Skip to main content
ARS Home » Southeast Area » Little Rock, Arkansas » Microbiome and Metabolism Research Unit » Research » Publications at this Location » Publication #208745
Title: Is Soy Estrogenic? Hepatic Gene Expression in the Presence or Absence of Endogenous Estrogen

Author
item SINGHAL, ROHIT - ACNC/UAMS
item SHANKAR, KARTIK - ACNC/UAMS
item BADGER, THOMAS - ACNC/UAMS
item RONIS, MARTIN - ACNC/UAMS

Submitted to: Federation of American Societies for Experimental Biology Conference
Publication Type: Abstract Only
Publication Acceptance Date: 2/15/2007
Publication Date: 4/28/2007
Citation: Singhal, R., Shankar, K., Badger, T.M., Ronis, M.J. 2007. Is soy estrogenic? Hepatic gene expression in the presence or absence of endogenous estrogen [abstract]. The FASEB Journal. 21(6):A1103.

Interpretive Summary: Soy foods contain compounds known as isoflavones, which have been reported to have effects on the body similar to estrogen hormones. However, there is controversy over whether soy exhibits estrogenic properties, and this is an important question because consuming estrogens can cause a host of medical disorders under certain conditions, especially during pregnancy and when estrogen-dependent cancers are present. To answer this question we conducted a study in which female rats were fed diets made with either casein (no soy) or soy as soy protein isolate (SPI) and ovaries were removed, followed by estrogen treatment (estradiol) for 14 days. This was done to see the effect of soy in the presence and absence of estrogens since on removing ovaries estrogen synthesis would be impaired and on estrogen supplementation physiological estrogen levels could be achieved. We performed a microarray analysis in the liver to determine which genes were switched on or off by estrogen and/or SPI. Our findings suggest that soy does have some effects of these genes, but only very weakly in the absence of estradiol. In the presence of estrogens (either supplemented or in the rats without ovariectomy), soy doesn't have any estrogenic effect on liver gene expression. These data expand our understanding of the synergistic, additive and interactive effects of soy on estrogen-mediated signaling and suggest that even when SPI is the total dietary protein source, has very few of the effects noted by estradiol. Therefore, SPI has little estrogenicity, especially in the presence of endogenous estrogens.

Technical Abstract: The objective of this study was to address the question of soy's estrogenicity by studying the hepatic gene expression (HGE) signature of diets made with soy protein isolate (SPI) fed in the presence or absence of estradiol (E2). Sprague-Dawley rats were ovariectomized (OVX) at PND50, infused with E2 using osmotic pumps for 14 d (5 'g/kg/d) and fed AIN-93G diets made with casein or SPI. Changes in HGE by E2 or SPI alone or together were examined by microarray analysis using rat 230 2.0 Affymetrix-GeneChip. In the OVX rats, E2 replacement resulted in significant (+/-2 fold, P<0.05) down-regulation of 191 genes and up-regulation of 110 genes, while SPI alone altered the expression of only 53 genes. Gene annotation revealed that out of 18 genes common between E2 and SPI, 95% were regulated similarly, suggesting a weak estrogenic effect of SPI in the absence of E2. A comparison between the effects of E2 in OVX and SPI in non-OVX rats revealed only 2 genes to be common suggesting almost no effect of SPI on estrogen responsive genes in the presence of E2. Only 2 genes were regulated in common between the groups fed with SPI with or without endogenous estrogens which suggests a completely different HGE signature in OVX and non-OVX rats fed SPI. These data expand our understanding of the synergistic, additive and interactive effects of soy on estrogen-mediated signaling and suggest that even when SPI is the total dietary protein source, there is very little estrogenicity, especially in the presence of endogenous estrogens.