopausal women. Acta Obstet Gynecol Scand 84,
474477.
de Valk-de Roo G, Stehouwer C, Meijer P, Mijatovic V, Kluft C,
Kenemans P et al. (1999). Both raloxifene and estrogen reduce
major cardiovascular risk factors in healthy postmenopausal
women. Arterioscler Thromb Vasc Biol 19, 29933000.
FDA (1999). Food and Drug Administration final rule: food labeling
health claims; soy protein and coronary heart disease. Fed Regist
64, 5769957733.
Greany KA, Nettleton JA, Wangen KE, Thomas W, Kurzer MS (2004).
Probiotic consumption does not enhance the cholesterol-lowering
effect of soy in postmenopausal women. J Nutr 134, 32773283.
Gruber C, Tshugguel W, Schneeberger C, Huber J (2002). Production
and actions of estrogens. N Engl J Med 346, 340351.
Hall W, Vafeiadou K, Bugel S, Koebnick C, Reimann M, Ferrari M et al.
(2005). Soy-isoflavone-enriched foods and inflammatory biomarkers of cardiovascular disease risk in postmenopausal women:
interactions with genotype and equol production. Am J Clin Nutr
82, 12601268.
Hanson LN, Engelman HM, Alekel DL, Schalinske KL, Kohut ML,
Reddy MB (2006). Effects of soy isoflavones and phytate on
homocysteine, C-reactive protein, and iron status in postmenopausal women. Am J Clin Nutr 84, 774780.
Hilpert KF, Kris-Etheron P, West CE (2005). Lipid response to a low-fat
diet with or without soy is modified by C-reactive protein status in
moderately hypercholesterolemic adults. J Nutr 135, 10751079.

Soy, homocysteine and inflammation


KA Greany et al

1425
Hwang SJ, Ballantyne CM, Sharret AR, Smith LC, Davis CE, Gotto AM
et al. (1997). Circulating adhesion molecules VCAM-1, ICAM-1,
and E-selectin in carotid atherosclerosis and incident coronary
heart disease cases. Circulation 96, 42194225.
Jenkins DJ, Kendall CW, Connelly PW, Jackson CJ, Parker T, Faulkner
D et al. (2002a). Effects of high- and low-isoflavone (phytoestrogen) soy foods on inflammatory biomarkers and proinflammatory
cytokines in middle-aged men and women. Metabolism 51, 919
924.
Jenkins DJ, Kendall CW, Jackson CJ, Connelly PW, Parker T, Faulkner
D et al. (2002b). 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 76, 365372.
Jenkins DJ, Kendall CW, Marchie A, Faulkner D, Wong J, de Souza R
et al. (2003). Effects of a dietary portfolio of cholesterol-lowering
foods vs lovastatin on serum lipids and C-reactive protein. JAMA
290, 502510.
Koh KK, Cardillo C, Bui MN, Hathaway L, Csako G, Waclawiw MA
et al. (1999). Vascular effects of estrogen and cholesterol-lowering
therapies in hypercholesterolemic postmenopausal women. Circulation 99, 354360.
Lampe JW, Karr SC, Hutchins AM, Slavin JL (1998). Urinary soy
excretion with a soy challenge: influence of habitual diet. Proc Soc
Exp Biol Med 217, 335339.
LHomme R, Brouwers E, Al-Maharik N, Lapcik O, Hampl R, Mikola H
et al. (2002). Time-resolved fluoroimmunoassay of plasma
and urine O-desmethylangolensin. J Steroid Biochem Mol Biol 81,
153157.
Lovati MR, Manzoni C, Gianazza E, Arnoldi A, Kurowska E, Carroll
KK et al. (2000). Soy protein peptides regulate cholesterol homeostasis in Hep G2 cells. J Nutr 130, 25432549.
Nettleton JA, Greany KA, Thomas W, Wangen KE, Adlercretuz H,
Kurzer MS (2004). Plasma phytoestrogens are not altered by
probiotic consumption in postmenopausal women with and
without a history of breast cancer. J Nutr 134, 19982003.
Nettleton JA, Greany KA, Thomas W, Wangen KE, Adlercretuz H,
Kurzer MS (2005). Short-term soy and probiotic supplementation
does not markedly affect concentrations of reproductive hormones in postmenopausal women with and without histories of
breast cancer. J Alt Comp Med 11, 10671074.
Nikander E, Metsa-Heikkila M, Tiitinen A, Ylikorkala O (2003).
Evidence of a lack of effect of a phytoestrogen regimen on the
levels of C-reactive protein, e-selectin, and nitrate in postmenopausal women. J Clin Endocrinol Metab 88, 51805185.
Olszewski AJ, McCully KS (1991). Homocysteine content of lipoproteins in hypercholesterolemia. Atherosclerosis 88, 6168.
Puska P, Korpelainen V, Hoie LH, Skovlund E, Lahti T, Smerud KT
(2002). Soy in hypercholesterolaemia: a double-blind, placebocontrolled trial. Eur J Clin Nutr 56, 352357.
Register T, Cann J, Kaplan J, Williams J, Adams M, Morgan T et al.
(2005). Effects of soy isoflavones and conjugated equine estrogens
on inflammatory markers in atherosclerotic, ovariectomized
monkeys. J Clin Endocrinol Metab 90, 17341740.
Reimann M, Dierkes J, Carlson A, Talbot D, Ferrari M, Hallund J et al.
(2006). Consumption of soy isoflavones does not affect
plasma total homocysteine or asymmetric dimethylarginine
concentrations in healthy postmenopausal women. J Nutr 136,
100105.
Ridker PM, Buring J, Shih JC, Matias M, Hennekens C (1998).
Prospective study of C-reactive protein and the risk of future
cardiovascular events among apparently healthy women. Circulation 98, 731733.
Ridker PM, Hennekens C, Buring J, Rifai N (2000). C-reactive protein
and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 342, 836843.

Ridker PM, Manson JE, Buring JE, Shih J, Matias M, Hennekens CH


(1999). Homocysteine and risk of cardiovascular disease among
postmenopausal women. JAMA 281, 18171821.
Roughead Z, Hunt J, Johnson J, Badger T, Lykken G (2005).
Controlled substitution of soy protein for meat protein: effects
on calcium retention, bone, and cardiovascular health indices in
postmenopausal women. J Clin Endocrinol Metab 90, 181189.
Ryan-Borchers TA, Park JS, Chew BP, McGuire MK, Fournier LR,
Beerman KA (2006). Soy isoflavones modulate immune function
in healthy postmenopausal women. Am J Clin Nutr 83, 11181125.
Setchell K, Brown N, Lydeking-Olsen E (2002). The clinical
importance of the metabolite equola clue to the effectiveness
of soy and its isoflavones. J Nutr 132, 35773584.
Skouby SO, Gram J, Andersen LF, Sidelmann J, Petersen KR, Jespersen
J (2002). Hormone replacement therapy: estrogen and progestin
effects on plasma C-reactive protein concentrations. Am J Obstet
Gynecol 186, 969977.
Steinberg F, Guthrie N, Villablanca AC, Kumar K, Murray M (2003).
Soy protein with isoflavones has favorable effects on endothelial
function that are independent of lipid and antioxidant effects in
healthy postmenopausal women. Am J Clin Nutr 78, 123130.
Teede HJ, Dalais FS, McGrath BP (2004). Dietary soy containing
phytoestrogens does not have detectable estrogenic effects on
hepatic protein synthesis in postmenopausal women. Am J Clin
Nutr 79, 396401.
Teede HJ, McGrath BP, Lakmini D, Cehun M, Fassoulakis A, Nestel P
(2003). Isoflavones reduce arterial stiffness: a placebo-controlled
study in men and postmenopausal women. Arterioscler Thromb
Vasc Biol 23, 10661071.
Tonstad S, Smerud K, Hoie L (2002). A comparison of the effects of 2
doses of soy protein or casein on serum lipids, serum lipoproteins,
and plasma total homocysteine in hypercholesterolemic subjects.
Am J Clin Nutr 76, 7884.
Walsh BW, Cox DA, Sashegyi A, Dean RA, Tracy RP, Anderson PW
(2001). Role of tumor necrosis factor-alpha and interleukin-6
in the effects of hormone replacement therapy and raloxifene on
C-reactive protein in postmenopausal women. Am J Cardiol 88,
825828.
Walsh BW, Paul S, Wild R, Dean RA, Tracy RP, Cox DA et al. (2000).
The effects of hormone replacement therapy and raloxifene on Creactive protein and homocysteine in healthy postmenopausal
women: a randomized controlled trial. J Clin Endocrinol Metab 85,
214218.
Wang G, Lapcik O, Hampl R, Uehara M, Al-Maharik N, Stumpf K et al.
(2000). Time-resolved fluoroimmunoassay of plasma daidzein and
genistein. Steroids 65, 339348.
West SG, Hilpert KF, Juturu V, Bordi PL, Lampe JW, Mousa SA et al.
(2005). Effects of including soy protein in a blood cholesterollowering diet on markers on cardiac risk in men and in
postmenopausal women with and without hormone replacement
therapy. J Womens Health 14, 253262.
Xu X, Duncan AM, Wangen KE, Kurzer MS (2000). Soy consumption
alters endogenous estrogen metabolism in postmenopausal
women. J Nutr 9, 781786.
Yildirir A, Aybar F, Tokgozoglu L, Yarali H, Kabakci G, Bukulmez O
et al. (2002). Effects of hormone replacement therapy on plasma
homocysteine and C-reactive protein levels. Gynecol Obstet Invest
53, 5458.
Yildiz MF, Kumru S, Godekmerdan A, Kutlu S (2005). Effects of
raloxifene, hormone therapy, and soy isoflavone on serum highsensitive C-reactive protein in postmenopausal women. Int J
Gynaecol Obstet 90, 128133.
Zhang SX, Shu X, Gao YT, Yang G, Li Q, Li H et al. (2003). Soy food
consumption is associated with lower risk of coronary heart
disease in Chinese women. J Nutr 133, 28742878.

European Journal of Clinical Nutrition