The effects of soy protein in women and men with elevated plasma lipids 1

Transcription

1 BioFactors 12 (2000) IOS Press Original report The effects of soy protein in women and men with elevated plasma lipids 1 R. Mackey a, A. Ekangaki a and J.A. Eden b, a Sydney Menopause Centre, Royal Hospital for Women, Barker Street, Randwick, NSW, 2031, Australia b School of Obstetrics and Gynaecology, University of New South Wales, Australia Abstract. Fifty four postmenopausal women with elevated cholesterol were recruited for a randomised, double-blind controlled trial of soy protein containing isoflavones. (ISP+) or a soy protein with a low isoflavone content (ISP ), taken daily for 12 weeks. There was an overall reduction after 12 weeks in total cholesterol (TC), LDL cholesterol (LDL-C), sex hormone binding globulin (SHBG), and luteinizing hormone (LH). There were no significant differences between treatment groups. In a separate study 27 male subjects with a TC > 5.5 mmol/l were given ISP+ for 12 weeks. In this male study there was a significant increase in HDL cholesterol (HDL-C) and SHBG. Soy protein has a cholesterol lowering effect in both women and men. These studies suggest that this effect is independent of isoflavones. Soy protein also reduces SHBG levels in both sexes. 1. Introduction As women pass through the menopause their lipoprotein profile worsens [1] and these changes can be reversed with Hormone Replacement Therapy (HRT) [2]. Soy protein has been shown to have significant lipid lowering effects in both men and women [3] but there is considerable debate as to whether this effect is due to the isoflavone content or some other component. We performed a series of studies in men and women using soy protein with or without isoflavones to study the effect on the lipoprotein profile as well as other biochemical indices such as sex hormones, pituitary hormones, markers of bone turnover and glucose tolerance. There has been some concern that isoflavones may have a feminizing effect on men and so in our male study we also measured andogen levels. 2. Materials and methods Female subjects were aged between 45 and 65 years, and were postmenopausal. Male subjects were over 40 years of age. Both groups were required to have a fasting total cholesterol (TC) > 5.5 mmol/l. 1 Financial support provided by Sanitarium Health Foods Pty Ltd. Correspondence to: Associate Professor John Eden, School of Obstetrics and Gynaecology, Royal Hospital for Women, Barker St., Randwick, NSW, 2031, Australia. Tel.: ; Fax: ; j.eden@unsw.edu.au /00/$ IOS Press. All rights reserved

2 252 R. Mackey et al. / Soy protein and elevated lipids Subjects were excluded if there was a history of allergy to soy or if any of they taking cholesterol lowing agents. The study was approved by the South Eastern Sydney Area Health Service Ethics Committee. The female study was a prospective, double-blind, randomised controlled study. The male study was an open prospective observational pilot study. All subjects were given the dietary guidelines from the National Heart Foundation (NHF) and were instructed to follow these for a period of four weeks. Subjects were invited to join the study if their serum TC was greater than 5.5 mmol/l after four weeks on the NHF diet. Pre-study baseline examinations included: 1) Body weight and height measurement 2) extraction of 30 ml of fasting venous blood for measurement of the following: Females: Total cholesterol; Triglycerides; HDL-C; LDL-C; Follicle Stimulating Hormone (FSH); LH; Thyroid Stimulating Hormone (TSH); SHBG; Osteocalcin; Bone-specific Alkaline Phosphatase. Males: All of the above and also Dehydroepiandrosterone (DHEA); Testosterone; Androstenedione. 3) Spot urine collection for measurement of deoxypyridinoline (in females only) and 4) a 24 hour urine collection for measurement of isoflavone metabolites. Fifty four female subjects were randomised to receive 28 g of protein powder; either a) a soy protein with an isoflavone content of 65 mg isoflavones daily (ISP+) or b) a soy protein isolate with less than 4 mg isoflavones per daily (ISP ). The product was provided by Protein Technologies International (St Louis, MI). The treatment was continued for 12 weeks and then stopped and further assessments were made 6 weeks later. They continued on their NHF diet and were instructed to avoid isoflavone rich foods during the study. In the male study all subjects received 28 grams of ISP+ for 12 weeks and continued on the NHF diet. All assays were performed by the staff of the Clinical Biochemistry Department at St George Hospital, Kogarah using commercial kits. All blood samples were collected after a minimum of 12 hours fasting. The blood was centrifuged and the serum was frozen at 70 C. Measurements were carried out in batches to minimise error. All intra-assay and inter-assay co-efficients of variation were less than 10%. Total cholesterol, HDL cholesterol and triglycerides were measured using enzymatic colorimetric tests (Boehringer Mannheim, Indianapolis, USA). Bone-specific alkaline phosphatase was measured using radioimmunoassay (Metra Biosystems, Inc). LH and FSH, osteocalcin and deoxypyridinoline were determined by immunoradiometric assay (Chiron Diagnostics Corporation, East Walpole, MA, Nichols Institute Diagnostics, CA and Diagnostic Products Corporation, CA, respectively). Serum LH, FSH and TSH were determined by ACS : 180 automated chemiluminescence system (Chiron Diagnostics Corporation, East Walpole, MA). Sex Hormone Binding Globulin and DHEA were measured using Immulite R chemiluminescent immunometric assays (DPC R, Los Angeles CA). Androstenedione and free testosterone were measured using Coat-A-Count R direct radioimmunoassays (DPC R Los Angeles, CA). The data from both studies were analysed separately using SAS software (version 6.12). Two sample students t-test and the Wilcoxon rank sum test were used to evaluate differences beween treatment groups in the female study. For variables measured at weeks 0 and 12, the paired t-test was used to evaluate change while linear regression was used to examine the effect of treatment group, age and initial weight on the change scores for each dependant variable, respectively. To accommodate the presence of repeated measurements, due to measurements taken at weeks 0, 12 and 18, regression analysis using generalised estimating equations was used to evaluate changes in lipoprotein profiles in relation to age, time and group where relevant. A significance level of 0.05 was used to determine statistical significance.

3 R. Mackey et al. / Soy protein and elevated lipids 253 Table 1 Baseline characteristics of the females in both treatment groups and male subjects. No significant differences were detected between the female treatment groups at baseline Female Female Male ISP (n =24) ISP+ (n =25) ISP+ (n =27) means (SD) means (SD) means (SD) Age (years) (4.21) (4.92) (8.38) Weight (kg) (9.82) (11.78) (13.62) TC (mmol/l) 7.47 (1.04) 7.29 (0.90) 6.86 (0.86) TC = total cholesterol; ISP = Soy protein with low isoflavone content; ISP+ =Soy protein with high isoflavone content. Table 2 Effects of diets containing a large isoflavone content (ISP+) or a small isoflavone content (ISP ) on lipid concentrations in the female subjects Treatment 0 weeks 12 weeks 18 weeks group means (SD) means (SD) means (SD) ON TREATMENT OFF TREATMENT Total cholesterol ISP (0.90) 6.94 (0.84) 7.03 (0.99) mmol/l ISP 7.47 (1.04) 7.15 (1.02) 7.30 (1.05) Total 7.38 (0.96) 7.04 (0.93) 7.16 (1.02) Triglycerides ISP (0.82) 1.54 (0.74) 1.61(0.56) mmol/l ISP 1.54 (0.72) 1.46 (0.82) 1.43 (0.66) Total 1.53 (0.77) 1.50 (0.77) 1.53 (0.61) HDL cholesterol ISP (0.39) 1.52 (0.31) 1.44 (0.37) mmol/l ISP 1.66 (0.45) 1.70 (0.45) 1.59 (0.45) Total 1.59 (0.42) 1.60 (0.39) 1.51 (0.41) LDL cholesterol ISP (0.73) 4.71 (0.73) 4.86 (0.81) mmol/l ISP 5.11 (1.02) 4.78 (0.96) 5.05 (0.93) Total 5.09 (0.88) 4.74 (0.84) 4.95 (0.87) p value for paired t-test = 0.04; p value for paired t-test = All results are expressed as means ± standard deviations (SD). Outliers were determined graphically using boxplots and Q-Q box. Observations lying beyond the whiskers of the boxplots are taken to be outliers. 3. Results female study Of the 54 women who were randomised into the study, 49 women completed the study. Three women discontinued the study and were not contactable to ascertain the reasons for this. There were five (5) drop-outs. No differences were found between the two female treatment groups at baseline for any of the variables measured (Table 1). The lipid results are summarised in Table 2. Only TC (mean: from 7.38 to 7.04; p =0.0003) and LDL cholesterol (mean: from 5.09 to 4.74; p =0.04) showed a significant reduction during the treatment phase. By week 18, after the 6 weeks off treatment, both total cholesterol and LDL cholesterol were seen to return to their baseline values (Figs 2 and 3). There were no significant differences between treatment groups for both triglycerides and HDL cholesterol in the 12 week treatment phase (Table 2). Urinary deoxypyridinoline was significantly reduced between weeks 0 and 12 (mean: from 8.90 ± 2.33 to 8.26 ± 1.87; p =0.01 from paired t-test). This observed change was due largely to treatment

4 254 R. Mackey et al. / Soy protein and elevated lipids Female study l Diet plus ISP + Low fat diet only Low fat diet only Diet plus ISP - WEEK 4 WEEK 0 WEEK 12 WEEK 18 Male study Low fat diet only Diet plus ISP + Low fat diet only WEEK 4 WEEK 0 WEEK 12 WEEK 18 Fig. 1. Study designs for the female and male studies. ISP = Soy protein with high isoflavone content; ISP+ = Soy protein with low isoflavone content. concentration(mmol/l) Mean LDL cholesterol Treatment ISP- ISP+ TIME (weeks) Fig. 2. Effects of diets containing soy protein with a large isoflavone content (ISP+) or a small isoflavone content (ISP ) on LDL cholesterol concentrations at baseline, 12 and 18 weeks in the female subjects. Values are displayed as means with standard error bars. group ISP, although no statistically significant difference was found between the groups. No changes in osteocalcin or bone-specific alkaline phosphatase levels were observed.

5 R. Mackey et al. / Soy protein and elevated lipids Mean total cholesterol concentration (mmol/l) Treatment ISP- ISP+ TIME(weeks) Fig. 3. Effects of diets containing soy protein with a large isoflavone content (ISP+) or a small isoflavone content (ISP ) on total cholesterol concentrations at baseline, 12 and 18 weeks in the female subjects. Values are displayed as means with standard error bars. Table 3 Effects of diets containing a large isoflavone content (ISP+) or a small isoflavone content (ISP ) on Sex Hormone Binding Globulin in the female subjects SHBG (nmol/l) 0 weeks 12 weeks 18 weeks ON TREATMENT OFF TREATMENT ISP+ (n =22) (22.89) (20.19) (18.74) ISP (n =24) (17.79) (19.18) (28.22) TOTAL (n =46) (20.17) (19.57) (25.34) p value for paired t-test = A significant reduction in LH levels was observed overall in the treatment phase (mean: from IU/L to IU/L; p = for paired t-test), but no significant differences were found between treatment groups. No change was observed with FSH or TSH, either in each treatment group or in the group as a whole. Results for SHBG are shown in Table 3. SHBG dropped significantly from baseline to 12 weeks of soy supplementation (mean: from ± to ± 19.57; p =0.002 for paired t-test). After excluding three outlying observations, no significant difference between treatment groups was found in the change SHBG from baseline to 12 weeks (2 sample t-test p =0.16). By week 18, that is 6 weeks after stopping the soy supplement, SHBG levels returned to their baseline values in both treatment arms. Menopause symptoms in both groups reduced by about 30% and there was no significant difference between ISP+ and ISP groups.

6 256 R. Mackey et al. / Soy protein and elevated lipids 4. Results male study Data from 27 men were available for analysis. The mean age of the subjects was 51.5 years and the mean baseline total cholesterol was 6.8 mmol/l. Of the lipid parameters measured, HDL cholesterol was the only lipid fraction in which a significant change was observed over the 12 week period of soy supplementation. This effect was observed at 6 weeks (mean = from 1.27 to 1.40; z =2.27, p =0.02 from regression model), and was still noticeable by 12 weeks (mean = from 1.27 to 1.36; z =2.17, p =0.03 from regression model). At 18 weeks, this effect was no longer seen (z =0.23, p =0.82 from regression model). No change in total cholesterol, LDL cholesterol or triglycerides was observed. A significant reduction in SHBG was observed during the treatment phase (from nmol/l to nmol/l; p = from paired t-test). This was seen to be independent of change in weight. SHBG levels had returned to baseline values by 18 weeks, that is, 6 weeks after stopping the soy supplement. No significant change was observed in TSH levels during the treatment phase. An increase in DHEA was observed over the twelve week soy supplementation period, although the increase only marginally failed to reach statistical significance (mean = from 4.01 to 4.28; p = 0.06 from paired t-test). No effect on FSH, LH, androstenedione or testosterone was seen. 5. Discussion Our results are consistent with those published elsewhere that soy has a significant cholesterol lowering effect [3]. In our study we found no difference between the ISP+ and ISP treatment groups, suggesting that isoflavones are not responsible for the lipid lowering effect of soy protein. Our results are consistent with those of Nestel [8] who observed no change in lipids after 10 weeks of purified soy isoflavones. Thus it would seem that the cholesterol lowering effect of soy protein is probably due to some other component or components of soy protein. Saponins have been suggested as a possible cholesterol lowering agent however saponins are extractable using alcohol and so saponin levels were very low in the ISP protein. The male study was designed as a pilot. The mean increase in HDL-C was 9.5%. Most importantly there was no measurable effect of soy protein on total serum levels of androgens but there was a significant reduction in SHBG with return to baseline levels after coming off the soy protein. Thus it would seem that soy protein does not have a feminizing effect on men but rather a lowering in SHBG levels would in fact increase biologically active free testosterone suggesting an androgenic effect. The observed decrease in SHBG in both our male and female studies is a unique result. SHBG levels are higher in vegetarians consuming high fibre diets [11] however this may be an effect of lignans rather than isoflavones. Possible mechanism for lowering SHBG levels include hypothyroidism, anti-oestrogen effect, and increased insulin. We were unable to show any effect on TSH or glucose levels in this study. A small but significant effect on the bone turnover marker, deoxypyridinoline was found This effect was small and probably not clinically relevant and no beneficial effect was seen on other bone markers. It would seem unlikely that soy protein in this dose will protect against osteoporosis. In conclusion soy protein significantly lowers cholesterol in both men and women probably via an isoflavone-independent mechanism. A surprise finding from these studies was that soy protein inhibited SHBG levels. Acknowledgement The authors would like to thank the Sanitarium Health Food Company for its financial support of the study and Protein Technology International for supplying the two soy proteins used in the study.

7 R. Mackey et al. / Soy protein and elevated lipids 257 References [1] K.A. Matthews, E. Meilahn, L.H. Kuller, S.F. Kelsey, A.W. Caggiula and R.R. Wing, Menopause and risk factors for coronary heart disease, N Engl J Med 321 (1989), [2] E. Barrett-Connor and T.L. Bush, Estrogen and coronary heart disease in women, JAMA 265 (1991), [3] J. Anderson, B. Johnstone and M. Cook-Newell, Meta-analysis of the effects of soy protein intake on serum lipids, N Engl J Med 333 (1996), [4] World health statistics annual 1993, Geneva, World Health Organisation, [5] M.S. Anthony, T.B. Clarkson, C.L. Hughes, T.M. Morgan and G.L. Burke, Soybean isoflavones improve cardiovascular risk factors without affecting the reproductive system of peri-pubertal rhesus monkeys, J Nutr 126 (1996), [6] J.L. Tang, J.M. Armitage, T. Lancaster, C.A. Silagy, G.H. Fowler and H.A.W. Neil, Systematic review of dietary intervention trials to lower blood total cholesterol in free-living subjects, BMJ 316 (1998), [7] M.R. Law, N.J. Wald and S.G. Thompson, By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease? BMJ 308 (1994), [8] P. Nestel, T. Yamashita, T. Sasahara, S. Pomeroy, A. Dart and P. Komesaroff et al., Soy isoflavones improve systemic arterial compliance but not plasma lipids in menopausal and perimenopausal women, Arteriosclerosis, Thrombosis, and Vascular Biology. 17 (1997), [9] S.M. Potter, R. Jimenez-Flores, J. Pollack, T.A. Lone and M.D. Berber-Jiminez, Protein-saponin interaction and its influence on blood lipids, J Agric. Food Chem 41 (1993), [10] Scandinavian Simvastatin Survival Study Group, Randomised trial of cholesterol lowering in 4444 patients with coronary artery disease: The Scandinavian Simvastatin Survival Study (4S), Lancet 344 (1994), [11] H. Adlercreutz, T. Fotsis, C. Bannwart, K. Wahala, T. Makela and G. Brunow et al., Determination of urinary lignans and phytoestrogen metabolites, potential antiestrogens and anticarcinogens, in urine of women on various habitual diets, J Steroid Biochem 25 (1986), [12] R.L. Divi, H.C. Chang and D.R. Doerge, Anti-thyroid isoflavones from Soybean, Chemical Pharmacology 54 (1997),