Background: Isoflavones are estrogen-like plant compounds that may protect against cardiovascular disease and endocrine-responsive cancer. Isoflavones may, because of their ability to act as selective estrogen receptor modulators, alter insulin-like growth factor (IGF) status.
Objective: The aim of this study was to assess the effect of 1-month isoflavone supplementation (86 mg/day red clover-derived isoflavones) on IGF status.
Design and subjects: Healthy pre- (n=16) and postmenopausal (n=7) women were invited to take part in a randomised, placebo-controlled crossover study with a minimum 2-month washout period.
Results: For premenopausal subjects, the change in IGF-1, IGF-BP1 and IGF-BP3 assessed at different points of the menstrual cycle did not differ between isoflavone and placebo phase. However, the change in IGF-1, when examined pre- and post-supplementation, was nonsignificantly reduced (P=0.06) on the isoflavone supplement compared to placebo. For postmenopausal subjects, the change in IGF-1, IGF-BP1 and IGFBP-3 concentrations over the supplementation period did not differ between isoflavone or placebo phase. Isoflavones increased HDL in postmenopausal women compared to placebo (P=0.02) but did not alter either cholesterol or triacylglycerol concentrations, and had no effect on antioxidant status.
Conclusions: This study shows that 1-month supplementation with red clover isoflavones has a positive effect on HDL cholesterol, but at most a small effect on IGF status in premenopausal and no effect in postmenopausal subjects. Further studies are required to ascertain the role these dietary compounds may have to play in breast cancer prevention.
Sponsorship: The isoflavone and placebo supplements were kindly supplied by Novogen (North Ryde, Sydney, Australia). This study was supported by Action Against Breast Cancer registered charity number 1020967.
High concentrations of insulin-like growth factor-1 (IGF-1) in serum are associated with an increased risk of premenopausal breast cancer (RR 7.28) (Hankinson et al, 1998). Tamoxifen has been shown to lower IGF-1 concentrations in breast cancer patients by up to 30% (Pollak et al, 1990; Campbell et al, 2001). Isoflavones are a class of phytoestrogens that may have a variety of potential health benefits in endocrine-responsive cancers and cardiovascular disease (Bingham et al, 1998; This et al, 2001). They are obtained in the diet predominantly from soy products (Bingham et al, 1998). Isoflavones are structurally similar to selective estrogen receptor modulators, for example, Tamoxifen. They may therefore be able to alter concentrations of IGF-1 and its binding proteins.
The aim of this study was to determine whether 1 month of isoflavone supplementation using red clover-derived isoflavones could alter serum concentrations of IGF-1 and its related binding proteins. The effect of menstrual cycle on IGF profile was accounted for in the study design. We also assessed whether the isoflavone supplement produces an alteration in lipid or antioxidant status as effects on these parameters have been demonstrated in some studies (Knight et al, 1999; Samman et al, 1999; Merz-Demlow et al, 2000; Takatsuka et al, 2000), but not in others (Nestel et al, 1999; Howes et al, 2000; Jenkins et al, 2000; Simons et al, 2000). The inclusion of a food frequency questionnaire (FFQ) and a 7-day food diary completed in each phase allowed the assessment of normal nutritional intake.
Healthy female subjects between the ages of 25 and 65 y were recruited from staff and students within University College London and the general public (through poster and radio adverts). The following eligibility criteria were applied:
Premenopausal—not on oral contraceptives, no antibiotics within the previous 4 months, a regular menstrual cycle of approximately 28 days.
Postmenopausal—not on hormone replacement therapy, no antibiotics within the previous 4 months, and postoophorectomy or amenorrhoea for at least 12 months.
The study was of a randomised, double-blind, placebo-controlled crossover design (Figure 1). Pre- and postmenopausal subjects were randomised to receive either placebo or 86 mg of red clover isoflavones daily (Promensil, Novogen, Australia, purified isoflavone pills containing 43 mg of total isoflavones consisting of 25 mg biochanin, 8 mg formononetin, 4 mg genistein and 5 mg daidzein) for one menstrual cycle followed by a 2-month washout period and the alternative intervention. During the intervention phases, subjects were asked to maintain their normal diet and to avoid the use of soy-containing foods. The supplement was taken as two pills, once a day. To monitor compliance, subjects were asked to return any unused pills. At the end of each intervention phase, subjects were asked if they had experienced any ill effects and whether their menstrual cycle had altered in length.
Fasting blood samples were obtained from postmenopausal subjects at baseline and at 28 days in each phase of the study. Fasting blood samples were collected at baseline, days 1–3, days 6–8, days 12–15, days 21–23 and days 26–28 from premenopausal subjects to assess the effects of the menstrual cycle on the end points measured. For all subjects, a 24-h urine sample was collected at baseline and at 28 days in each phase. In addition, a fully validated FFQ (McKeown et al, 2001) was filled in by all subjects at the baseline stage of each phase and a 7-day food diary completed over the last 7 days of each phase. All biological samples were stored at −80°C until analysis. The study was approved by the Royal Free and University College London Ethical Committee.
A variety of end point measures were assessed. Concentrations of IGF-1, IGF-BP1 and IGF-BP3 were measured in serum using ELISA methods (Diagnostic Systems Laboratories). Daidzein and genistein isoflavones and the isoflavone metabolite equol were assessed in urine at baseline (days 1–3 in premenopausal women) and at the end (days 26–28 in premenopausal women) of each supplementation period by GC-MS according to Morton et al (1994). Urinary creatinine was assessed by a colorimetric kit (Sigma Diagnostics). Serum lipids were assessed, again at baseline and at the end of each supplementation phase, using standard spectrophotometric assays. Antioxidant status was assessed by the measurement of vitamin E (Craft et al, 1992), malondialdehyde (Young & Trimble, 1991) and vitamin C (Vuillemier & Keck, 1989).
Fully validated EPIC FFQs and 7-day food diaries (McKeown et al, 2001) were analysed using DietPlan 5 for Windows 98.
The intervention was double-blinded and was not unblinded until all analyses, including statistical analyses, had been completed. Pre- and postmenopausal women were considered separately in the analyses. Changes from baseline were compared between treatment groups using a paired samples t-test (for two time-point comparisons) or repeated measures ANOVA (for multiple time-point comparisons). The association of continuous variables was assessed using Pearson correlation coefficients. Subjects with urinary equol concentrations post-isoflavone supplement of >1 mg/ml (Lampe et al) were defined as good equol excretors. For the premenopausal treatment subjects, at the 5% level of significance, the study had 80% power to detect as statistically significant a 35% difference in change in IGF-1 between the treatment groups over the supplementation period.
A total of 16 pre- and seven postmenopausal subjects completed both phases of the study. Mean (s.d.) age was 34 (9) y in premenopausal and 57 (6) y in postmenopausal subjects. Mean (s.d.) body mass index was 23.3 (4.3) kg/m2 in premenopausal and 24.4 (5.0) kg/m2 in postmenopausal subjects. The waist-to-hip ratio was 0.75 (0.04) in premenopausal and 0.77 (0.05) in postmenopausal subjects. Among premenopausal subjects, three out of 16 were parous, and among postmenopausal subjects, five out of six were parous. The percentage of total energy derived from protein, carbohydrate, fat and alcohol was 14.1, 41.8, 38.8 and 5.3 in premenopausal and 14.3, 47.0, 35.8 and 3.0 in postmenopausal subjects. The mean (range) number of wash-out days was 95 (56, 206) in premenopausal subjects and 70 (62, 91) in postmenopausal subjects. One menstrual cycle length in premenopausal women was similar in the two treatment groups.
Excretion of the isoflavones genistein, daidzein and equol was expressed as micromoles per mole of creatinine. Median (interquartile range) excretions in premenopausal subjects at baseline and after supplementation were as follows: placebo baseline: genistein 127.2 (45.4, 266.9), daidzein 147.8 (48.0, 308.7), equol 20.8 (13.3, 56.9); placebo 28 days: genistein 186.0 (137.4, 476.5), daidzein 102.5 (63.7, 194.2), equol 49.9 (17.8, 57.5); supplement baseline, genistein 319.1 (39.5, 731.5), daidzein 94.0 (66.9, 231.4), equol 32.2 (2.6, 82.8); supplement 28 days: genistein 1523.1 (793.7, 2254.5), daidzein 1785.7 (925.1, 3416.1), equol 443.3 (152.2, 760.9); (μmol/mol creatinine). Median (interquartile range) excretions in postmenopausal subjects at baseline and after supplementation were as follows: Placebo baseline: genistein 155.3 (104.0, 333.8), daidzein 259.4 (187.4, 496.6), equol 67.2 (34.5, 73.5); placebo 28 days: genistein 617.0 (141.8, 785.9), daidzein 195.0 (122.2, 562.3), equol 58.0 (25.5, 127.9); supplement baseline: genistein 411.0 (336.9, 517.9), daidzein 301.9 (100.5, 794.2), equol 50.5 (45.3, 63.2); supplement 28 days: genistein 3538.2 (1461.1, 6756.3), daidzein 5717.0 (2464.9, 5932.5), equol 704.8 (282.3, 2019.6); (μmol/mol creatinine). The change in concentration of genistein, daidzein and equol over the supplementation period was significantly larger during the supplement phase compared to placebo in both pre- and postmenopausal subjects. As defined above, 23% of premenopausal subjects were equol excretors compared to 20% of postmenopausal subjects.
For premenopausal subjects, while there was an effect of time on IGF-1, IGF-BP1 and IGF-BP3 confirming that the IGF profile is influenced by menstrual cycle, this did not differ between placebo and isoflavone supplement (Table 1a). When the change in IGF-1 over the whole supplementation period was compared between the isoflavone and placebo phases, there was a nonsignificant reduction in change in IGF-1 (P=0.06) on isoflavone supplement compared to placebo. However, this may be because of the nonsignificantly higher baseline concentrations of IGF-1 during the isoflavone phase. In postmenopausal subjects, there was no effect of isoflavone supplementation in comparison with placebo on IGF-1, IGF-BP1 or IGF-BP3 (Table 1b). Equol excretion (in μg/day) was associated with IGF-BP3 concentrations in postmenopausal women both at the end of the placebo phase (r=0.895, P=0.04) and supplement phase (r=0.984, P=0.002).
Total cholesterol and triacylglycerol were not affected by the supplement in either pre- or postmenopausal women. However, HDL-cholesterol concentrations were significantly elevated by isoflavone supplementation in the postmenopausal subjects (Tables 2a and b). There were no significant differences in any of the antioxidant end points assessed in either pre- or postmenopausal women (Table 2a and b).
When nutrient intake was assessed by an FFQ or a 7-day food diary, there was no significant difference in energy, carbohydrate, fat, SFA, PUFA, MUFA, cholesterol, fibre, vitamin E or carotene intake between placebo or supplement phase, whether assessed by an FFQ or a food diary in pre or postmenopausal subjects. For protein and vitamin C intake, there was no significant difference between phases when assessed by an FFQ. However, when assessed by a food diary, protein intake (in grams) was lower during the supplement phase than the placebo phase in premenopausal women (77 (78) vs 69 (19) g; placebo vs supplement, mean (s.d.) P<0.05), while vitamin C intake was lower during the supplement phase than the placebo phase in postmenopausal women (105 (41) vs 74 (30) mg; placebo vs supplement, mean (s.d.) P<0.05).
Red clover isoflavone supplementation had no effect on IGF-1, IGF-BP1 or IGF-BP3 in postmenopausal women. In premenopausal women, isoflavone supplementation had no effect when multiple time points over the menstrual cycle were considered, although there seemed to be a slight protective effect of isoflavone supplementation when the analysis was carried out simply before and after the supplementation period. This is likely to be due, however, to the large difference in IGF-I levels at baseline and cannot be attributed to the isoflavone supplement. The effect of isoflavone supplementation on IGF status would therefore appear to be relatively weak. The clinical relevance of this small change may, in fact, be negligible. Our numbers were small, and the variance observed in IGF-1 and its binding proteins was high, so the area clearly required further study.
IGF appears to be influenced by the stage of the menstrual cycle (Juul et al, 1997; Helle et al, 1998; Wangen et al, 2000). We observed a significantly higher concentration of IGF-1 and IGF-BP3 in premenopausal subjects on placebo in the luteal phase (days 21–23) compared to the menstrual phase (days 1–3). These results indicate that controlling closely for menstrual cycle phase may be necessary to observe IGF changes of small magnitude.
This study has used a red clover-derived isoflavone supplement over a relatively short time period (1 month). There is a possibility that a regular soy diet may have different effects on IGF-1 than this short-term supplement.
Four other studies have examined the effect of soy on IGF status (unpublished observations, Wangen et al, 2000; Aukema and Housini, 2001; Woodside & Campbell, 2001; Khalil et al, 2002). In a rat model of polycystic kidney disease, soy protein feeding resulted in reduced IGF-1 in both normal and affected male animals (Aukema & Housini, 2001). In a 1-week human supplementation study using a soy, rye and linseed bar, both IGF-1 and IGF-BP3 were significantly elevated after intervention with the bar (unpublished observations, Woodside & Campbell, 2001), although this may have been because of increased kilocalorie and protein intake (Clemmons & Underwood, 1991). However, in a soy protein supplementation study in men, with a milk protein control, serum IGF-1 was higher in those supplemented with soy protein than those consuming milk protein (Khalil et al, 2000), suggesting that soy protein had an effect on IGF-1 independent of its protein content. By contrast, in a 3-month study looking at markers of bone turnover (Wangen et al, 2000), in premenopausal women, IGF-1 and IGF-BP3 were increased by a low isoflavone diet, while in postmenopausal women, there were trends towards decreased IGF-1 and IGF-BP3 concentrations with increasing isoflavone concentration. The authors concluded that, although soy isoflavones do affect markers of bone turnover, the changes observed were of small magnitude and were not likely to be clinically relevant.
We have shown an effect of isoflavone supplementation within 1 month on HDL-cholesterol in postmenopausal women. Again, numbers were small, and the difference between placebo and isoflavone phase only reached significance at the 5% level in postmenopausal women, so the area clearly requires further study. Several studies have provided evidence that phytoestrogens in general can modulate plasma lipid and lipoprotein concentrations, but the data are not entirely consistent (Anderson et al, 1995). A recent overview concluded that soy could improve blood lipid parameters in both normocholesterolaemic and hypercholesterolaemic subjects, although the use of soy alone may not allow patients with hyperlipidaemia to achieve target lipid parameters (Costa & Summa, 2000). Soy, with which most of the studies have been carried out, has a variety of properties unrelated to isoflavone content that may contribute to its lipid-modulating properties (Potter, 1995). Our postmenopausal subjects had normal lipid levels, but displayed an increase in HDL-cholesterol during the isoflavone supplementation phase. This may be pertinent in light of the fact that, although one recent short-term study using the same purified red clover isoflavone pills found no effect on plasma lipid and lipoprotein concentrations (Nestel et al, 1999), other longer term studies using this supplement have demonstrated an effect of isoflavones on HDL3 (Samman et al, 1999) in premenopausal women and total HDL (Knight et al, 1999; Clifton-Bligh et al, 2001) in menopausal and postmenopausal women. The apparent difference in these findings compared to those obtained with soy-derived isoflavones could be accounted for by the high levels of biochanin and formononetin in red clover isoflavone extracts.
Isoflavones can act as antioxidants (Bingham et al, 1998), both in in vitro test systems (Kapiotis et al, 1997; Lissin & Cooke, 2000) and in vivo (Tikkanen et al, 1998; Jenkins et al, 2000; Wiseman et al, 2000). However, the data are not entirely consistent, as other studies have not shown evidence of antioxidant effects in vivo (Hodgson et al, 1999; Samman et al, 1999; Hsu et al, 2001). We found little effect of isoflavone supplementation on markers of antioxidant status. Our subjects were healthy and had normal antioxidant levels; it may be that isoflavones will exert marked antioxidant effects in vivo only in those with an already compromised antioxidant status.
Background diet did not seem to affect the findings of this study. Protein and vitamin C intakes differed significantly by phase in pre- and postmenopausal subjects, respectively, but only when assessed by food diary and not by FFQ. The magnitudes of the differences observed are unlikely to affect our conclusions.
Red clover-derived isoflavone supplementation over 1 month has no effect on IGF status in postmenopausal women, and at most has a small effect on IGF-1 in premenopausal women. Red clover isoflavones seem to increase HDL-cholesterol in postmenopausal women, but have little or no effect on other lipids or antioxidant status. It must be remembered, in the interpretation of these data, that the study was small and had low power to detect a difference between placebo and supplement, particularly when multiple comparisons were being carried out. Further work in larger numbers will determine whether these dietary compounds may play a role in endocrine-responsive cancer and CVD prevention.
Anderson JJB, Anthony M, Messina M & Garner SC (1999): Effects of phyto-estrogens on tissues. Nutr. Res. Rev. 12, 75–116.
Anderson JW, Johnstone BM & Cook-Newell ME (1995): Meta-analysis of the effects of soy protein intake on serum lipids. N. Engl. J. Med. 332, 276–282.
Aukema HM, Housini I (2001): Dietary soy protein effects on disease and IGF-1 in male and female Han:SPRD-cy rats. Kidney Int. 59, 52–61.
Bingham SA, Atkinson C, Liggins J, Bluck L & Coward A (1998): Phyto-estrogens: where are we now? Br. J. Nutr. 79, 393–406.
Campbell MJ, Woodside JV, Secker-Walker J, Titcomb A & Leathem AJC (2001): IGF status is altered by Tamoxifen in breast cancer patients. Mol. Pathol. 54, 307–310.
Clemmons DR & Underwood LE (1991): Nutritional regulation of IGF-1 and IGF binding proteins. Ann. Rev. Nutr. 11, 393–412.
Clifton-Bligh PB, Baber RJ, Fulcher GR, Nery ML & Moreton T (2001): The effect of isoflavones extracted from red clover (Rimostil) on lipid and bone metabolism. Menopause 8, 259–265.
Costa RL & Summa MA (2000): Soy protein in the management of hyperlipidemia. Ann. Pharmacother 34, 931–935.
Craft NE, Wise SA & Soares JH (1992): Optimisation of an isocratic high performance liquid chromatography separation of carotenoids. J. Chromatogr. 589, 171–176.
Hankinson SE, Willett WC, Colditz GA et al (1998): Circulating concentrations of insulin-like growth factor-1 and risk of breast cancer. Lancet 351, 1393–1396.
Helle SI, Anker G, Meadows KA et al (1998): Alterations in the insulin-like growth factor system during the menstrual cycle in normal women. Maturitas 28, 259–265.
Hodgson JM, Puddey IB, Croft KD, Mori TA, Rivera J & Beilin LJ (1999): Isoflavonoids do not inhibit in vivo lipid peroxidation in subjects with high-normal blood pressure. Atherosclerosis 145, 167–172.
Howes JB, Sullivan D, Lai N et al (2000): The effects of dietary supplementation with isoflavones from red clover on the lipoprotein profiles of postmenopausal women with mild to moderate hypercholesterolaemia. Atherosclerosis 152, 143–147.
Hsu C-S, Shen WW, Hsueh Y-M & Yeh S-L (2001): Soy isoflavone supplementation in postmenopausal women. Effects on plasma lipids, antioxidant enzyme activities and bone density. J. Reprod. Med. 46, 221–226.
Jenkins DJA, Kendall CWC, Vidgen E et al (2000): Effect of soy-based breakfast cereal on blood lipids and oxidized low-density lipoprotein. Metabolism 49, 1496–1500.
Juul A, Anker G, Pederson AT et al (1997): Changes in serum concentrations of growth hormone, insulin, insulin-like growth factor and insulin-like growth factor-binding proteins 1 and 3 and urinary growth hormone excretion during the menstrual cycle. Hum. Reprod. 12, 2123–2128.
Kapiotis S, Hermann M, Held I, Seelos C, Ehringer H & Gmeiner BMK (1997): Genistein, the dietary derived angiogenesis inhibitor, prevents LDL oxidation and protects endothelial cells from damage from atherogenic LDL. Arterioscler. Thromb. Vasc. Biol. 17, 2868–2874.
Khalil DA, Lucas EA, Juma S, Smith BJ, Payton ME & Arjmandi BH (2002): Soy protein supplementation increases insulin-like growth factor-1 in young and old men but does not affect markers of bone metabolism. J. Nutr. 132, 2605–2608.
Knight DC, Howes JB & Eden JA (1999): The effect of Promensil, an isoflavone extract, on menopausal symptoms. Climacteric 2, 79–84.
Lissin LW & Cooke JP (2000): Phytestrogens and cardiovascular health. J. Am. Coll. Cardiol. 35, 1403–1410.
McKeown NM, Day NE, Welch AA et al (2001): Use of biological markers to validate self-reported dietary intake in a random sample of the European Prospective Investigation into Cancer United Kingdom Norfolk cohort. Am. J. Clin. Nutr. 74, 188–196.
Merz-Demlow BE, Duncan AM, Wangen KE et al (2000): Soy isoflavones improve plasma lipids in normocholesterolemic premenopausal women. Am. J. Clin. Nutr. 71, 1462–1469.
Morton MS, Wilcox G, Wahlqvist ML & Griffiths K (1994): Determination of lignans and isoflavonoids in human female plasma following dietary supplementation. J. Endocrinol. 142, 252–259.
Nestel PJ, Pomeroy S, Kay S et al (1999): Isoflavones from red clover improve systemic arterial compliance but not plasma lipids in menopausal women. J. Clin. Endocrinol. Metab. 84, 895–898.
Pollak M, Costantino J, Polychronakos C et al (1990): Effect of tamoxifen on serum insulin like growth factor-1 levels in stage I breast cancer patients. J. Natl. Cancer. Inst. 82, 1693–1697.
Potter SM (1995): Overview of proposed mechanisms for the hypocholesterolemic effect of soy. J. Nutr. 125, S606–S611.
Samman S, Lyons Wall PM, Chan GSM, Smith SJ & Petocz P (1999): The effect of supplementation with isoflavones on plasma lipids and oxidisability of low density lipoprotein in premenopausal women. Atherosclerosis 147, 277–283.
Simons LA, von Konigsmark M, Simons J & Celermajer DS (2000): Phytestrogens do not influence lipoprotein levels or endothelial function in healthy postmenopausal women. Am. J. Cardiol. 85, 1297–1301.
Takatsuka N, Nagata C, Kurisu Y, Inaba S, Kawakami N & Shimizu H (2000): Hypocholesterolemic effect of soy milk supplementation with usual diet in premenopausal normolipidemic Japanese women. Prev. Med. 31, 308–314.
This P, de al Rochefordiere A, Clough K, Fourquet A & Magfelenat H (2001): Phytestrogens after breast cancer. End Relat. Cancer 8, 129–134.
Tikkanen MJ, Wahala K, Ojala S, Vihma V & Adlercreutz H (1998): Effect of soybean phytestrogen intake on low density lipoprotein oxidation resistance. Pro. Natl. Acad. Sci. USA 95, 3106–3110.
Vuillemier JP & Keck E (1989): Fluorometric assay of vitamin C in biological materials using a centrifugal analyser with fluorescence attachment. J. Micronut. Anal. 5, 25–34.
Wangen KE, Duncan AM, Merz-Demlow BE et al (2000): Effects of soy isoflavones on markers of bone turnover in premenopausal and postmenopausal women. J. Clin. Endocrinol. Metab. 85, 3043–3048.
Wiseman H, O'Reilly JD, Adlercreutz H et al (2000): Isoflavone phytoestrogens consumed in soy decrease F2-isoprostane concentrations and increase resistance of low-density lipoprotein to oxidation in humans. Am. J. Clin. Nutr. 72, 395–400.
Young IS & Trimble ER (1991): Measurement of malondialdehyde in plasma by high performance liquid chromatography with fluorimetric detection. Ann. Clin. Biochem. 28, 504–508.
The isoflavone and placebo supplements were kindly supplied by Novogen (North Ryde, Sydney, Australia). This study was supported by Action Against Breast Cancer registered charity number 1020967.
About this article
Molecular and Cellular Biochemistry (2015)