Polycystic ovary syndrome—a hormonal disorder that can manifest as irregular menstrual cycles, increased testosterone levels, or enlarged ovaries with numerous cysts—is a leading cause of infertility in women, affecting up to one in five of childbearing age. Yet the underlying mechanisms and causes of PCOS remain poorly understood.
PCOS often runs in families. Up to 70 percent of daughters of women with PCOS also develop it, but genetic variation doesn’t fully explain the high incidence within families—some genome-wide association studies of PCOS susceptibility reckon genetics explains less than 10 percent of the condition’s heritability. That has scientists suspecting that other factors, such as epigenetic mechanisms, might play a role in passing the condition on to future generations.
A study published this week (February 3) in Cell Metabolism suggests that mice can pass down PCOS-like symptoms for at least three generations. This is likely transmitted as epigenetic modifications, which—like a set of instructions dictating which genes are to be expressed—are inherited from parents to offspring. The researchers also analyzed blood from women with PCOS, reporting that the samples exhibited epigenetic alterations similar to those observed in the mouse models.
After two or three weeks of treatment, this epigenetic drug was able to restore both the ovulation of these animals and significantly improve their metabolic alterations. I was not expecting such a rapid recovery.—Paolo Giacobini, French National Institute of Health and Medical Research
To Margrit Urbanek, a geneticist specializing in polycystic ovary syndrome at Northwestern University, “the paper does demonstrate a clear transgenerational epigenetic inheritance of PCOS-like traits in mice,” she writes to The Scientist in an email. “This suggests that it could also be the case in humans but does not prove it,” adds Urbanek, who wasn’t involved in the study.
In previous research, developmental neuroscientist Paolo Giacobini of the French National Institute of Health and Medical Research and his colleagues had identified a potential cause of PCOS. Injecting pregnant mice with excess anti-Müllerian hormone (AMH) would cause their young to develop PCOS symptoms, including elevated testosterone, irregular reproductive cycles, and smaller litters. They also found AMH levels were elevated in pregnant women with PCOS compared to healthy pregnant women. Evidently, “we have a pathology that develops in the womb of the mother and is most likely transmitted to the fetus due to . . . this abnormal in utero [environment],” says Giacobini.
In the new research, he and his colleagues set out to investigate if mice could pass these PCOS-like symptoms down multiple generations. To do so, they mated male and female animals whose mothers had been exposed to AMH in the womb, then mated their female offspring with males exposed to AMH in utero, creating a third generation of mice. For this last generation, the most recent fetal AMH exposure was to their father and maternal grandparents. To Giacobini’s surprise, when examining around 15 third-generation females, “we saw all reproductive defects that are typical of PCOS,” he recalls. They had ovulatory dysfunction, smaller-than-usual litters, as well as some metabolic symptoms that occur in people with PCOS, such as weight gain and features of type 2 diabetes. The team observed similar ovulation and fertility defects when they repeated the breeding protocol with healthy males from lineages that had never been exposed to excess AMH, “suggesting that those defects were inherited from the . . . females,” Giacobini says.
Methylation profiles of PCOS
Suspecting that the symptoms might be transmitted through epigenetic mechanisms, the team took a close look at the third-generation females’ genomes using techniques to survey the genomes’ methylation patterns. Methylated genes are typically repressed. The researchers revealed a scattering of genes with unusual methylation patterns, often with strikingly low levels of methylation compared to the methylation profiles of control mice. The most affected gene pathways were involved in regulating reproductive and metabolic functions, including insulin signaling and inflammation, says Giacobini. (He and several of his colleagues disclosed having submitted a patent application involving methods to diagnose and treat PCOS.)
Notably, the methylation-deprived genes included Tet1, which encodes ten-eleven translocation methylcytosine dioxygenase, an enzyme critical for methylation removal.
They also conducted DNA analyses on blood samples from more than 30 women with PCOS—some of whom were the daughters of mothers with PCOS. Interestingly, they found that Tet1 was among several genes that exhibited the same methylation-deprived signature they had observed in mice.
“We really think that [Tet1] is one of the key genes whose alteration could be responsible for the . . . hypomethylation that we see in both the PCOS mice and the women with PCOS,” Giacobini says. The team hypothesizes that an initial burst of AMH in the womb—either directly or indirectly through effects on other hormones—somehow reprograms the methylation status of key genes such as Tet1, thereby influencing the expression of other genes, leading to PCOS symptoms.
It supports the idea that AMH has a direct role in the pathogenesis of PCOS rather than be a consequence of PCOS.—Margrit Urbanek, Northwestern University
Further bolstering this hypothesis, the team discovered that they could reverse some of the symptoms in third-generation females by injecting them with a naturally occurring, methylation-promoting molecule called S-adenosylmethionine. Remarkably, “after two or three weeks of treatment, this epigenetic drug was able to restore both the ovulation of these animals and significantly improve their metabolic alterations,” Giacobini says. “I was not expecting such a rapid recovery.”
To Elisabet Stener-Victorin, a reproductive physiology researcher at the Karolinska Institute, this last result is particularly intriguing. Noting that the agent used in the study is a relatively unspecific epigenetic modifier that affects all cells, perhaps it could be worthwhile exploring the effect and mechanisms of more targeted agents in future studies with the eventual aim of finding treatments for PCOS.
Perhaps the most intriguing finding “is that they can translate this and see at least some of these epigenetic changes in the serum of women with PCOS,” she says, adding that some of her previous research has also found epigenetic changes in PCOS patients. Such discoveries could pave the way for developing potential biomarkers to predict PCOS, adds Stener-Victorin, who has collaborated with Giacobini in the past but wasn’t involved in the new research.
The role of AMH in PCOS
Urbanek writes that she finds the close resemblance between the AMH-induced mouse model and human patients striking, as well as the fact that the epigenetic effects were maintained so strongly for three generations. “It supports the idea that AMH has a direct role in the pathogenesis of PCOS rather than be a consequence of PCOS,” she writes. “It also raises a cautionary flag of what a significant impact hormone exposure can have on a multigenerational time scale. For humans this means that a woman’s hormone exposure would still be impacting her grandchildren and great grandchildren 50 or more years later.”
Both Urbanek and Stener-Victorin say they’re curious to know more about the underlying mechanisms—for instance, how exactly AMH causes these epigenetic changes to begin with, and whether it’s directly caused by AMH or indirectly through a resulting elevation of testosterone, which was elevated in the second- and third-generation offspring mice, Stener-Victorin notes.
Another puzzle is how this epigenetic pattern is maintained across generations: Is it transmitted directly via the germ cells, or are elevated hormone concentrations somehow generating the epigenetic modifications in each generation? To answer this question, it would be helpful to see epigenetic analyses from germ cells specifically, rather than whole ovaries, as well as data on AMH concentrations in the second and third generation, both researchers note. It’s “important to separate what is driven by the germ cells and what is an in-utero effect,” Stener-Victorin says. Another limitation of the research, she adds, is that the team compares the methylation patterns in second- and third-generation rodents with those of first-generation control mice; ideally, each generation is compared with an equivalent generational control.
An outstanding question Giacobini hopes to answer is whether such epigenetic effects could influence male progeny as well. His study focused on female mice because, by definition, PCOS affects women. But if PCOS is mediated by changes in the expression of influential epigenetic players such as Tet1, then that could affect all offspring—as some of his lab’s preliminary data already suggest, he says. So far, because clinicians rarely follow up about the health of boys born to women with PCOS—let alone track them over time—“we don’t really know in humans whether the sons of women with PCOS might [also] have long-term health consequences.”
N. El Houda Mimouni et al., “Polycystic ovary syndrome is transmitted via a transgenerational epigenetic process,” Cell Metab, doi:10.1016/j.cmet.2021.01.004.