Hormones are complicated. They work like a complex arrangement of dominos, one knocking over the next. Misplace one domino, and it all falls apart.

When trying to piece together hormone signaling pathways, scientists try to figure out how they are arranged by testing which dominoes fall when one is pushed over. They want to find that first domino — the one hormone that rules them all.

Scientists previously thought that they knew the elusive first dominos for reproduction. Estrogen regulates female reproduction, and testosterone regulates male reproduction. But as time went on, it became clear that things weren’t so black and white. Piraye Beim, the CEO of the precision medicine company Celmatix felt boxed in by this absolutism. 

Piraye Beim is the CEO of Celmatix, a precision medicine company specializing in ovarian health.

Credit: Celmatix

“We kept positioning ourselves as a women’s health therapeutics company, and there’s this perception that ‘Oh, so you’re doing estrogen blockers?’ We wanted to put out into the world this concept of ovarian health as a new therapeutic category,” said Beim. “And yes, the ovary is clearly very important for reproductive function, but it’s important for so much more. It really is the central command center of the endocrine system.”

Beim, like many other researchers, is turning her sights to a hormone called anti-Müllerian hormone (AMH). AMH is produced by the ovaries and critically regulates not only early reproductive development, but female fertility. Researchers in academia and industry are learning more about how AMH regulates the ovary and its eggs to develop therapeutics for infertility, polycystic ovarian syndrome (PCOS), and ovarian cancer in humans and animals.

“AMH and its receptor AMHR2 are really the missing hormonal pathway in the pantheon of key reproductive pathways that are regulating the ovaries,” said Beim. “We are seeing it emerge as a very exciting therapeutic target with broad applicability across women’s health.”

A new female hormone

Researchers discovered that AMH was involved in female development by accident. It was first identified as a hormone that led to male reproductive organ development by regulating the Müllerian duct, the structural precursor to the female reproductive tract (1).

About halfway through pregnancy, a fetus’ reproductive organs start to develop. Fetuses carrying two X chromosomes develop female reproductive organs, while fetuses carrying an X and a Y chromosome develop male organs. This is primarily driven by hormone production from the primordial gonads, which will eventually become ovaries or testes.

Expression of a certain gene on the Y-chromosome instructs the gonads to produce AMH, which prevents the Müllerian duct from developing into female reproductive organs. If a fetus with a Y chromosome doesn’t properly produce AMH, the fetus will not only develop male reproductive organs, but also a uterus and fallopian tubes; this is a condition called persistent Müllerian duct syndrome.

Several decades ago, clinicians studying newborns born with both male and female organs anticipated that females would not express AMH, which might provide them with a diagnostic marker for sex. To their surprise, every newborn they tested expressed AMH.

“Women kept testing positive for AMH, and there was this sort of head scratching amongst the folks who were developing the diagnostic test. They said ‘wait, this is supposed to be a tiebreaker test because it’s in males, but not females. What in the world is it doing in females?’” said Beim.

Female mice begin expressing AMH after birth. Once the reproductive organs are developed, AMH becomes a master regulator of a key ovarian structure: follicles.

Follicles are sacs of fluid that contain a developing oocyte. Follicles produce hormones that regulate the menstrual cycle, and when a mature oocyte is released from the follicle, if it is not fertilized within a day or two, it dies. When researchers made AMH knockout mice, female mice seemed normal and fertile. But when they looked a bit closer, they found that the female mice used up their supply of oocytes earlier in life than wild type mice (1).

“Just until the last 15 years, people assumed it was a male hormone, so there was no focus on its utility or application in women’s health. Now, it’s the number one biomarker to gauge everything from timeline to menopause to ovarian reserve,” said Beim. “It’s the bread-and-butter assay that’s used in fertility treatments.”

AMH levels taper off as women age and approach menopause, a reflection of their decreasing oocyte reserve. But AMH levels present a Goldilocks conundrum; AMH levels must be just right for the ovary to work properly. 

A promising suspect in PCOS

High AMH levels often indicate a condition called polycystic ovarian syndrome (PCOS). Patients with PCOS have enlarged ovaries and many follicles containing immature eggs. They often struggle to conceive, and they experience complications during pregnancy if they do. They also develop other symptoms such as cystic acne, rapid hair growth, and baldness. Patients are often overweight and carry increased risk for developing diabetes later in life. Currently, treatment consists of lifestyle changes and medication for symptoms. But a lack of clarity for what causes PCOS has made producing a drug targeting the root cause impossible.

Paolo Giacobini, a neuroendocrinologist from the University of Lille, is on a mission to find that root cause. He looked for answers in an unexpected place: the brain.

Paolo Giacobini, a neuroendocrinologist from the University of Lille, thinks dysregulation in AMH signaling may contribute to PCOS.

Credit: Paolo Giacobini

Reproduction is probably not the first word that comes to mind when thinking about the brain, but gonadotropin releasing hormone (GnRH) neurons in the brain produce hormones that direct the pituitary gland to produce other hormones. These hormones tell the ovaries to release mature eggs from their follicles. 

GnRH neurons don’t relay messages well in women with PCOS; they release hormones into the body inconsistently. Giacobini found that AMH partly controls how much and how often GnRH neurons release hormones, and high AMH levels may overstimulate this pathway in people with PCOS (2). The receptor for AMH, AMHR2, decorates GnRH neurons, allowing AMH to directly bind to and regulate those neurons. 

“You have a vicious cycle here that starts from the brain and impacts the ovaries, and then you basically induce ovulation and also increase testosterone levels,” said Giacobini.

AMH levels are almost always high in people with PCOS, but levels of lutenizing hormone (LH), which GnRH neurons produce to regulate egg release from the follicles, vary. Symptoms and hormone levels can differ depending on a patient’s weight. LH is high in women who are normal weight, but low in women who are overweight. This presents a complication for drug development: How can therapeutically targeting this mechanism help both groups of women? 

“The criticism that I had by some clinicians is that this treatment might not be interesting for women who are obese; it could only be interesting for women who are leaner. But I do not agree because if you look at the pulsivity of LH, it is elevated regardless of if women are obese or lean,” said Giacobini.

Giacobini wondered if AMH’s effects on GnRH neurons could explain why PCOS seems to be heritable. Since healthy pregnant women have low levels of AMH that decline as pregnancy progresses, clinicians assumed that this would also be the case in women with PCOS. But when he and his team checked AMH levels in pregnant women with PCOS, their levels increased during pregnancy (3). He wanted to know if increasing AMH levels were “a sufficient or contributing factor to the transmission of disease over multiple generations.”

Emely Larios, a reproductive biology researcher from the University of Texas MD Anderson Cancer Center, captured this image of a normal adult ovary section from a female bat. The large, light pink circles are follicles containing an oocyte. Women with PCOS have an excessive number of these follicles, which make it appear that ovaries are decorated with cysts.

Credit: EmeLy Larios

Giacobini and his team injected AMH directly into mice to mimic the increased AMH levels found in pregnant women. These mice experienced symptoms of PCOS, including reduced fertility, fewer mature follicles, and GnRH neurons that released more LH. Mice injected with AMH passed PCOS symptoms onto their offspring as well, providing the perfect model for understanding how PCOS is inherited.

Although the fetuses were never directly exposed to circulating AMH in these pregnant mice, the placenta contains enzymes that break down AMH and convert testosterone into estrogen. These enzymes are inhibited by AMH, leaving the fetus exposed to high levels of testosterone and low levels of estrogen. When Giacobini’s team blocked receptors that recognized the hormones released by GnRH neurons, essentially silencing their hormonal messages, they partially reversed the PCOS-like symptoms in the mice given AMH.

Giacobini is still trying to understand how all these moving parts connect and lead to PCOS. He recently discovered that high levels of AMH or testosterone may lead to epigenetic changes common among some patients with PCOS, and that these changes are heritable (4).

“Now the million-dollar question is what is the culprit for PCOS? Is it AMH that directly impacts the epigenetic programming and induces PCOS-like traits in the progeny? Or is it testosterone, or the combination of both that predisposes [offspring to] these changes?” said Giacobini. “This is something we are currently working on to actually dissect the mechanism of each one of these factors.”

A new cancer vaccine

Although AMH may not cause ovarian cancer, targeting its receptor, AMHR2, is a good strategy for treating (and maybe preventing) epithelial ovarian carcinoma (EOC), the most common and lethal type of ovarian cancer (5-6). Less than half of women diagnosed with EOC survive. The symptoms — bloating, frequent urination, constipation — aren’t a dead giveaway that something is seriously wrong, so women often miss it before it’s too late. 

“It’s a disease that sneaks up on women; the median age is 63,” said Vincent Tuohy, an immunologist and cancer researcher from the Cleveland Clinic. “There’s really no good biomarker for early detection. It’s a disease to me, as an immunologist, that screams ‘prevent me.’”

Tuohy and his colleague Suparna Mazumder, another immunologist and cancer researcher from the Cleveland Clinic, set out to develop a preventative vaccine for EOC by looking for a protein that was specifically expressed in EOC cells. They focused on proteins expressed specifically in the ovary, but that nearly disappeared after menopause when EOC usually arises; Tuohy calls these “retired” proteins.

Vincent Tuohy, a cancer researcher from the Cleveland Clinic, develops antibodies against AMHR2 to treat ovarian cancer.

Credit: Vincent Tuohy

The extracellular domain of AMHR2 (AMHR2-ED), the point where AMH binds the receptor, seemed to be the perfect drug target because AMHR2-ED expression decreases as women age and reaches nearly undetectable levels after menopause. AMHR2-ED is also highly expressed in EOC. 

“Our goal was to vaccinate women against this extracellular domain, the component where AMH binds. If we vaccinated in postmenopausal women, we wouldn’t get any autoimmunity because the proteins are no longer there in normal cells. But we could provide immunity against any emerging tumor,” said Tuohy.

Tuohy and Mazumder showed that mice vaccinated against recombinant AMHR2-ED protein could inhibit the growth of EOC tumors introduced only seven days after vaccination (5). The vaccination produced an immune response so robust that transferring antibody-producing B cells from a vaccinated mouse to an unvaccinated mouse enabled the unvaccinated mouse to fight off EOC before it took hold.

“When we vaccinated the mice against the mouse ovarian carcinoma, the vaccination was mediated by IgG. This is very unusual because most vaccines are mediated by T cells,” said Tuohy. “We thought if this was mediated by IgG, we could make a monoclonal antibody that could be mass produced to treat [EOC].” 

Mazumder screened antibodies against AMHR2-ED and identified one that killed EOC cells. The antibody also stopped the growth of patient-derived EOC tumors implanted into mice. 

“It’s a wonderful antibody that inhibits the growth of ovarian tumors, and we want to bring that to clinical trials. But first we want to play with it,” said Tuohy. 

The researchers’ vaccine is already in development in collaboration with the National Cancer Institute PREVENT Cancer Preclinical Drug Development Program, which helps researchers move their cancer treatments to clinical trials.

“We’re trying to develop a 21^st century vaccine program that prevents adult-onset diseases like breast cancer and ovarian cancer,” said Tuohy.

A new fertility treatment

David Pepin and Patricia Donahoe, cancer researchers at Massachusetts General Hospital and co-founders of Oviva Therapeutics, originally planned to use recombinant AMH as an ovarian cancer therapeutic similar to Tuohy’s AMHR2-ED vaccine. When Pepin put primary ovarian tumors into immunocompromised mice and treated them with AMH, he was delighted to find that tumor growth slowed, but he started considering other effects that AMH treatment may have.

“David had trained in reproductive biology, so he had the smarts to look at the ovaries, and the ovaries were shrunken in these animals. Anybody else would have missed it,” said Donahoe. “[AMH] is usually made in developing follicles, but there were no developing follicles in these animals because of all of the excess [AMH] we had been giving them. David came up with the hypothesis that we were inhibiting development of these follicles.”

Normally when developing follicles are depleted, a woman is likely in menopause. However, in this case, the excess AMH slowed follicle development and egg release prior to menopause, preserving the ovarian reserve. 

Pepin confirmed his observation that excess AMH affected the number of developing follicles by giving mice recombinant AMH protein or using a gene therapy platform to deliver the AMH gene (7). Again, he found that mice overexpressing AMH had smaller ovaries and no developing follicles. The number of eggs was stagnant, however, which hinted to Pepin that this approach might lead to a form of birth control. While mice were treated with AMH, they couldn’t have pups. But as soon as Pepin stopped giving the mice AMH, they could become pregnant. When he transplanted the ovaries of mice treated with AMH into mice that never received treatment, the recipient mice could reproduce. 

“The contraceptive effect was so profound. That’s probably the most impressive part of this paper since this is a completely different method of contraception. All of the other methods act on later stages of [follicle development],” said Pepin. “We tried to think how we could use this unique mechanism to solve a technical problem, and one of them was damage from chemotherapy in young women.”

Women who undergo chemotherapy often suffer decreased or total loss of fertility in part due to damage done to developing oocytes. Researchers have tested types of contraception that inhibited follicle development at stages later than Pepin’s AMH therapy to preserve fertility, but the effects were “modest.” 

“Maybe we had another drug that could do ovarian suppression, but could do it at the right stage. If you could prevent the follicle from activating, it would never grow and be exposed to chemotherapy,” said Pepin.

Donahoe and Pepin next tested if AMH treatment could protect the ovarian reserves of mice undergoing chemotherapy. They treated mice with AMH for one month, then treated the mice with chemotherapy for about 12 days. Mice treated with AMH had a larger ovarian reserve than mice that were not treated with AMH before therapy.

Donahoe and Pepin are not only exploring how they can translate AMH therapy to women, but they are also investigating its use as a contraceptive for animals including cats and dogs in shelters as an alternative to surgery. Researchers at the Cincinnati Zoo are currently testing this method on feral cats and reporting  promising results.

Although AMH looks like a promising therapeutic for preserving the ovarian reserve in people undergoing chemotherapy, Beim cautioned that it’s not a good option for day-to-day birth control.

“It would be a very risky way to achieve birth control in women. Birth control pills block ovulation, but they do not block follicular development, so they’ll still have all the normal ovarian hormones for bone health or normal glucose metabolism,” said Beim. “It would be like removing the ovaries. It would basically be like putting women into menopause for contraception, which is overkill.”

Beim added, “It’s not that we think it’s dangerous for women to be on AMH chronically. We think it’s dangerous for women to be on it at a level that would be required to suppress the ovary enough that they would not be producing eggs or not be getting pregnant.” She further clarified that the dose needed for patients undergoing chemotherapy is low.

But there are some instances where a healthy woman may want to take AMH. Pepin and Donahoe showed that women pursuing in vitro fertilization (IVF) may turn to AMH in the future to improve their odds of success (8). 

One strategy to improve chances of becoming pregnant during IVF is to use drugs to induce superovulation, where a woman produces more than one mature egg per month. Pepin determined that pretreating mice with AMH strengthened the effect of  drugs used to induce superovulation; treated mice yielded double the number of eggs.

“One of the big problems is to have the follicles in the right stage of maturation to respond to stimulation,” said Pepin. “The wave of follicles is coordinated and synchronized [after AMH pretreatment], and so that makes it easier to get a better response to ovarian stimulation.”

Donahoe added, “It’s like all the horses are coming out of the starting gate at the same time.” 

A new therapeutic for a new therapeutic category

Although there are many potential applications for an AMH therapeutic, making the therapeutics themselves is a huge challenge. Celmatix is working to systematically navigate these obstacles and develop effective, easily produced, AMH-based therapeutics. 

One big obstacle, and a main reason why there isn’t yet an AMH therapeutic, is that it’s virtually impossible to isolate AMH from a natural source. “Other hormonal drugs that have really built the foundation of women’s health therapeutics — estradiol, progesterone, gonadotropin — the first generation of those drugs was purified from natural sources such as smashed up placentas,” said Beim. “There are still drugs that women take that are sourced from urine. It’s not even an antiquated thing.”

AMH is found at relatively low levels in the body, and is only active in the ovaries, so the amount of usable, active AMH isolated from urine, or any natural source, would be minimal. 

Paolo Giacobini’s team at the University of Lille uses animal models of PCOS to understand AMH’s role in the disease and its potential as a therapeutic for the disorder.

credit: paolo giacboni

Researchers and clinicians are turning more and more to recombinant protein technology to develop fertility hormones. Making recombinant female hormones currently used in women’s health is relatively easy. The molecules are small, naturally expressed at a high level, and don’t require activation. 

“The problem with AMH is that it has evolved to be expressed at very low levels, which then translates into recombinant systems as well. It doesn’t express well in recombinant systems,” said Beim. “What we saw as the clear therapeutic strategy to undertake at targeting AMHR2 was to really tackle the protein engineering problem.”

AMH is a huge protein, especially compared to other hormones. This not only makes producing the protein in a recombinant system more challenging, but it also makes it difficult to produce a small molecule drug that causes the same effect as the large protein. But there are benefits to using a large molecule drug.

“What’s nice about a large protein ligand is that it doesn’t get processed in the liver the way that a small molecule would. You don’t have the risk of the same kinds of off target effects. It’s not going to cross the blood brain barrier. Taking a protein engineering approach really just makes sense on multiple levels because of the biology of the target, but also because of the safety profile of what you’re trying to achieve therapeutically,” said Beim.

References

1. Moses, M.M. & Behringer, R.R. A gene regulatory network for Müllerian duct regression. Environ Epigenet  5, dvz017 (2019).
2. Cimino, I. et al. Novel role for anti-Müllerian hormone in the regulation of GnRH neuron excitability and hormone secretion. Nat Comms  7, 1055 (2016).
3. Tata, B. et al. Elevated prenatal anti-Müllerian hormone reprograms the fetus and induces polycystic ovary syndrome in adulthood. Nat Med  24, 835-846 (2018).
4. Minomouni, N. et al. Polycystic ovary syndrome is transmitted via a transgenerational epigenetic process. Cell Metab  33, P513-530 (2021).
5. Mazumder, S. et al. Primary Immunoprevention of Epithelial Ovarian Carcinoma by Vaccination against the Extracellular Domain of Anti-Müllerian Hormone Receptor II. Cancer Prev Res  10, 612-124 (2017).
6. Mazumder, S. et al. Immunotherapy of ovarian cancer with a monocolonal antibody specific for the extracellular domain of anti-Müllerian hormone receptor II. Oncotarget  11, 1894-1910 (2020).
7. Kano, M. et al. AMH/MIS as a contraceptive that protects the ovarian reserve during chemotherapy. PNAS  10, E1688-E1697 (2016).
8. Kano, M. et al. Neoadjuvant Treatment With Müllerian-Inhibiting Substance Synchronizes Follicles and Enhances Superovulation Yield. J of the Endocr Soc  3, 2123-2134 (2019).
9. Hart, K. et al. Structure of AMH bound to AMHR2 provides insight into a unique signaling pair in the TGF-β family. PNAS  118, e2104809 (2021).