Location: Arkansas Children's Nutrition CenterTitle: Unique Gene Expression Profiles in the Mammary Gland of Prepubertal and Adult Female Rats Treated With Estradiol or Soy Protein Isolate (SPI)) Author
Submitted to: Endocrine Reviews
Publication Type: Abstract only
Publication Acceptance Date: 3/1/2011
Publication Date: 5/1/2011
Citation: Ronis, M.J., Singhal, R., Shankar, K., Badeaux, J., Badger, T.M. 2011. Unique Gene Expression Profiles in the Mammary Gland of Prepubertal and Adult Female Rats Treated With Estradiol or Soy Protein Isolate (SPI). Endocrine Reviews. 32(3):OR34-4. Interpretive Summary: There is much argument in the popular press and scientific literature about the possible health effects of eating soy and soy-based foods such as edameme, tofu and soy infant formulas. On the one hand, people have argued that soy foods are cancer protective based on the lower incidence of certain cancers in Asian populations. On the other hand some have argued that soy foods promote breast cancer and cause infertility, as a result of the weakly estrogenic properties of a small number of phytochemicals, (the isoflavones genistein and daidzein), found only in soy products. The scientific arguments hinge on whether eating soy foods (as opposed to treatment with pure isoflavones) result in estrogenic responses. The current study was designed to answer this question by comparing global gene changes in breast tissue of female rats in response to treatment with the female sex hormone 17 beta-estradiol (E2) or to feeding the soy product, soy protein isolate (SPI) which is used to make soy infant formula and is used as a protein source in many processed foods. Two experiments were done – one in adult female rats ovariectomized (ovaries removed) to remove endogenous estrogens and one in intact prepubertal females where endogenous sex steroid levels are low. Rats were fed control diets with casein as the protein or diets where casein was replaced with SPI. Additional groups were treated with E2. Breast tissue was collected and global gene expression assessed by gene chip. In both experiments, there was little overlap in gene expression patterns generated in response to feeding SPI or to E2 treatment suggesting that SPI is not estrogenic. In groups with combined SPI feeding and E2 treatment it appeared that SPI feeding actually prevented some E2 actions, particularly on pathways involved in cell division. These results are consistent with the population-based studies in Asians suggesting that soy consumption reduces the risk of breast cancer.
Technical Abstract: Concerns have arisen regarding infertility and increased breast cancer risk in women consuming soy foods, primarily because of the perceived estrogenicity of soy isoflavones such as genistein and daidzein. Two studies were conducted in mammary gland to determine if consumption of soy products induces a gene expression profile similar to that of estradiol. In Experiment 1, Sprague-Dawley rats (N = 20/group) were fed AIN-93G diets with casein or SPI from PND30. From PND50-64, rats (N = 10/group) were ovariectomized and infused s.c. with 5 'g/kg/d 17 beta-estradiol (E2) to restore physiological levels. In Experiment 2, intact female rats were fed AIN-93G or SPI from PND21-33 or were additionally infused s.c. with 10 micro g/kg/d E2 (N = 10/group). Mammary glands were collected and total mRNA isolated for microarray analysis. Array data were normalized using GeneSpring. Statistical analysis of genes expressed greater than 1.5-fold in Experiment 1 revealed 651 genes regulated by E2 (P< 0.05) compared to 27 genes regulated by SPI with only 6 in common between the two treatments. In Experiment 2, 391 genes were regulated by E2 (P<0.05) and 71 by SPI with only 17 genes in common. Data were confirmed by real time RT-PCR and suggest that gene signatures of E2 and SPI in the mammary gland have little overlap in either adults or prior to puberty. Sub-sets of 164 and 116 E2-regulated genes were significantly modulated by SPI feeding in the E2 + SPI vs. E2 + casein groups. There was little evidence of additive effects, and particularly in the adults, SPI feeding actually antagonized E2 actions. In Experiment 1, SPI feeding inhibited E2-mediated induction of cyclin D1, the oncogene c-myc, several c-myc regulated genes and E2 induction of ER beta-specific genes, such as Knsl2 and Top2a. These data suggest that foods such as SPI, which are complex mixtures of proteins, peptides and many phytochemicals, are not estrogenic. In fact, anti-estrogenic actions of SPI on mammary cyclin D1 and c-myc pathways are consistent with epidemiological data demonstrating reduced breast cancer risk in Asian soy consumers.