This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Anthony, M. S. Articles by Wagner, J. D. Search for Related Content PubMed PubMed Citation Articles by Anthony, M. S. Articles by Wagner, J. D.

(Arteriosclerosis, Thrombosis, and Vascular Biology. 1997;17:2524-2531.)
© 1997 American Heart Association, Inc.

Articles

Soy Protein Versus Soy Phytoestrogens in the Prevention of Diet-Induced Coronary Artery Atherosclerosis of Male Cynomolgus Monkeys

Mary S. Anthony; Thomas B. Clarkson; Bill C. Bullock; ; Janice D. Wagner

From the Comparative Medicine Clinical Research Center, Bowman Gray School of Medicine, Wake Forest University, Winston-Salem, NC.

Correspondence to Mary S. Anthony, MS, Department of Comparative Medicine, Bowman Gray School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1040.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Abstract Soy protein, long recognized as having cardiovascular benefits, is a rich source of phytoestrogens (isoflavones). To distinguish the relative contributions of the protein moiety versus the alcohol-extractable phytoestrogens for cardiovascular protection, we studied young male cynomolgus macaques fed a moderately atherogenic diet and randomly assigned to three groups. The groups differed only in the source of dietary protein, which was either casein/lactalbumin (casein, n=27), soy protein with the phytoestrogens intact (soy⁺, n=27), or soy protein with the phytoestrogens mostly extracted (soy^-, n=28). The diets were fed for 14 months. Animals fed soy⁺ had significantly lower total and LDL plus VLDL cholesterol concentrations compared with the other two groups. The soy⁺ animals had the highest HDL cholesterol concentrations, the casein group had the lowest, and the soy^- group was intermediate. A subset was necropsied for atherosclerosis evaluations (n=11 per group). Morphometric and angiochemical measures were done to quantify atherosclerosis. Coronary artery atherosclerotic lesions were smallest in the soy⁺ group (90% less coronary atherosclerosis than the casein group and 50% less than the soy^- group), largest in the casein group, and intermediate in the soy^- group. The effects of the diets on lesion size and arterial lipid measures of the peripheral arteries were similar to those in the coronary arteries, with greatest prevention of atherogenesis with soy⁺ and intermediate benefit with soy^- relative to casein. We could not determine whether the beneficial effects seen in the soy^- group relate to the protein itself or to the remaining traces of phytoestrogens. The beneficial effects of soy protein on atherosclerosis appear to be mediated primarily by the phytoestrogen component. Testicular weights were unaffected by the phytoestrogens.

Key Words: cynomolgus monkeys • isoflavones • males • phytoestrogens • soy protein

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Recently, several compounds that are estrogen agonists for bone, liver, and the cardiovascular system that appear to have no estrogen agonist effect on the female reproductive system have been studied.^1–5 Target tissue-specific estrogens are currently the focus of extensive research because although there is a well-known association between estrogen treatment and reduced coronary heart disease,⁶ there are also adverse effects on the reproductive system of females^7,8 and males.^9–12 Because we have seen no estrogen agonist effects of soybean phytoestrogens on the reproductive system in male or female macaques,³ it seemed reasonable to explore the therapeutic potential of these naturally occurring plant estrogens in the prevention of diet-induced atherosclerosis in males.

We reported previously that the phytoestrogens (isoflavones) of soybeans both favorably affect the plasma lipoproteins (significantly lower total and LDL plus VLDL cholesterol) of male rhesus monkeys and have no adverse effects on their reproductive systems.³ That observation prompted us to extend those studies to further ascertain whether the soy phytoestrogens could diminish or prevent diet-induced CAA in young male cynomolgus monkeys. In the present study, the monkeys were fed moderately atherogenic diets that differed only with respect to the protein component (casein/lactalbumin, alcohol-extracted soy protein isolate, or unextracted soy protein isolate). This was done to evaluate the relative contribution of the protein moiety versus the alcohol-extractable components of soy on atherosclerosis and its risk factors.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Animals
The subjects were 82 colony-born male cynomolgus monkeys (Macaca fascicularis) that ranged in age from 0.7 to 4.3 years at the start of the study. The animals were fed a moderately atherogenic diet (described below) throughout the baseline period and until the start of the experimental period. Animals were weaned at about 6 months of age and housed in single-gender peer social groups (5 to 10 animals per group) until the start of the experimental period. During the experimental period, the animals were housed in larger single-gender social groups (24 to 28 animals per group) in indoor/outdoor housing facilities containing 52 m² of indoor space and 125 m² of outdoor space.

During the study, three monkeys died of causes that appeared to be unrelated to the study, two from bacterial gastrointestinal disease and one from bronchopneumonia. All three deaths were in the casein group. All procedures involving animals were conducted in compliance with state and federal laws, standards of the US Department of Health and Human Services, and guidelines established by our institution's Animal Care and Use Committee.

Study Design
The study was a randomized, three-arm design with the treatment period lasting for 14 months after a 3-month baseline period. A stratified randomization, based on age and pretreatment TPC/HDL-C ratio, was used to assign monkeys to the following three treatment groups: (1) casein, a group fed a diet with casein and lactalbumin as the source of protein (n=27); (2) soy^-, a group fed a diet with soy protein isolate, from which the phytoestrogens had been extracted, as the protein source (0.17 mg phytoestrogens/g isolate) (n=27); and (3) soy⁺, a group fed a diet with soy protein isolate, with the phytoestrogens intact, as the protein source (1.5 mg phytoestrogens/g isolate) (n=28). Plasma lipid and lipoprotein concentrations (TPC, HDL-C, and triglycerides) and body weight were measured both at baseline and periodically during the treatment periods. After 14 months, the youngest 11 monkeys in each treatment group (ages 0.7 to 1.8 years at the start of treatment) were necropsied, and atherosclerosis evaluations and testicular weights were analyzed.

Diet Composition and Feeding
Throughout the baseline period and until the start of the experimental period, all monkeys were fed a diet with casein and lactalbumin as the protein source; the diet contained 0.2 mg cholesterol/kcal and 37% of calories from fat from the time of weaning. Although this initial diet was somewhat lower in fat and cholesterol than the experimental diets, it was simply a challenge diet. Because all the animals were fed the same diet, all baseline data were collected under the same dietary conditions. Three moderately atherogenic experimental diets were used, and their compositions are shown in Table 1.

View this table: [in this window] [in a new window]   Table 1. Experimental Diet Compositions

The isolated soy proteins used for this study were provided by Protein Technologies International. The soy protein with the phytoestrogens intact (Supro 670-HG^®) contained about 1.10 mg genistein and 0.37 mg daidzein (the two principal soy phytoestrogens) per gram of soy protein isolate. The alcohol-extracted soy protein (Supro 670-IF^®) contained only 0.12 mg genistein and 0.05 mg daidzein per gram of isolate. The alcohol-extracted soy protein isolate contained 90.7% protein and 0.7% fat compared with the unextracted protein, which was 87.0% protein and 4.0% fat. (Soy protein isolate analyses were done by Ralston Analytical Laboratories) Using high-performance liquid chromatography, we analyzed the soy protein isolates for ß-sitosterol as an indicator of other sterols that may have been extracted and found no detectable amount of ß-sitosterol in either isolate. The isolated soy protein is carefully processed before ethanol extraction to reduce lactins, phytase, and trypsin inhibitors, so the extracted and unextracted proteins were similar for components other than the phytoestrogens and lipid. The casein and lactalbumin were purchased from Harlan Teklad. Analyses of these proteins were performed by Covance Laboratories, Inc. Casein contains about 87.5% protein and 1.1% fat, and lactalbumin contains about 89.3% protein and 2.8% fat.

Diets were adjusted for the differences in the amounts of protein and fat to be isocaloric for macronutrients. The amounts of calcium and phosphorus were also adjusted to be approximately equal in all diets. Five grams per kilogram of D,L-methionine was added to the soy protein diets to ensure that the requirements of this essential amino acid were met and to approximately equilibrate the amounts of sulfur-containing amino acids in the three diets. The monkeys were fed about 75 kcal/kg body weight twice a day. Animals were given water ad libitum.

All diets were prepared in our diet laboratory in 10-kg batches and kept frozen until needed. One day's worth of diet was thawed overnight at 4°C before feeding.

Plasma Lipid and Lipoprotein Measurements
Blood for lipid and lipoprotein analyses was collected from the animals into evacuated tubes containing EDTA (final concentration, 1 g/L) after food was withheld for 18 hours. These analyses were done at baseline and at 2, 4, 6, 9, and 12 months of treatment. TPC was measured by enzymatic techniques based on the methods of Allain et al.¹³ Plasma triglycerides were determined by the methods of Fossati and Principe.¹⁴ HDL-C concentrations were measured using the heparin-manganese precipitation procedure described in the Manual of Laboratory Operations: Lipid Research Clinics Program.¹⁵ LDL-C plus VLDL-C was calculated as the difference between TPC and HDL-C. All analyses were done on a COBAS FARA II autoanalyzer. The laboratory subscribes to the Centers for Disease Control and Prevention Lipid Standardization Program. Body weight was monitored when the animals were sedated for collection of other samples.

Postmortem Evaluations
At the end of the study the monkeys were euthanized with sodium pentobarbital (100 mg/kg IV), a method consistent with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association. The cardiovascular system was flushed with lactated Ringer's solution. The right common iliac and right common carotid arteries were removed and frozen for chemical analyses. The left common iliac and left common carotid arteries, left carotid bifurcation, and abdominal aorta were removed, opened longitudinally, and immersion-fixed in 10% neutral buffered formalin for morphometric analyses. The hearts were removed, and the coronary arteries were fixed by perfusion for 1 hour with 10% neutral buffered formalin at 100 mm Hg pressure. Testicles were removed immediately after perfusion fixation, and weights were subsequently measured. Average testicular weights were calculated for each animal and used for subsequent analyses. Histological sections of the testicles were made and examined with light microscopy by one of the authors (B.C.B.).

Coronary artery¹⁶ and peripheral artery^17,18 morphometric evaluations were done using methodology established previously and described briefly below. To quantify the extent of CAA, we took five consecutive blocks from the LCX, five from the LAD, and five from the RCA (each 3 mm in length) beginning at the proximal portion of each artery. In animals of this age, these five serial blocks constitute nearly all of the coronary artery. Three blocks of carotid artery were taken, one each from the proximal, middle, and distal portions, using a crow's-foot template to standardize the sites relative to the total artery length. Three consecutive blocks of iliac artery were taken, beginning just distal to the abdominal aorta, which constituted the entire artery. Three blocks of abdominal aorta were studied, one near its origin (just distal to the celiac artery), one from the middle portion (just distal to the renal arteries), and one just proximal to the iliac artery bifurcation. One block of the carotid bifurcation was cut for analysis. For all arterial blocks, two 5-µm sections were cut from each block and stained with either hematoxylin and eosin or Verhoeff-van Gieson's stain.

Morphometric Evaluations
For morphometric measurements, sections of arteries stained with Verhoeff-van Gieson's stain were projected, using a projection microscope, onto a digitizer plate. Using a hand-held stylus and a computer-assisted digitizer, the component parts of the artery were traced. The measurements included intimal area (plaque size), area within the internal elastic lamina (artery size), lumen area, intimal thickness at the site of maximum lesion thickness in each block, and medial thickness at the site where intimal thickness was measured. Average intimal areas were calculated from the measures of the three blocks for each peripheral artery (abdominal aorta, left common carotid, and left common iliac) for each animal and used to generate the mean for each group. The average intimal areas and lumen areas of the five blocks from each of the three coronary arteries (LAD, LCX, and RCA) were calculated, the numbers were averaged to derive a coronary artery mean for each animal, then group means were calculated.

To measure the percentage of surface area with fatty streaks or plaque, the fixed left common iliac, left common carotid, and abdominal aorta sections were stained with Sudan IV in a 70% isopropanol solution, and the areas of staining were digitized.

Monkeys were considered to have CAA, as opposed to fatty streaks, when the intimal thickness was, on average, equal to or greater than half the thickness of the media. The prevalence was then calculated as the percentage of individuals in each group with atherosclerotic plaques.

Angiochemistry
Angiochemical analyses were done on lipid extracts of samples of the right common carotid and right common iliac arteries, which were prepared using a chloroform/methanol solution (2:1, vol/vol), following the method of Folch et al.¹⁹ Total and free cholesterol concentrations were determined enzymatically.²⁰ Cholesteryl ester concentrations were calculated as the difference between the measured total and free cholesterol.

Statistical Analyses
All analyses were done using BMDP Statistical Software, version 7.0. All variables measured at multiple time points during the treatment phase were analyzed using repeated-measures ANOVA or repeated-measures ANCOVA to assess changes over time and to determine whether there were important groupxtime interactions. If there were no significant groupxtime interactions (P>.05), the means for the treatment period for each group were compared by ANCOVA using the baseline measurement of that variable as the covariate. Body weight data were analyzed by ANCOVA using age and body weight at baseline as covariates. Testicular weights were analyzed by ANCOVA using age and body weight at necropsy as covariates.

Atherosclerosis and angiochemical measures were analyzed by ANCOVA using baseline lipid and lipoprotein concentrations and age as covariates. Comparisons were made using log-transformed intimal area and angiochemical data and square-root transformations of percent surface area data. These transformations were done to meet the statistical criteria of equivalence of variances between groups and to improve normality of distribution of the data. Because the group sizes for atherosclerosis measurements are relatively small (n=11 per group), we report all ANCOVA P values rather than only P values .05. Between-group comparisons were done using t tests. Group means that are significantly different (P<.05) are indicated by different letter superscripts in the tables.

Multivariate regression analysis, using a stepwise technique, was used to assess the association between age and lipid and lipoprotein variables measured during the treatment period and CAA. The log-transformed coronary artery intimal area variable was the dependent variable, and age and plasma lipid and lipoprotein concentrations were the independent variables. Variables were entered into the model until there was no further improvement in the explanatory capability of the model as determined by the change in the adjusted r².

	Results

Top Abstract Introduction Methods Results Discussion References

 
The plasma lipid and lipoprotein data for both the entire study population and the subset that was necropsied for atherosclerosis evaluations are shown in Table 2. Results are similar for the entire group and the necropsy subset. Because of the lower power with the smaller sample size in the necropsy subset, the differences between the soy^- and casein groups for HDL-C and the TPC:HDL-C ratio are not statistically significant (unlike the entire study population). TPC and LDL-C+VLDL-C concentrations of animals fed the soy^- diet were not different from those in the casein group. The animals fed the soy⁺ diet had significantly lower total plasma and LDL-C+VLDL-C concentrations compared with both the Casein group and the soy^- group. There was a stepwise increase in HDL-C, with the casein group having the lowest concentrations, the soy^- group having intermediate concentrations, and the soy⁺ group having the highest concentrations. For the entire study population, all between-group comparisons were significantly different, whereas for the necropsy subset, the casein and soy^- groups were not significantly different. There also were significant improvements in the TPC:HDL-C ratio, with both the soy^- and soy⁺ groups significantly lower than the casein group for the entire study population but not the necropsy subset and the soy⁺ group significantly lower than the soy^- group for both the larger population and subgroup.

View this table: [in this window] [in a new window]   Table 2. Plasma Lipids and Lipoproteins in Male Cynomolgus Monkeys in Response to Experimental Diets

Body weight and testicular weight data are presented in Table 3. Body weights were not different between diet groups. Likewise, testicular weights were not different between groups. Histological sections of the testicles were examined, and no differences could be detected in any of the microscopic features.

View this table: [in this window] [in a new window]   Table 3. Body Weight and Testicular Weight in Male Cynomolgus Monkeys in Response to Experimental Diets

Table 4 contains data on intimal area (lesion size) for the coronary arteries, abdominal aorta, left carotid bifurcation, and left common carotid and left common iliac arteries and percent of surface area that stained sudanophilic (lipid-containing area) for the abdominal aorta and left common carotid and left common iliac arteries. In general, average intimal area was always smallest in the soy⁺ group, largest in the casein group, and intermediate in the soy^- group.

View this table: [in this window] [in a new window]   Table 4. Atherosclerosis Measurements in Male Cynomolgus Monkeys in Response to Experimental Diets

Fig 1, a, shows the percentage of each group with atherosclerotic plaques, defined as mean intimal thickness equal to or greater than half the medial thickness (as opposed to no lesion or fatty streaks only). The intimal area (plaque size) for those with atherosclerotic plaques is shown in Fig 1, b. Thus, even considering only those with atherosclerotic plaques, the average lesion size was largest in the casein group (n=8), smallest in the soy⁺ group (n=5), and intermediate in the soy^- group (n=7).

View larger version (15K): [in this window] [in a new window]   Figure 1. a, Proportion of each group with CAA plaques, defined as intimal thickness greater than half the medial thickness (reflecting those with intimal lesions thicker than fatty streaks). b, Average lesion size for those monkeys with atherosclerotic plaques (mean intimal area±SEM, adjusted for baseline TPC).

We measured artery size and lumen size because compensation or remodeling of coronary arteries as a part of atherogenesis is important to clinical outcome.²¹ The lumens of the coronary arteries of the soy⁺ group were slightly larger than those of the soy^- or casein groups, although the differences were not statistically significant (soy⁺=0.50±0.03 mm², soy^-=0.42±0.03 mm², and casein=0.42±0.03 mm²; P=.13).

The results of the angiochemical measurements (total cholesterol, free cholesterol, cholesteryl ester, and triglycerides) are shown in Table 5. The angiochemical data support well the morphometric data. In both the common carotid and common iliac arteries, cholesterol concentrations are lowest in the soy⁺ group, intermediate in the soy^- group, and highest in the casein group.

View this table: [in this window] [in a new window]   Table 5. Angiochemical Measurements in Male Cynomolgus Monkeys in Response to Experimental Diets

Multivariate regression analysis was performed to determine to what extent age and plasma lipid and lipoprotein concentrations explained the variability in extent of CAA, which was measured as intimal area. The model that best fit the data included LDL-C+VLDL-C and HDL-C concentrations measured during the treatment period. In this model, LDL-C+VLDL-C and HDL-C explained about 50% of the variability in CAA (r²=0.509, P<.001). In this model the partial correlation coefficient was 0.397 for LDL-C+VLDL-C and -0.349 for HDL-C. The correlation coefficient between CAA and LDL-C+VLDL-C alone was 0.664 (P=.00003) and between CAA and HDL-C was -0.646 (P=.00005); thus, the strength of the association between atherosclerosis and either variable is about equal.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The evidence in the literature supporting the role of soy in the reduction of TPC and LDL-C is extensive.²² There are also experimental studies demonstrating the protection against atherosclerosis development by soy protein relative to animal proteins such as casein.^23,24 Many studies have attempted to determine the components of soy protein responsible for its hypocholesterolemic effect, but at best only a partial answer has been found. The amino acid composition of soy has been thoroughly investigated for its effect on plasma lipids and its role in atherosclerosis prevention.^25–30 Generally, vegetable protein amino acid reconstitutions were not as effective in lowering plasma cholesterol as the intact proteins, but animal protein amino acid reconstitutions induced a similar degree of hypercholesterolemia as the intact proteins.^25–27 An outgrowth of the amino acid composition hypothesis was investigation of the lysine:arginine ratio. Both Kritchevsky's group^28,29 and Sugano and coworkers³⁰ found that when lysine was added to soy protein to match the lysine:arginine ratio of casein or arginine was added to casein to equal the lysine:arginine ratio of soy, the intrinsic effect of the proteins was not appreciably modified. Taken together, the results of these studies suggest that components in soy protein concentrates other than the amino acids explain much of its hypocholesterolemic effects. Other attributes of soy have been hypothesized as being responsible for the lipid lowering, including saponins,^31–35 protein digestibility,³⁶ and protein phosphorylation,^37,38 but with no conclusive evidence.³⁹ Setchell⁴⁰ was the first to speculate that compounds in soy with estrogenic activity, termed phytoestrogens, might be responsible for the plasma cholesterol-lowering effect of soy but presented no experimental evidence.

Several lines of evidence have suggested that the components of soy protein that result in lipid lowering are alcohol extractable. One group reported that addition of an alcohol extract of soy protein to a casein-based diet lowered LDL-C concentrations in rats.⁴¹ Another study showed that a methanol-extracted, undigested fraction of soy protein was not as effective as the untreated product in preventing cholesterol increases in rats.⁴² Finally, our own work with nonhuman primates has shown that when the alcohol-extractable phytoestrogens are removed from soy protein, it is less effective in improving atherogenic diet-induced dyslipoproteinemia.³

Other than phytoestrogens, the alcohol-extractable components of soy that might be hypocholesterolemic are saponins, phytosterols, and soy globulins. Using ß-sitosterol as an indicator of phytosterols present in soy, we found no detectable amounts of ß-sitosterol in either the unextracted or extracted soy protein isolates. This is likely a result of the extensive processing of the isolates before alcohol extraction. Although the saponin content of these protein isolates was not evaluated, published values for protein isolates average 0.81%.⁴³ The saponins in soy have been studied rather extensively in both humans and various animal models.^31–34 There now appears to be a consensus that saponins, in the presence of soy protein, do not affect plasma cholesterol concentrations. There is evidence to suggest that this is due to protein-saponin interactions forming insoluble complexes that are biologically inactive.³⁴ Malinow,³⁵ who has extensively studied the hypocholesterolemic actions of alfalfa saponins, has also suggested that the molecular structures of the soy saponins may account for their lack of effect.

Sirtori et al⁴⁴ suggest that soy globulins, in particular a 7S globulin, may be the hypocholesterolemic component of soy protein. It is possible that during the aqueous alcohol-extraction process of the one soy protein isolate we used (soy^-), the globulins were denatured or removed, thereby leading to its relative ineffectiveness in improving plasma lipid concentrations and inhibiting atherogenesis. Experimental evaluation of the possible role of the 7S globulin must await technical developments that will allow the selective addition of the purified 7S globulin or purified phytoestrogens to the alcohol-extracted soy isolate or to casein/lactalbumin-based diets. Somewhat relevant to this question is a study by Balmir et al,⁴¹ in which acetone was added to the alcohol extract of soy to remove saponins, sugars, and other impurities. The resulting extract was about 79% phytoestrogens and 21% other components; when added to a casein-based diet, it resulted in LDL-C lowering in rats. Although this study supports the phytoestrogens as the cardioprotective component of soy, there is not yet conclusive proof.

The effects of the soy⁺ treatment on plasma lipids and lipoproteins were distinctly beneficial compared with those in the casein group. The effects of the soy^- diet on plasma lipids and lipoproteins fell between the improved concentrations noted in the soy⁺ group and those in the casein group. It remains unclear whether this intermediate effect is due to the protein moiety of the soy isolate, to its remaining phytoestrogen concentration (0.17 mg/g isolate), or to a combination of the two.

Because of the clinical implications, we studied atherosclerosis in the coronary arteries in more detail than in the other arteries. For the entire group, regardless of whether they had atherosclerotic plaques (Table 4), average lesion size (intimal area) was largest in the casein group, smallest in the soy⁺ group, and intermediate in the soy^- group. The treatments also affected both prevalence and plaque size for those with atherosclerotic plaques, ie, lesions thicker than fatty streaks (Fig 1). Treatment with soy with its phytoestrogens resulted in both the lowest prevalence and the smallest plaque sizes among those affected. Considering plaque size, in the group with atherosclerotic plaques, the soy^- group was not significantly different from the casein group (P=.16), but the soy⁺ group had strikingly smaller plaques than those in the casein group (P=.003) and, indeed, smaller plaques than the soy^- group (P=.05). Again, it is not possible from these data to know whether the trend toward some improvement in CAA in the soy^- group relates to the protein moiety, to its trace amounts of phytoestrogens, or to a combination of the two.

Table 4 summarizes the intimal areas of the coronary arteries of all three groups. Interestingly, the effect of soy with its phytoestrogens on coronary artery intimal area is much larger than we noted previously with estradiol treatment.⁴⁵ Estradiol resulted in a 56% decrease in coronary artery intimal area relative to an untreated control group, whereas we noted a 90% lower extent of CAA, measured as intimal area, with soy in this study (soy⁺ versus casein groups). Although not quite reaching a conventional level of statistical significance (P=.17), the soy⁺ group had a 50% lower extent of CAA compared with the soy^- group, which we presume to be attributable to the phytoestrogens.

The reduction in abdominal aorta atherosclerosis by soy treatment was slightly less than the reduction in coronary arteries (85%). In the previous study of estradiol's effects, reductions in plaque size of 24% were seen in the abdominal aorta versus the 56% decrease noted in coronary arteries.⁴⁵ Additionally, Wagner et al⁴⁶ have reported that the difference in LDL accumulation, an early end point in atherogenesis, is greatest in the coronary arteries and least in the abdominal aorta when compared with estrogen-treated and control monkeys. In a large autopsy study of humans, there appeared to be no "female protection" in the abdominal aorta versus a clear beneficial effect in coronary arteries.⁴⁷

Carotid and iliac arteries are also somewhat less responsive to the antiatherosclerotic effects of estradiol. We found reductions resulting from estradiol treatment of cynomolgus monkeys to be 31% and 43% in the carotid and iliac arteries, respectively.⁴⁵ In the present study, the effect of treatment with soy containing its phytoestrogens was as pronounced in carotid arteries as in coronary arteries (90% reduction) and only slightly less in the iliac arteries (83% reduction).

The lack of a statistically significant effect of soy treatment on carotid bifurcation atherosclerosis was not surprising. Although pathogenesis of carotid bifurcation is affected by the usual risk factors for atherosclerosis, there can be lesion development at this site in the absence of risk factors such as atherogenic diet and elevated TPC:HDL-C ratio.¹⁸ Lesion progression at this site may be modulated to a greater extent by heart rate and pulse wave velocity, factors that may not be affected by these dietary interventions. Consistent with this finding of a weaker effect of soy on carotid bifurcation atherosclerosis, a previous study showed essentially no effect of estradiol at the carotid bifurcation in the cynomolgus model (8% difference in estradiol-treated versus control animals).⁴⁵ Although there did appear to be some effect of soy⁺ treatment on atherogenesis at this site in this study (intimal area in the soy⁺ group is about one-third the size of the lesions in the soy^- and casein groups), it appeared to be less robust than in the other arterial sites.

The angiochemical and percent surface data (Table 5) support the intimal area effects already discussed. Because of the relatively small group sizes for atherosclerosis data, we made multiple measures of atherosclerosis (ie, intimal area, percent sudanophilia, and angiochemistry). The different measures are generally consistent in magnitude of the effect of the different treatments.

The correlation between HDL-C and LDL-C+VLDL-C concentrations during treatment and coronary artery intimal area is quite strong (multiple correlation coefficient=0.714). However, together they still explain only about 50% of the variation in CAA. The greater magnitude of effect seen with phytoestrogen-containing soy relative to estradiol could be because of the more pronounced effect of soy phytoestrogen treatment on lipids and lipoproteins, because of intervention at a younger age, or because soy phytoestrogens may have multiple mechanisms of action (eg, estrogenic effects⁴⁸ or tyrosine kinase inhibition⁴⁹).

We confirmed our previous observation³ that soy phytoestrogens had no effect on testicular weight and also found no effect on testicular development. In the previous study, we also found no significant differences between soy^- and soy⁺ groups in serum testosterone concentrations, although group sizes were small. More studies are required to investigate further whether soy phytoestrogens affect sexual behavior.

Monkeys in this study received the human equivalent of 143 mg/d of soy phytoestrogens, assuming a human intake of 2000 calories per day. We have no indication from this study (or the literature) what might constitute an effective antiatherosclerotic dose of soy phytoestrogens. Conclusive proof of whether the phytoestrogens are the cardioprotective component in soy protein remains to be determined. Further research into whether there are protein-soy phytoestrogen interactions that affect the actions of the phytoestrogens, effective doses of the phytoestrogens, and the relative effects of genistein and daidzein (the principal phytoestrogens in soy protein) are required.

	Selected Abbreviations and Acronyms

 

ANCOVA = analysis of covariance ANOVA = analysis of variance CAA = coronary artery atherosclerosis HDL-C = HDL cholesterol LAD = left anterior descending coronary artery LCX = left circumflex coronary artery LDL-C = LDL cholesterol RCA = right coronary artery TPC = total plasma cholesterol VLDL-C = VLDL cholesterol

	Acknowledgments

 
Supported in part by grant P01 HL-45666, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. The authors thank Protein Technologies International for provision of soy protein isolates; Karen Potvin Klein for editorial assistance; and Vickie Hardy, Tim Vest, and Maryanne Post for technical assistance. Presented in part at the 69th Scientific Sessions of the American Heart Association, New Orleans, La, November 10–13, 1996; at the Second International Symposium on the Role of Soy in Preventing and Treating Chronic Disease, Brussels, Belgium, September 15–18, 1996; and in abstract form (Circulation. 1996;94(suppl I):I-265).

Received January 16, 1997; accepted April 23, 1997.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Washburn SA, Adams MR, Clarkson TB, Adelman SJ. A conjugated equine estrogen with differential effects on uterine weight and plasma cholesterol in the rat. Am J Obstet Gynecol. 1993;169:251–256.[Medline] [Order article via Infotrieve]
2. Washburn SA, Honoré EK, Cline JM, et al. Effects of 17-dihydroequilenin sulfate on atherosclerotic male and female rhesus monkeys. Am J Obstet Gynecol. 1996;175:341–351.[Medline] [Order article via Infotrieve]
3. Anthony MS, Clarkson TB, Hughes CL, Morgan TM, Burke GL. Soybean isoflavones improve cardiovascular risk factors without affecting the reproductive system of peripubertal rhesus monkeys. J Nutr. 1996;126:43–50.
4. Black LJ, Sato M, Rowley ER, et al. Raloxifene (LY139481 HCl) prevents bone loss and reduces serum cholesterol without causing uterine hypertrophy in ovariectomized rats. J Clin Invest. 1994;93:63–69.
5. Ke HZ, Simmons HA, Pirie CM, Crawford D, Thompson DD. Droloxifene, a new estrogen agonist/antagonist, prevents bone loss in ovariectomized rats. Endocrinology. 1995;136:2435–2441.[Abstract]
6. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med. 1991;20:47–63.[Medline] [Order article via Infotrieve]
7. Colditz GA, Egan KM, Stampfer MJ. Hormone replacement therapy and risk of breast cancer: results from epidemiologic studies. Am J Obstet Gynecol. 1993;168:1473–1480.[Medline] [Order article via Infotrieve]
8. Grady D, Gebretsadik T, Kerlikowske K, Ernster V, Pettiti D. Hormone replacement therapy and endometrial cancer risk: a meta-analysis. Obstet Gynecol. 1995;85:304–313.[Abstract]
9. Robinson RW, Higano N, Cohen WD. Long-term effects of high-dosage estrogen therapy in men with coronary heart disease. J Chron Dis. 1963;16:155–161.[Medline] [Order article via Infotrieve]
10. Oliver MF, Boyd GS. Influence of reduction of serum lipids on prognosis of coronary heart disease. Lancet. 1961;2:499–505.[Medline] [Order article via Infotrieve]
11. Marmorston J, Moore FJ, Hopkins CE, Kuzma OT, Weiner J. Clinical studies of long-term estrogen therapy in men with myocardial infarction. Proc Soc Exp Biol Med. 1962;110:400–408.
12. Stamler J, Pick R, Katz LN, et al. Effectiveness of estrogens for therapy of myocardial infarction in middle-age men. JAMA. 1963;183:632–638.
13. Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem. 1974;20:470–475.[Abstract]
14. Fossati P, Principe L. Serum triglycerides determined colorimetrically with an enzyme that produces hydrogen peroxide. Clin Chem. 1982;28:2077–2080.[Abstract/Free Full Text]
15. Manual of Laboratory Operations: Lipid Research Clinics Program. Vol 1. Lipid and Lipoprotein Analysis. Bethesda, Md: National Heart, Lung, and Blood Institute, National Institutes of Health; 1974. US Dept of Health, Education, and Welfare publication NIH 75–268.
16. Clarkson TB. Progression and regression of nonhuman primate coronary artery atherosclerosis: considerations of experimental design. In: Malinow MR, Blaton VH, eds. Regression of Atherosclerotic Lesions. New York, NY: Plenum Publishing; 1984:43–60.
17. Clarkson TB, Bond MG, Bullock BC, McLaughlin KJ, Sawyer JK. A study of atherosclerosis regression in Macaca mulatta. V. Changes in abdominal aorta and carotid and coronary arteries from animals with atherosclerosis induced for 38 months and then regressed for 24 or 48 months at plasma cholesterol concentrations of 300 or 200 mg/dL. Exp Mol Pathol. 1984;41:96–118.[Medline] [Order article via Infotrieve]
18. Kaplan JR, Clarkson TB, Manuck SB. Pathogenesis of carotid bifurcation atherosclerosis in cynomolgus monkeys. Stroke. 1984;15:994–1000.[Abstract/Free Full Text]
19. Folch J, Lees M, Sloane-Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem. 1957;226:497–509.[Free Full Text]
20. Carr TP, Andresen CJ, Rudel LL. Enzymatic determination of triglyceride, free cholesterol, and total cholesterol in tissue lipid extracts. Clin Biochem. 1993;26:39–42.[Medline] [Order article via Infotrieve]
21. Clarkson TB, Prichard RW, Morgan TM, Petrick GS, Klein KP. Remodeling of coronary arteries in human and nonhuman primates. JAMA. 1994;271:289–294.[Abstract]
22. Anderson JW, Johnstone BM, Cook-Newell ME. Meta-analysis of the effects of soy protein intake on serum lipids. N Engl J Med. 1995;333:276–282.[Abstract/Free Full Text]
23. Huff MW, Roberts DCK, Carroll KK. Long-term effects of semipurified diets containing casein or soy protein isolate on atherosclerosis and plasma lipoproteins in rabbits. Atherosclerosis. 1982;41:327–336.[Medline] [Order article via Infotrieve]
24. Meeker DR, Kesten HD. Effect of high protein diets on experimental atherosclerosis of rabbits. Arch Pathol. 1941;31:147–162.
25. Huff MW, Hamilton RMG, Carroll KK. Plasma cholesterol levels in rabbits fed low fat, cholesterol-free, semipurified diets: effects of dietary proteins, protein hydrolysates and amino acid mixtures. Atherosclerosis. 1977;28:187–195.[Medline] [Order article via Infotrieve]
26. Huff MW, Carroll KK. Effects of dietary proteins and amino acid mixtures of plasma cholesterol levels in rabbits. J Nutr. 1980;110:1676–1685.
27. Nagata Y, Ishiwaki N, Sugano M. Studies on the mechanism of antihypercholesterolemic action of soy protein and soy protein-type amino acid mixtures in relation to the casein counterparts in rats. J Nutr. 1982;112:1614–1625.
28. Kritchevsky D. Vegetable protein and atherosclerosis. J Am Oil Chem Soc. 1979;56:130–140.[Medline] [Order article via Infotrieve]
29. Kritchevsky D, Tepper SA, Czarnecki S, Klurfeld DM. Atherogenicity of animal and vegetable protein: influence of the lysine to arginine ratio. Atherosclerosis. 1982;47:429–431.
30. Sugano M, Ishiwaki N, Nagata Y, Imaizumi K. Effects of arginine and lysine addition of casein and soya-bean protein on serum lipids, apolipoproteins, insulin and glucagon in rats. Br J Nutr. 1982;48:211–221.[Medline] [Order article via Infotrieve]
31. Topping DL, Storer GB, Calvert GD, Illman RJ, Oakenfull DG, Weller RA. Effects of dietary saponins on fecal bile acids and neutral sterols, plasma lipids, and lipoprotein turnover in the pig. Am J Clin Nutr. 1980;33:783–786.[Abstract/Free Full Text]
32. Calvert GD, Blight L, Illman RJ, Topping DL, Potter JD. A trial of the effects of soya-bean flour and soy-bean saponins on plasma lipids, faecal bile acids and neutral sterols in hypercholesterolaemic men. Br J Nutr. 1981;45:277–281.[Medline] [Order article via Infotrieve]
33. Pathirana C, Gibney MJ, Taylor TG. The effect of dietary protein source and saponins on serum lipids and the excretion of bile acids and neutral sterols in rabbits. Br J Nutr. 1981;46:421–430.[Medline] [Order article via Infotrieve]
34. Potter SM, Jiminez-Flores R, Pollack J-A, Lone TA, Berber-Jiminez MD. Protein-saponin interaction and its influence on blood lipids. J Agric Food Chem. 1993;41:1287–1291.
35. Malinow MR. Saponins and cholesterol metabolism. Atherosclerosis. 1984;50:117–119. Letter.[Medline] [Order article via Infotrieve]
36. Woodward CJH, Carroll KK. Digestibilities of casein and soya-bean protein in relation to their effects on serum cholesterol in rabbits. Br J Nutr. 1985;54:355–366.[Medline] [Order article via Infotrieve]
37. Samman S, Roberts DCK. The phosphorylation state of casein and its hypercholesterolaemic effect: the role of divalent cations. Atherosclerosis. 1984;52:347–348.[Medline] [Order article via Infotrieve]
38. Van der Meer R, de Vries H, West CE, de Waard H. Casein-induced hypercholesterolaemia in rabbits is calcium-dependent. Atherosclerosis. 1985;56:139–147.[Medline] [Order article via Infotrieve]
39. Forsythe WA, Green MS, Anderson JJB. Dietary protein effects on cholesterol and lipoprotein concentrations: a review. J Am Coll Nutr. 1986;5:533–649.[Abstract]
40. Setchell KDR. Naturally occurring non-steroidal estrogens of dietary origin. In: McLachlan JA, ed. Estrogens in the Environment II: Influences on Development. New York, NY: Elsevier; 1985:69–85.
41. Balmir F, Staack R, Jeffery E, Jimenez MDB, Wang L, Potter SM. An extract of soy flour influences serum cholesterol and thyroid hormones in rats and hamsters. J Nutr. 1996;126:3046–3053.
42. Sugano M, Koba K. Dietary protein and lipid metabolism: a multifunctional effect. Ann N Y Acad Sci. 1993;676:215–222.[Medline] [Order article via Infotrieve]
43. Anderson RL, Wolf WJ. Compositional changes in trypsin inhibitors, phytic acid, saponins and isoflavones related to soybean processing. J Nutr. 1995;125:581S–588S.
44. Sirtori CR, Lovati MR, Manzoni C, Monetti M, Pazzuconi F, Gatti E. Soy and cholesterol reduction: clinical experience. J Nutr. 1995;125:598S–605S.
45. Adams MR, Kaplan JR, Manuck SB, et al. Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomized monkeys: lack of an effect of added progesterone. Arteriosclerosis. 1990;10:1051–1057.[Abstract/Free Full Text]
46. Wagner JD, St Clair RW, Schwenke DC, Shively CA, Adams MR, Clarkson TB. Regional differences in arterial low density lipoprotein metabolism in surgically postmenopausal cynomolgus monkeys: effects of estrogen and progesterone replacement therapy. Arterioscler Thromb. 1992;12:717–726.[Abstract/Free Full Text]
47. Tejada C, Strong JP, Montenegro MR, Restropo C, Solberg LA. Distribution of coronary and aortic atherosclerosis by geographic location, race, and sex. Lab Invest. 1968;18:509–526.[Medline] [Order article via Infotrieve]
48. Wang TTY, Sathyamoorthy N, Phang JM. Molecular effects of genistein on estrogen receptor mediated pathways. Carcinogenesis. 1996;17:271–275.[Abstract/Free Full Text]
49. Shimokado K, Umezawa K, Ogata J. Tyrosine kinase inhibitors inhibit multiple steps of the cell cycle of vascular smooth muscle cells. Exp Cell Res. 1995;220:266–273.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				E. Grossini, C. Molinari, D. A. S. G. Mary, F. Uberti, P. P. Caimmi, N. Surico, and G. Vacca Intracoronary Genistein Acutely Increases Coronary Blood Flow in Anesthetized Pigs through {beta}-Adrenergic Mediated Nitric Oxide Release and Estrogenic Receptors Endocrinology, May 1, 2008; 149(5): 2678 - 2687. [Abstract] [Full Text] [PDF]

				S. Nagarajan, R. L. Burris, B. W. Stewart, J. E. Wilkerson, and T. M. Badger Dietary Soy Protein Isolate Ameliorates Atherosclerotic Lesions in Apolipoprotein E-Deficient Mice Potentially by Inhibiting Monocyte Chemoattractant Protein-1 Expression J. Nutr., February 1, 2008; 138(2): 332 - 337. [Abstract] [Full Text] [PDF]

				H. Qiu, C. Depre, K. Ghosh, R. G. Resuello, F. F. Natividad, F. Rossi, A. Peppas, Y.-T. Shen, D. E. Vatner, and S. F. Vatner Mechanism of Gender-Specific Differences in Aortic Stiffness With Aging in Nonhuman Primates Circulation, August 7, 2007; 116(6): 669 - 676. [Abstract] [Full Text] [PDF]

				J. A Nettleton, L. M Steffen, M. B Schulze, N. S Jenny, R G. Barr, A. G Bertoni, and D. R Jacobs Jr Associations between markers of subclinical atherosclerosis and dietary patterns derived by principal components analysis and reduced rank regression in the Multi-Ethnic Study of Atherosclerosis (MESA) Am. J. Clinical Nutrition, June 1, 2007; 85(6): 1615 - 1625. [Abstract] [Full Text] [PDF]

				M. R. Peluso Flavonoids Attenuate Cardiovascular Disease, Inhibit Phosphodiesterase, and Modulate Lipid Homeostasis in Adipose Tissue and Liver Experimental Biology and Medicine, September 1, 2006; 231(8): 1287 - 1299. [Abstract] [Full Text] [PDF]

				S. Nagarajan, B. W. Stewart, and T. M. Badger Soy Isoflavones Attenuate Human Monocyte Adhesion to Endothelial Cell-Specific CD54 by Inhibiting Monocyte CD11a J. Nutr., September 1, 2006; 136(9): 2384 - 2390. [Abstract] [Full Text] [PDF]

				B. L McVeigh, B. L Dillingham, J. W Lampe, and A. M Duncan Effect of soy protein varying in isoflavone content on serum lipids in healthy young men Am. J. Clinical Nutrition, February 1, 2006; 83(2): 244 - 251. [Abstract] [Full Text] [PDF]

				M. Reimann, J. Dierkes, A. Carlsohn, D. Talbot, M. Ferrari, J. Hallund, W. L. Hall, K. Vafeiadou, U. Huebner, F. Branca, et al. Consumption of Soy Isoflavones Does Not Affect Plasma Total Homocysteine or Asymmetric Dimethylarginine Concentrations in Healthy Postmenopausal Women J. Nutr., January 1, 2006; 136(1): 100 - 105. [Abstract] [Full Text] [PDF]

				M. R. Adams, D. L. Golden, J. K. Williams, A. A. Franke, T. C. Register, and J. R. Kaplan Soy Protein Containing Isoflavones Reduces the Size of Atherosclerotic Plaques without Affecting Coronary Artery Reactivity in Adult Male Monkeys J. Nutr., December 1, 2005; 135(12): 2852 - 2856. [Abstract] [Full Text] [PDF]

				V. V. Lozovaya, A. V. Lygin, A. V. Ulanov, R. L. Nelson, J. Dayde, and J. M. Widholm Effect of Temperature and Soil Moisture Status during Seed Development on Soybean Seed Isoflavone Concentration and Composition Crop Sci., August 26, 2005; 45(5): 1934 - 1940. [Abstract] [Full Text] [PDF]

				H. J. Teede, F. S. Dalais, D. Kotsopoulos, B. P. McGrath, E. Malan, T. E. Gan, and R. E. Peverill Dietary Soy Containing Phytoestrogens Does Not Activate the Hemostatic System in Postmenopausal Women J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 1936 - 1941. [Abstract] [Full Text] [PDF]

				D. Liu, H. Jiang, and R. W. Grange Genistein Activates the 3',5'-Cyclic Adenosine Monophosphate Signaling Pathway in Vascular Endothelial Cells and Protects Endothelial Barrier Function Endocrinology, March 1, 2005; 146(3): 1312 - 1320. [Abstract] [Full Text] [PDF]

				D. Liu, L. L. Homan, and J. S. Dillon Genistein Acutely Stimulates Nitric Oxide Synthesis in Vascular Endothelial Cells by a Cyclic Adenosine 5'-Monophosphate-Dependent Mechanism Endocrinology, December 1, 2004; 145(12): 5532 - 5539. [Abstract] [Full Text] [PDF]

				K. A. Greany, J. A. Nettleton, K. E. Wangen, W. Thomas, and M. S. Kurzer Probiotic Consumption Does Not Enhance the Cholesterol-Lowering Effect of Soy in Postmenopausal Women J. Nutr., December 1, 2004; 134(12): 3277 - 3283. [Abstract] [Full Text] [PDF]

				C. E. Wood, T. C. Register, M. S. Anthony, N. D. Kock, and J. M. Cline Breast and Uterine Effects of Soy Isoflavones and Conjugated Equine Estrogens in Postmenopausal Female Monkeys J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3462 - 3468. [Abstract] [Full Text] [PDF]

				C. Atkinson, W. Oosthuizen, S. Scollen, A. Loktionov, N. E. Day, and S. A. Bingham Modest Protective Effects of Isoflavones from a Red Clover-Derived Dietary Supplement on Cardiovascular Disease Risk Factors in Perimenopausal Women, and Evidence of an Interaction with ApoE Genotype in 49-65 Year-Old Women J. Nutr., July 1, 2004; 134(7): 1759 - 1764. [Abstract] [Full Text]