Isoflavone-Rich Soy Protein Isolate Suppresses Androgen Receptor Expression Without Altering Estrogen Receptor-[Beta] Expression or Serum Hormonal Profiles in Men at High Risk of Prostate Cancer1-3

By Hamilton-Reeves, Jill M Rebello, Salome A; Thomas, William; Slaton, Joel W; Kurzer, Mindy S

Abstract The purpose of this study was to determine the effects of soy protein isolate consumption on circulating hormone profiles and hormone receptor expression patterns in men at high risk for developing advanced prostate cancer. Fifty-eight men were randomly assigned to consume 1 of 3 protein isolates containing 40 g/d protein: 1) soy protein isolate (SPI+) (107 mg/d isoflavones); 2) alcohol-washed soy protein isolate (SPI-) (<6 mg/d isoflavones); or 3) milk protein isolate (0 mg/d isoflavones). For 6 mo, the men consumed the protein isolates in divided doses twice daily as a partial meal replacement. Serum samples collected at 0, 3, and 6 mo were analyzed for circulating estradiol, estrone, sex hormone- binding globulin, androstenedione, androstanediol glucuronide, dehydroepiandrosterone sulfate, dihydrotestosterone, testosterone, and free testosterone concentrations by RIA. Prostate biopsy samples obtained pre- and postintervention were analyzed for androgen receptor (AR) and estrogen receptor-beta expression by immunohistochemistry. At 6 mo, consumption of SPI+ significantly suppressed AR expression but did not alter estrogen receptor-beta expression or circulating hormones. Consumption of SPI- significantly increased estradiol and androstenedione concentrations, and tended to suppress AR expression (P = 0.09). Although the effects of SPI- consumption on estradiol and androstenedione are difficult to interpret and the clinical relevance is uncertain, these data show that AR expression in the prostate is suppressed by soy protein isolate consumption, which may be beneficial in preventing prostate cancer. J Nutr. 137: 1769- 1775, 2007.

Introduction

Steroid hormones modulate growth of the prostate gland, and elevated levels of androgens have been associated with prostate cancer risk (1,2). Consumption of soy foods is thought to contribute to prostate cancer prevention as a result of the hormonal properties of soy isotlavones, either through altered endogenous circulating hormones or hormone-receptor signaling. Cell culture studies have suggested that the isoflavonoids, genistein and equol, exert the most noteworthy hormonal effects. Genistein inhibits the activity of 5alpha-reductase and 17beta-hydroxysteroid dehydrogenase, enzymes required for androgen synthesis (3,4). The isoflavonoid equol, a bacterially derived metabolite of the isoflavone daidzein, sequesters dihydrotestosterone (DHT)8 from the androgen receptor (AR) in rat prostate tissue (5). Both isotlavonoids accumulate in the prostate gland (6-9) and may mimic or modulate endogenous hormones relevant to prostate carcinopcncsis.

Despite evidence from in vitro studies, huamn intervention studies report inconsistent effects of soy or isotlavone consumption on circulating hormone profiles in men. Although reports show statistically significant suppression of total testosterone (10,11), sex hormone hinding globulin (SHBG) (12), DHT (13), dehydroepi.uidrosterone (14), estrone (15), and free androgen index (13), and increased concentrations of SHBG (16) and DHT (17), the majority of the 22 intervention studies to date have not found significant changes in circulating sex steroid hormones (10-31). Generally, the studies that report significant changes were carried out in older men for a relatively long duration. None of the published studies reported equol-excretor status effects on circulating hormone response to soy isoflavone interventions in men.

Circulating hormone profiles may tail to accurately reflect prostate tissue exposure, and evaluating hormone receptor expression patterns in the prostate may provide additional evidence concerning the role of soy as a cancer preventive dietary agent. The AR mediates the action of androgens, and AR expression is a potential marker for prostate cancer prognosis (32). Dietary genistein has been shown to downregulate AR mRNA expression in rodents (33,34), and genistein has been shown to suppress AK activity through an estrogen receptor-beta (ERbeta)-dependent mechanism in LNCIaP cells (35). Despite these data, to our knowledge, there are no studies published to date that evaluate the effects of soy protein isolate consumption on AR and ERbeta expression in men, although one study reported that an isoflavone extract derived from red clover failed to alter AR expression compared with historically matched controls (26).

The objective of this project was to evaluate the effects of isoflavone-rich soy protein isolate consumption on circulating concentrations of reproductive hormones and prostate tissue markers of estrogen and androgen receptor expression in men at high risk of prostate cancer. The effects of an isoflavone-rich soy protein isolate were compared with those of an isoflavone-poor soy protein isolate to determine whether the isoflavones are the responsible bioactive constituents. The underlying hypothesis was that isoflavone-rich soy protein isolate consumption would reduce circulating hormones, down regulate AR expression, and upregulate ERbeta expression.

Material and Methods

Subjects. Fifty-eight men, aged 50-85 y, were recruited at the Minneapolis Veteran’s Administration Medical Center Urology Clink from a group of patients that had already undergone a transrectal ultrasound and hiopsy. Patients in this study were either at high risk for developing prostate cancer (n = 53), or had low-grade prostate cancer that was being followed by active surveillance (n = 5). Subjects were considered high risk if they had high-grade prostatic intraepithelial neoplasia (PIN) (n = 50) and/or atypical small acmar proliferation (ASAP) (n= 14). The subjects with prostate cancer had Cileason scores of <6 and were not receiving any other prostate cancer therapy. Subjects were recruited by urologic physicians, and the research nurse reviewed tinpatients' medical records to determine that eligibility criteria were met. Exclusionary criteria included BMI >40 kg/m^sup 2^, prostate cancer that required medical treatment, prostatitis, alcohol consumption >14 drinks/ wk, soy or milk allergy, regular antibiotic use, or renal insufficiency.

Fighty-seven subjects were screened tor the study; 21 chose not to participate after attending the orientation session, and 66 subjects began the study. Fight subjects withdrew from the study before their 3-mo appointment [disliked the study treatment powder (M = 3), inconvenienced by study demands (n = 2), gastrointestinal discomfort (n = 1), chose conventional prostate cancer treatment (n = 1 ), weight gain (n = D]. Three subjects completed 3 mo of the study with good compliance but chose not to finish due to inconvenience of the study demands, and 55 subjects completed the full fi-mo study.

Data from 58 subjects were included in the serum hormone analysis, and 42 subjects were included in the hormone reveptor expression analysis. Fewer participants were eligible for the hormone expression analysis because 3 subjects did not undergo the final prostate biopsy (liver cancer diagnosis (n = 1), heart condition (H = 1), not clinically indicated (n = 1)], and 13 subjects had insufficient hiopsy tissue at either baseline or postintervention for the analyses. All 58 subjects who completed the study were Caucasian.

Study design. The University of Minnesota Institutional Review Board Human Subjects Committee, the Minneapolis Veterans Affairs Institutional Review Boad, and the U.S. Army Medical Research and Materiel Command’s Human Subjects Research Review Board approved the study protocol, and all subjects provided informed consent, attended an orientation session, and were provided with a study handbook. During the study orientation, subjects were interviewed and prompted about incidental exposure to dietary isoflavones (e.g., snack bars, shakes, soy nuts, canned tuna, legumes, breads) to determine whether they were soy consumers. Only one participant reported regular soy consumption, but he did not consume soy-containing products tor 1 mo prior to beginning the study. The 6-mo intervention study used a randomized, single-blinded, placebo-controlled, parallel design. Free-living subjects supplemented their diets with 1 of 3 randomly assigned protein isolates: 1) soy protein isolate high in isoflavones (SPI+); 2) soy protein isolate that had nu ist of the isoflavones removed by alcohol extraction (SPI-); or 3) milk protein isolate (MPI) (The Solac Company). The protein isolates were consumed in divided doses twice daily and contributed 40 g/d protein and 200-400 kcal/d (1 kcal = 4.18 kJ). The isoflavone content of the protein isolates expressed as aglycone equivalents was 107 +- 5.0 m g/d tor the SPI+ ; <6 +- 0.7 mg/d for the SPI-; and 0 mg/d for the MPI (mean +- SD). The mean distribution of isoflavones was 53% genistein, .15% daidzein, and 11% glycitein in SPI+ , and 57% genistein, 20% daidzein, and 23% glycitein in SPI- as analyzed by Dr. Pat Murphy (Department of Hood Science and Human Nutrition, Iowa State University). The packets of protein isolate were numbered and patients were unaware of the treatment protein isolate they had been assigned until all subjects completed the intervention. Only the study coordinators who administered the protein isolates knew the group to which each participant belonged. Compliance was assessed by counting the number of times the patient consumed the protein isolate, as self-reported in recording calendars given to them, and mean compliance was 94%. Dietary and herbal supplements were allowed, and participants were asked to avoid changing dosages or adding new supplements to their regimen during the study. Subjects consumed their habitual diets, and received detailed instructions to exclude soy products to minimize isoflavone consumption from other sources. Serum collection and analysis. Fasting blood was collected in the morning at 0, 3, and 6 mo. Serum was separated and aliquots were frozen at -70[degrees]C until analysis. All serum samples were analyzed for testosterone, free testosterone, DHT, androstanediol glucuronide (3alpha-AG), androstenedione, dehydroepiandrosterone sulfate [DMFAS), SIIBCi, estradiol, and estrone. Steroid hormones were analyzed in duplicate by RIA, and SHBG was analyzed by immimoradiometric assay (Diagnostics Systems Laboratories). Hormone analyses were performed in 3 batches and all assays required ^sup 125^I-labeled analyte. Intraassay variabilities were 3.7% for testosterone, 4.4% for free testosterone, 6.1 % for DHT, 4.5% for 3alpha-AG, 4.4% for androstenedione, 2.3% for DHEAS, 4.4% for SHBG, 3.9% for estradiol, and 4.1% for estrone. An internal control was utilized to determine variability among batches, and interassay van abilities were between 9 and 10% for all analytes. All 3 serum samples for each participant were analyzed in the same batch.

Urine collection and analysis. To assess equol-producer status, 24-h urine was collected in plastic containers containing 1 g/L of ascorbic acid and separated into aliquots after the addition of sodium azide to a final concentration of 0.1%. Aliquots were frozen at -20[degrees]C until analysis. Equol was determined by HPLC and MS as previously described (16). The intraassay CV for equol was 8.2%, and the interassay CV was 12.5%. Subjects were classified as equol excrctors if 24-h urine equol levels exceeded 1000 nmol/d.

Dietary intake and analysis, hood records were completed for 1 d before each clinic visit. A registered dietitian taught study participants how to keep accurate food records. Patients were encouraged to use household scales and volumetric tools and to submit food labels from unusual foods. Study coordinators reviewed each food record for ci impleteness and clarified ambiguities with the participant at each clinic visit. Food records were analyzed with Nutritionist V, version 2.3 (17), and, for each 3-d food record, mean intakes of energy, macronutrieiits, saturated fat, cholesterol, fiber, vitamin D, vitamin E, calcium, selenium, and zinc were calculated.

Tissue collection and analysis. Biopsies were performed before the initial screening and at the 6-mo clinical visit. Biopsy cores were formalin-preserved tor 24 h and paraffin embedded. The histological diagnoses were determined during a routine pathological evaluation. Imiminohistochemistry was performed to assess AR and ERbeta expression on primarily normal, hyperplastic, or preneoplastic glands collected from eligible study participants. Antigen retrieval was achieved by pressure cooking deparaffnized and rehydrated tissue sections at 103 kPa in citrate buffer. Sections were treated in quenching solution (3% H^sub 2^O^sub 2^ in 100% methanol), and then incubated with a protein-blocking solution ( 10% milk, 5%. serum, and 1 % ISSA). Samples were incubated overnight at 4[degrees]C with rabbit polyclonal anti-ERbeta antibody (ab3577; Abcam; 1:1000) for the FRbeta assay, or for 30 min at room temperature with the mouse monoclonal anti-AR antibody (AM256-2M; BioCienex; RTU) for the AR assays. Next, the avidin-biotin peroxidase method was carried out (Vectastain Elite ABC kit, Vector Laboratories). Color reaction was developed using diaminohenzidine as the chromagen. Appropriate positive and negative controls were included in all staining runs. Disrupted glands and glands on the edge ol tissue sections were excluded from analysis to avoid false positives. A technician without prior knowledge of histological grading scored both the intensity of immunostaining and the percentage of imnumopositive areas at 40 x magnification using the HSCORF. system as previously described (.1H). The range of the HSCORK is a minimum of 1 and a maximum of 4 (1 indicated absent staining; 4 indicated intense staining). A mean of 6 intact glands (range: 2-15) per slide for ERbeta and a mean of 8 intact glands (range: 3-19) per slide for AR were averaged to derive the HSCORF (Fig. 1).

Excluded from analysis. The following data were excluded from statistical analysis: 6-mo dietary intake from one participant reporting unusually low consumption (mean <500 kcal) (1 kcal = 4,184 kj) during the 3-d food diary as a result of illness; 3 mo DHEAS that was above normal range (16 [mu]mol/L) and inconsistent with the participant's baseline and 6-mo measurements; all DHEAS measurements from one participant with abnormally high 3-mo and 6-mo DHEAS concentrations (9 and 10 [mu]mol/L, respectively) compared with baseline; and all SHBG measurements from one subject with undetectable SHBG in the serum (<3 nmol/L). One subject did not consume the treatment powder for 3 d prior to his 6-mo appointment as a result of illness, so he was excluded from the 6-mo equol excretion analysis.

Statistical analysis. The data appeared normally distributed and had similar variance among groups. Demographic comparisons between groups were performed with 1-way ANOVA for continuous endpoints, and chi-square for categories of prostate cancer markers, ANCOVA was used to compare groups adjusted tor the baseline value of the final endpoint. For androstenedione, the model included a treatment by baseline interaction. Preplanned pairwise comparisons of all groups are reported for each endpoint as dictated by the study hypotheses: each group’s adjusted mean (least squares mean) was compared with the other 2 groups’ adjusted means, Paired t tests were used to test for significant within-group changes over time. In addition, these covariates were screened as adjusters: baseline body weight, equol excretor status, and energy and nutrient intake. P < 0.05 was considered significant. All analyses were performed using SAS, version 9.1 (39).

Results

Baseline. Baseline anthropometries, cancer status, and dietary intake did not differ among the groups (Table 1), except that the MPI group had a higher body weight and the SPI- group consumed significantly more protein, calcium, and zinc at baseline (Table 2). Baseline prostate steroid receptor expression patterns (Table 3) and serum hormone and SHBG concentrations (Table 4) did not differ among the groups.

Anthropometries and dietary intake. Body weight did not change from baseline to 3 or 6 mo m any group (Table 2), and the significant differences in body weight among the groups at baseline were maintained. Protein, calcium, and vitamin D intakes increased in all groups during the study as a result of their concentrations in the protein isolates, and the differences in protein, calcium, and zinc intake at baseline were not present at 3 and 6 mo. At 3 mo, total and saturated fat consumption were reduced in the SPI- group relative to kisfline. During the study, energy, carbohydrate, cholesterol, fiber, vitamin H, selenium, and zinc intakes did not change for any group. Dietary and herbal supplement usage did not differ among groups (data nor shown). Body weight and protein intake differences among groups were unrelated to altered hormone concentrations or steroid receptor expression patterns.

Steroid receptors. Baseline-adjusted AR expression was lower in prostate biopsies after 6 mo in the SPI+ group compared with the MPI group (P = 0.04) and tended to he lower in the SPI-group compared with the MPI group (P = 0.09; Table 3), AR expression significantly increased from baseline in the MPI group, hut not in the other 2 groups. There were no changes from baseline in ERbeta expression among the groups (Table 3).

Serum estrogens. The serum estradiol concentration was significantly increased in the SPI- group at 3 and 6 mo relative to baseline, and by 6 mo, baseline-adjusted estradiol concentrations were significantly higher in the SPI- group compared with the other 2 groups (Table 4). Serum estronc was also significantly increased in the SPI- group at 3 and 6 mo, and was significantly higher than in the MPI group at i mo but not at h mo.

Serum androgens and SHBG. The serum audrostenedione concentration was significantly higher in the SPI+ group than in the MPI group at 3 mo. At 6 mo it was significantly greater than at baseline in the SPI- group, resulting in a significantly higher concentration than in the SPI+ group (Table 4). At both 3 and 6 mo, serum DHEAS was higher in the SPI group than in the other 1 groups, and at 3 mo, 3a- AG was higher in the SPI- group than the other 2 groups. At 3 mo, the DHT concentration decreased from baseline in the SIM group. Serum SHBG concentrations were decreased significantly from baseline at 3 and 6 mo in all groups, with no difference among the groups.

Equol-excretor status and hormone profiles. Equol excretor status was assessed only in the SIP+ group, because only they consumed sufficient daidzein to excrete equol. At 3 mo, there were 4 excretors and 15 nonexcretors, but of this group, only 1 excretor remained at 6 mo [dropped out after 3 mo (n = 1), apparently changed status (n = 1), and excluded data (n = 1)]. Baseline characteristics (Supplemental Table 1) and serum hormone concentrations at 3 mo (Supplemental Table 2) did nut differ between excretors and nonexcretors.

TABLE 1 Baseline characteristics of subjects1 TABLE 2 Anthropometries and dietary intake of men at high risk of prostate cancer that consumed various protein isolates for 6 mo1

FIGURE 1 Representative immunohistochemical staining of AR in human prostate core biopsies for HSCORE. Arrow indicates stained acinar cell in MPI control group (enlarged view inset in lower right).

Discussion

The present study evaluated men at high risk of prostate cancer to determine the effects of soy protein consumption on serum hormones and prostate tissue steroid receptor expression levels. The major finding was lower AR expression levels and no differences in ERbeta expression or circulating hormones in men consuming SPI+ compared with those consuming MPI.

AR increased significantly from baseline in the MPI group, but did not change from baseline in the soy groups. Because AR expression is expected to increase in this population (40), we inter that SPI+ apparently prevented or suppressed a rise in AR expression. Lower tissue AR expression in the SPI+ group is consistent with research m which dietary phytoestrogens down- regulated AR mRNA expression in adult male rats (33,34,41). Our data differ, however, from those of Jarred et al. (26), who reported no differences in AR expression patterns between radical prostatecromy patients treated with isoftavones and historically matched controls. The inconsistent results between the 2 studies can be explained by several methodological differences. In the study by Jarred et al. (26), the subjects, who consumed 160 mg/d of isoflavones in extracts derived from red clover, were men with advanced prostaric neoplasms treated for short and varied time periods (7-54 d). The tissue sections studied from the radical prostatectomies taken from treated subjects represented cancerous glandular acinne and were compared with sections of cancers from historically matched controls. Our subjects consumed 107 mg/d of isoflavones in isoflavone-rich SPI, were earlier in the carcmogenesis continuum, were treated for 6 mo each, and all biological samples were evaluated within the same subject before and after the intervention. Furthermore, the gland acinac studied presented either benign, hyperplastic, or preneoplastic tissue.

TABLE 3 Steroid receptor expression of men at high risk of prostate cancer that consumed various protein isolates for 6 mo1

Consumption of SPI+ did not affect ERbeta expression or circulating hormones. The ERbeta expression results are inconsistent with studies in animals in which prolonged isoflavone exposure decreased ERbeta expression (33,42), and may he explained by the variability in commercially available ERbeta antibodies (43). Our hormone results, however, are consistent with most published reports trom the clinical setting. The testosterone results are consistent with numerous soy or isoflavone intervention studies in which no change in total testosterone was observed (12-31), hut differ from 2 studies of short duration ( 10,11 ). Our rinding of no effect on directly measured tree testosterone is similar to published soy or isoflavone intervention studies to date ( 11,14,15,20,22,24), and our rinding of no effect on circulating DHT is consistent with most reports (10,14,16,19-21,23,30), although it differs from results of 2 studies ( 13,17), one of which used red clover extract (17). The lack of effect on circulating cstradiol or cstrone is consistent with the literature (10,1 1,15,16,19,22,29,30), although there is one report of decreased estrone in men consuming soymuk for 8 wk (15).

Serum SHBG decreased significantly from baseline in all study groups. The finding that consumption of SPI+ decreased SHBG is similar to a report by Mackey et al. (12); however, they did not rind a significant decrease in SHBG with an isoflavonepoor protein isolate as we did. In contrast to our findings, Hahito et al. (16) reported increased SHBG in men consuming 35 g of tofu daily for 2 wk, and others have reported no significant changes of SHRG with isoflavone-rich foods or extracts (13,15,17,20-23,30). Decreased SHHG is a potentially harmful effect because SHBCi-bound hormones are less biologically available to stimulate hormone-sensitive cancers. Because high protein intake has been associated with decreased SHBG (44), it is likely that the decrease in SHRG from baseline in all groups in our study resulted from the subnets’ significantly increased protein intake during the study (45).

The hormonal effects in the SIP- group were unexpected. Although AR expression was not significantly lower in the SPI-group, AR expression appeared to be intermediate between that of SPI+ and MPI groups. In addition, serum estradiol was increased in the SPI- group. These results are similar to a study in young men by Dillingham ct al. (20) in which a low-isofbvone protein isolate containing <2 mg/d isoflavones significantly increased estradiol and estrone compared with a milk protein isolate after a 8-wk intervention. Our results differ, however, from a study in older men by Goldin et al. ( 19) in which a low-isoflavone soy protein isolate containing <2 mg/d isotlavones did not change estradiol or estrone concentrations after a 6-wk intervention. Interestingly, we found serum est radio! was significantly higher in the SPI- group than in the SPI+ group, whereas in Dilingham et al. (20) found that estradiol in the low-isotlavonegnmp did not differ from the high- isoflavone group (20).

TABLE 4 Serum hormones and SHBG m men at high risk of prostate cancer that consumed various protein isolates for 6 mo1

Serum androstenedione and DHEAS concentrations were increased in the SPI- group compared with both SPI+ and MPI groups. No other soy protein or isoflavone intervention study has reported a change in circulating androstenedione (12,17,19,20,30), hut till other studies to date have intervened for a shorter duration. Higher DHtAS is consistent with orht-r low-isoflavone soy protein isolate interventions ( 19,20). Although DHKAS and androstenedione can he converted by 17/3-hydroxysteroid dehydrogenase to testosterone, no significant changes were observed in circulating testosterone, free testosterone, or DHT. Instead, our study population had low, but normal, testosterone concentrations throughout the study. Although DHIiAS and androstenedione concentrations have been associated with aggressive prostate cancer (46), our Hndings of unchanged testosterone and .1 trend toward lower AR expression (/J – 0.09) suggest neutral effects of SPI- consumption. In fact, because DHIiAS and aiidrostenedione may be converted to estradiol and estrone in the prostate gland (47), the increase in DHKAS and androstenedioiie may have contributed to the observed increases in circulating estradiol and estrone. The hormonal effects of SIM- consumption are likely due to the effects of the alcohol extraction process on SIM constituents.

In conclusion, we found that consumption of isoflavone-rich soy protein for 6 mo lowered AR expression levels in the prostate, but did not change KR/i expression or circulating hormones in men .it high risk of prostate cancer. Although consumption of the alcohol- extracted soy protein did not significantly lower AR expression, its effect appeared to be intermediate to that of SPI + and MPI consumption, suggesting that the isotlavones alone may not be responsible for the AR expression decrease, or, alternatively, that the low level of isotlavoues in SPI- were sufficient to alter the AR. Unexpectedly, consumption of SPl-, but not SPI + , significantly increased estradiol and androstenedione concentrations. None of these results were influenced by equol excretion status. These data suggest that consumption of isotlavone-rich and isoftavone-poor soy protein isolate exert differing effects on endogenous hormones and receptor expression, which may mediate prostate cancer preventive effects.

Acknowledgments

ImimmolmtoJieiuistry was performed by Kenji Takamura. The authors thank Kayla Vettlingand Kllie Wiener for tissue scoring, and Lori Sorensen, Nicole Nelson, and Mary McMulleii foiassistance with clinic visits and data entry.

1 The Soy and Prostate Cancer Prevention (SoyCaP) trial was supported by grant DAMD 17-02-1-0101 (M.S.K.I and W81XWH-06-1-0075 (J.H.R) from the United Slates Army Department of Defense Prostate Cancer Research Program. The protein isolates were donated by The Solae Company, St. Louis, MO. Neither sponsor was involved in writing this report.

2 Author disclosures: J. M. Hamilton-Reeves, S. A, Rebello, W. Thomas. J. W Slaton, and M. S. Kurzer, no conflicts of interest.

3 Supplemental Tables 1 and 2 are available with the online posting of this paper at jn.nutrition.org.

8 Abbreviations used: 3alpha-AG, androslanediol glucuronide; AR, androgen receptor; DHEAS, dehydroepiandrosterone sulfate; DHT, dihydrotestosterone; ER0. estrogen receptor-beta, MPI, milk protein isolate; SHBG, sex hormone-binding globulin, SPI-, alcohol- extracted sov protein isolate; SPI+ , isotlavone rich soy protein isolate.

Literature Cited

1. Shaneyfelt T. U.K., Biihlcy G.Mantzoros C. Hormonal predictors of prostate i.incer: A meta-analysis. J t.’lin Oncol. 2000;18:847- 53.

2. Bosland M. The role ni steroid hormones in prostate carcinogenesis. J Nail Cancer lust Monngr. 2000;27:.19-66.

3. Hvans BA, Griffiths K, Morton MS. Inhibition of 5 alpha- rcductase in genital skin h’hrohlasts .ind prostate tissue by dietary lign.ms and isollavniu.ids. J Knd.K;riiiol. 1995;147:295- 302.

4. Maki-la S, Poutancn M, Kostian ML, Lehtimaki N, Strauss L, Santti R, Vihko R. lnliihitioii

5. Lund TD, Munson DJ, Haldy MK, Setchell KDR, I epluirt KD, Manda RJ. Kqtiol is a novel anti-androgen that inhibits prostate pritwrh and hiirinonL- ieedback. Biol Rt-prnd. 2004;70:1188-95.

6. Riiwland I, Fuinhii.in M, Hoey L, Wahal.i K, Williamson G, Cassidy A. Bkuvjilahility of phyto-oestrogens. Br J Nutr. 200;89: Suppl 1 :S45-.S8. 7. Morion MS, Chan PS, Chcng C:, Black lock N, Matos-Fcrreira A, Abranchfs-Monteiru L, C’orrcia R, Lloyd S, Griiriihs K. Lignans and isollavonoids in plasma and prostatic fluid in men; Samples from portugal, hong kong, and ihc united kingdom. Prostate. 1997;32:I22-S.

8. Hedlund TL, Maroni I’D, Ferucci PG,, Dayton K, Barnes S, Junes K, Moore R, Ogden LG. W.ili.ila K, et .il. Long-term dietary habits – tttect soy isollavonc metabolism and accumulation in prust.itit tlind in Caucasian men. J Nuir. 2005;135:1400-6.

9. Hong SJ, Kiin Sl, K won SM, Lee JR, Chung IiC. Comparative study of concent ration of isollavones and lignans in plasma and prostatic tissues of normal control and benign prostatic hyperpl.isia. Yonsei Mfd J. 2002;43:236-41.

10. Gardncr-Thorpe D, O’Hagen C, Young 1, Lewis SJ. Dietary supplements of soya Hour lower scrum testosterone concentrations and improve markers of oxidative stress in men. Kur j Clin Nutr. 2003;57:IOO-6.

11. Spentzus D, Mantzoros Cl, Regan MM, Mnrrissey MK, Duggan S, Klickner-Garvey S, McCormick H, IX-WoIl W, Balk S, Buhley GJ. Minimal effect of a low-fat/high soy diet for asymptomatic, horimmally naive prostate tauter patients. Um Cancer Res. 2003;9:3282-7.

12. M.itkey R, Kkangaki A, Kden JA. The effects of soy protein in women and men with elevated plasma lipids. Biotactors. 2000; 12:251- 7.

13. Kranse R, Dagnelie PC, van Kemenade MC, de Jong KH, Blom JH, Tijburg LB, Weststrate JA, Schroder KH. Dietary intervention in prostate cancer patients: PSA response in a randomized double-Wind placebo-controlled study, lnt J Cancer. 2005;113:835-40.

14. Kischer L, Mahoney C, Jeffcoat AR, Kinrh MA, Thomas BK, Valentine JL, Stinchcomhc T, Boan J, Crnwell JA, Zeisel Sl 1. Clinical characteristics and pharmatokinetics of purified soy isollavones: Multiple-dose administration to men with prostate neoplasia. Nutr Cancer. 2004;4X:160-70.

15. Nagata C, Takatsiika N, Shimizu H, Hayashi H, Akamatsu T, Murase K. Effect of soyrnilk consumption on serum estrogen and androgcn concentrations in Japanese men. Canter Kpidemiol Biomarkers Prev. 2001;10:179-84.

16. Habim RC, Momalto J, Leslie K, Halt MJ. Fffetts ol replatiiif; meat with soyahean in the diet on sex hormone concentrations in healthy adult males. Br J Nutr. 2000;84:557-63.

17. Lewis JG, Morris JC, Clark BM, Elder PA. The effect of isoflavonc extract ingestion, as trinovin, on plasma steroids in normal men. Steroids. 2002;h7:25-9.

18. Ornish D, Weidncr G, Kair WR, Marlin R, Pettengill EB, Raisin CJ, Dunn-Linke S, Crutchfield L, Jacobs KN, et al. Intensive lifestyle changes may affect the progression of prostate cancer. J Urol. 2005; 174:106 S,9; discussion 1069-70.

19. Goldin BR, Brauner K, Adlercreutz. H, Ausman LM, l.ichtenstein All. Hormonal response to diets high in soy or animal protein without and with isollavones in moderately hypertholesterolemic subjects. Nuir Cancer. 2005;5l:l-fr.

20. Dillim-ham BL, McVeigh Bl, lampe JW, Duncan AM. Soy protein isolates of varying isotlavone content exert miiinr ellects on serum reproductive hormones m healthy young men. J Nutr. 2005;135:584-91.

21. Schroder KH, Roohol MJ, Boevr KR, de Mutscrt R, Zuijdgcest- van Leeuwen SD, Kersten I, Wildliagen MK, van Helvoort A. Randomized, double-blind, placebo-controlled crossover study in men with prostate cancer and rising PSA: Klfettiveness of a dietary supplement. Eur Urol. 2005;48:922,30; discussion 930-1.

22. kumar NB, Cantor A, Alien K, Riccardi D, Besteriiian-Dahaii K, Seigne J, HeIaI M, Salup R, Pow-Sang J. The specific role of isoflavones in reducing prostate cancer risk. Prostate. 2004;59;14l- 7.

23. Dalais FS, Meliala A, Wattanapenpaiboon N, Krydenherg M, SHUT DA, Thomson WK, Wjhlqvist ML. Kftetts of a diet rith in plmoestrogens on prostate-specific antigen and sex hormones m men diagnosed with prostate tancer. Urology. 2004;64:5 10-5.

24. deVere White RW. Hackman RM, Soares SK., Beckett LA, Li Y, Sun B. Efflects of a genistein-rich extract on PSA levels in men with a history of prostate cancer. Urology. 2004;6.1:259-63.

25. Hussain M, Banerjce M, Sarkar KH, Djuric Z, Pollak MN, Docr^e D, Kontana J, Chinni S, Davis J, et al. Soy isotlavones in the treatment of prostate cancer. Nutr Cancer. 2003;47:111-7.

26. Jarred RA, Keikh.i M, Howling C, McPherson SJ, Clare AM, Husband AJ, Pederscn |S, Krydenberg M, Kisbndger (if*. Induction of apopmsis in li’w to moderate-grade human prostate carcinoma hy red cloverderived dietary isotlavones. Cancer Kpidemiol Biomarkers Prev. 2002; ll:l6K9-9n.

27. Teede HJ, Dalais KS, Kotsopouli.s I), Li.mg YL, Davis S, Mc(irath HP. Dietary soy has both beneficial and potentially adverse cardiovascular eltects: A placebo-controlled study in men and posrmeiiopausal women. J Clin Kndocrinol Metah. 2001;86:3053-60.

28. Higashi K, Ahata S, Iwamoto N, Ogura M, Yamashita T, Ishikawa O, Ohslzu H, Nakamura 11. Effects of soy protein on levels of remnant-like particles cholesterol and vitamin E in healthy men. | Nutr Sa Vit.iminnl (Tokyo). 2001;47:2K1-N.

29. Mitchell JH, Cawood E, Kinnihurgh D, I’rovan A, Collins AR, Irvine DS. F.itect of a phytoestrogcn fooil supplement on reproductive health in normal males. Clin Sci (Lond). 2001;100:613- 8.

30. Ranmkko A, I’etas A, R.iivio T, Janne OA, Ranmkko S, Adlercrent? H. The effects ut short-term oral phy(oestrogen supplementation on the hypothalamic-pituitary-testicular axis in prostate cancer patients. Prostate. 2006;66:1086-91.

31. MaskarinecCi, Morimoto Y, Hehshi S.Sharma S, Franke AA, Stanczyk FZ. Serum prostate-specific antigen hut not testosterone levels decrease in a randomized soy intervention among men. Eur J Clin Nutr. 2006;60:1423-9.

32. Quinn DI, Henshall SM, Sutherland RI.. Molecular markers of prostate cancer outcome. [review| [415 reis| liurj Cancer. 2005;41:858-87.

33. Fritz WA, Wang J, F.ltnum It, Lamartiniere C]A. Dietary genistein down-regulates androgeri and estrogen receptor expression in the rat prostate. MoI Cell Hndocrinnl. 2002; 186:89-99.

34. l-.imartiniere CA, Cotronco MS, Frit/ WA, Wang J, Mentor-M a reel R, F-lgaviih A. Genistein chemoprevention: timing and mechanisms of action in murine mammary and prostate. J Nutr. 2002;132:5S2S-8S.

35. Bektic J, Berger AP, Pfeil K, Dohlcr C,, Bartsch Ci, Klotkcr H. Androgen receptor regulation hy physiological concentrations of the isoflavonoid gcnistein in androgen-dependent LNCaP cells is mediated hy estrogen receptor, (beta] Eur Urol. 2004;45:245-51.

36. Franke AA, Custer LJ, Wilkens LR, Le Marchand LL, Nomura AM. Goodman MT, Kolonel LN. Liquid chromatographic-photodiode array mass spectrometric analysis of dietary phytoestrogens from human urine and NtxiJ. J Chromatogr B Analyt Technol Biomed Life Sci. 2002;777:45-S9.

37. Nutritionist V Version 2.3. First D.itaBand Division. San Bruno, CA: The Hcarst Corporation, 2000.

38. Budwit-Novotny DA, McCarty KS, Cox EB, Super JT, Mutch DC,, Creasman WT, Flowers JL, McCarty KS, Jr. lmmunohistochemical analyses of esirogcn receptor in endometrial adenocarcinoma using a monoclonal antibody. Cancer Res. l986;46;54iy-25.

39. Version SAS 9.1, Gary, NC: SAS Institute, Inc., 2004.

40. Burnstfin Ki.. Regulation ofandrogen receptor levels: Implications for prostate cancer progression and therapy. J CxII Biochern. 2005;95: 617-69.

41. l.und TD, Munson DJ, Adlercreutz H, Handa KJ, Lephart KD. Androgcn receptor expression in the rat prostate is down-regulated by dietary phytoestrogens. Reprod Biol Endocrinol. 2004;2:5.

42. DaIu A, Blaydes BS, Bryant CW, L.itendresse JR, Weis CC, Barry DeIcIi)S K. Estrogen receptor expression in the prostate ot rats treated with dietary genistcin. J Chromatogr B Analyt Technol Biomed Life Sci. 2002;777:249-60.

43. Skliris CP, Parkes AT1 Limer JL, Burdall SE, Carder I’J, Speirs V. Evaluation of seven oestrogen receptor beta antibodies for immunohistochemistry, western blotting, and flow cytometry in human breast tissue. J Pathol. 2I)02;197:1 5.S-M.

44. Anderson KE, Kappas A, Conney AH, Bradlow HL, Hshman J. The tnHuence of dietary protein and carbohydrate on the principal oxidative biotranslormations of estradiol in normal subjects. J Clin Endocrinol Metab. 1984;59:103-7.

45. Adlercreutz H. Diet and sex hormone metabolism. In: Rowland I, editor. Nutrition, toxicity, and cancer. Boca Raton, FL: CRC Press; 1991. p. 137-194.

46. Severi G, Morris HA, Maclnms RJ, Lnghsh DR, Tilley W, Hopper Jl , Boyle P, Giles GG. Circulating steroid hormones and the risk of prostate cancer. Cancer Epidemiol Biomarkers Prev. 2006;15:86-91.

47. lakase Y, Levesque MH, Luu-The V, EI-Alfy M, Labrie F, Pelletier G. Expression of enzymes involved in estrogen metabolism in human prostate. J Histochem Cytochcm. 2006;54:911-21.

Jill M. Hamilton-Reeves,4 Salome A. Rebello,4 William Thomas,5 Joel W. Slaton,6,7 and Mindy S. Kurzer4*

4 Department of Food Science and Nutrition; Division of Biostatistics in the School of Public Health; and 6 Department of Urologic

Surgery, University of Minnesota, Minneapolis, MN 55455 and 7 Department of Urology, Veterans Administration Medical Center,

Minneapolis, MN 55417

* To whom correspondence should be addressed. E-mail: mkurzer@umn.edu.

Copyright American Institute of Nutrition Jul 2007

(c) 2007 Journal of Nutrition, The. Provided by ProQuest Information and Learning. All rights Reserved.