Epigenetics, Evolution, Endocrine Disruption, Health, and Disease

A possible mechanism for how DES exerts its action epigenetically has been proposed recently

2006 Study Abstracts

Endocrine-disrupting chemicals (EDCs) in the environment have been linked to human health and disease. This is particularly evident in compounds that mimic the effects of estrogens. Exposure to EDCs early in life can increase risk levels of compromised physical and mental health. Epigenetic mechanisms have been implicated in this process. Transgenerational consequences of EDC exposure is also discussed in both a proximate (mechanism) and ultimate (evolution) context as well as recent work suggesting how such transmission might become incorporated into the genome and subject to selection. We suggest a perspective for exploring and ultimately coming to understand diseases that may have environmental or endocrine origins.

That epigenetic mechanisms may play a role in endocrine disruption helps explain the transgenerational effects of some hormonally active chemicals. Treatment with diethylstilbestrol (DES) during pregnancy results in vaginal adenocarcinoma in female offspring in humans and mice. Female offspring of mice exposed to DES during pregnancy, when mated to control males, produce a second generation of females who, although not exposed to DES themselves, express this same rare genital tract cancer. This transgenerational transmission of a specific reproductive tract lesion would be hard to explain without invoking an epigenetic mechanism for heritable change and, given the finding of altered DNA methylation patterns in a specific uterine gene in mice treated developmentally with DES, we think a strong case can be made for such a conclusion. Newbold and colleagues next showed that specific, rare genital tract cancers (rete testes cancers) are also expressed and therefore transmitted to the male offspring of females treated in utero with DES. In colloquial terms, this demonstrated the occurrence of reproductive tract tumors in the grandsons and granddaughters of mothers treated with DES. A possible mechanism for how DES exerts its action epigenetically has been proposed recently. The transmission of uniquely specific changes in the program of development in mice has implications for similarly exposed humans as well as the biology of hormonally induced disease.

We have already mentioned that DES treatment of mice during development results in reproductive tract cancers, persistent up-regulation of key estrogen-responsive genes, and altered patterns of gene methylation in the affected genes and that the cancer can be transmitted through two generations. In addition to DES, methoxychlor has been reported to increase global DNA methylation in uterine ribosomal DNA after in utero exposure; the alteration in methylation remains months after treatment.

To understand the role of estrogens in development, there is hardly a more powerful model than that of the outcomes observed in humans and mice developmentally exposed to the synthetic estrogen DES. Female offspring of humans or mice exposed prenatally to DES have a risk for vaginal clear cell adenocarcinoma. The mechanisms underlying these developmentally induced lesions have been sought for three decades. There was the suggestion by clinical investigators that DES had altered the normal differentiation of the epithelial cells of the fetal cervix and vagina such that they responded abnormally to estrogen at puberty, because no cancers had been seen in prepubescent girls. Similarly, ovariectomy of developmentally DES-treated mice prevented the subsequent expression of uterine adenocarcinomas.

Epigenetic change in the molecular program of cell differentiation in the affected tissues may be a common mechanism. The clear cell cancers of the vagina in DES-exposed women displayed genetic instability consistent with epigenetic imprints in the absence of any expected mutation in classical oncogenes or tumor suppressor genes. Using a well validated mouse model for DES genital tract tumors, Li and colleagues discovered that one of the estrogen-inducible genes in the mouse uterus, lactotransferrin, that had been shown earlier to be persistently up-regulated by developmental DES exposure, had an altered pattern of CpG methylation in the promoter region of the gene upstream from the estrogen response element. Subsequent work demonstrated that other developmentally up-regulated genes such as fos and jun also had persistent changes in the pattern of methylation of the gene after DES exposure during development. These experiments raise the possibility that DES (and other environmental estrogens) alter the program of differentiation of estrogen target cells in the reproductive tract through an epigenetic mechanism.

Other studies support this hypothesis. In addition to cervicovaginal adenocarcinomas in female mice and humans exposed prenatally to DES and uterine adenocarcinoma in mice, it has been shown that developmental exposure to DES results in excess risk of uterine leiomyomas (fibroids) in mice, rats, and women. It was also recently reported that sea lions in areas contaminated with EDCs have a higher prevalence of uterine fibroids. The Eker rat carries a germ-line mutation in a tumor suppressor gene and is predisposed to uterine leiomyoma. Cook and colleagues used this model system to demonstrate a DES-induced alteration in developmental imprinting as analyzed by tumor suppressor gene penetrance, concluding that developmental programming by estrogen works in concert with preexisting genetic change. In a population of 819 black and 504 white women, fibroid status was determined by ultrasound screening or surgical record review, whereas prenatal DES exposure was determined by interview. DES-exposed women had a significantly greater risk for uterine fibroids and tended to have larger tumors. The authors conclude that their study, as well as animal studies, indicate a role for prenatal estrogen in the etiology of uterine leiomyoma in women.


  • Full study (free access) : Epigenetics, Evolution, Endocrine Disruption, Health, and Disease, Endocrinology, Volume 147, Issue 6, Pages s4–s10, doi.org/10.1210/en.2005-1122, June 2006.
  • Featured image academic.oup.

Have your say ! Share your views