false
zh-CN,zh-TW,en,fr,de,hi,ja,ko,pt,es
Catalog
Recent Advances in Gonadotropin Signaling and Ther ...
Recent Advances in Gonadotropin Signaling and Ther ...
Recent Advances in Gonadotropin Signaling and Therapy
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Good morning, everyone. I have to say there are more of you here than I was expecting on the last morning, and it's probably because I was lobbying you all to be here, so thank you for that. I also wanna thank the organizers for the opportunity to present our work, and hopefully all of us, the three speakers, will make it worth your while getting up early. Okay, so I'm gonna talk more about how gonadotropins are made, and in particular, the regulation of FSH, and let's see if I can manage this remote. Okay, I have no disclosures, sadly. Here's your QR code. I'm not seeing any phones, so no one wants the QR code. Okay, so I don't need to tell this audience about the endocrine control of reproduction. We've heard a lot at this meeting, everyone's seen the pointer, about cispeptins regulating GnRH, which then goes on to regulate the gonadotropins, which then regulate gonadal function, for example, sex steroids, which then feed back to regulate their own synthesis at the level of the brain and the pituitary. Our lab is interested in that part of the axis, but we also spend a lot of time talking about something that I don't think I've heard mentioned once at this meeting, which is strange, because 20 years ago, it was talked about a lot, and that's the role of the TGF-beta ligands, the inhibins and the activins, which, again, I probably don't need to tell this audience, but I'll just remind you, are selective regulators of FSH without regulating LH. So in response to FSH, in particular, the gonads will make the inhibins, and females inhibit A and B, and males just inhibit B, and then these proteins, members of the TGF-beta family, will feed back at the level of the gonadotrope cell and the pituitary to suppress FSH production. And the current dogma about how they do this is that they act as competitive receptor antagonists, blocking the local actions of activin B, which stimulate FSH production by regulating the transcription of the FSH-beta subunit. So this is just a schematic to remind you that activins and inhibins are structurally related dimeric proteins. Inhibins are heterodimers, I'm not really seeing the pointer, of the inhibin alpha subunit with one of two beta subunits, inhibin beta A to form inhibin A, or inhibin beta B to form inhibin B. And then interestingly, the dimerization, either HOMO or heterodimerization of the beta subunits is how we get the activins. And again, these molecules have opposite activities, inhibins suppressing FSH, and activins stimulating FSH. So what I wanna do this morning is tell you two stories. One was recently published, so I might go through that one a little bit more quickly. And then a second story, which is completely unpublished. So again, just to reiterate what I said, the current model is that within the pituitary, there's production of activin B by the gonadotrope itself, which then acts in either an autocrine or paracrine fashion to stimulate the expression of the FSH beta subunit, and then the production of the dimeric FSH hormone. So here we have activin B binding to complexes of transmembrane serine threonine kinase receptors, referred to as type II and type I receptors, where the ligand binds to a pair of type II receptors. You get a recruitment of a pair of type I receptors. Type II transphosphorylates type I, which activates the type I. And then type I phosphorylates substrates that initiate intracellular signaling that eventually culminates in the expression of that FSH beta subunit. In the case of inhibin, they're one of these rare examples of endogenous competitive receptor antagonists. Because inhibins have structural similarity to the activins, they share a beta subunit. They can also bind to the type II receptors, but they can only engage one at a time rather than a pair. And in the presence of a high affinity co-receptor, referred to as beta glycan, you get this high affinity ternary complex where inhibin essentially steals the type II receptor away from activin, and that's how you get a suppression of FSH. So it's really a passive process in that you're inhibiting the ability of activin to regulate FSH production. That's the model, and I'm gonna tell you that everything I told you here is not completely wrong, but not mostly wrong, which hopefully is provocative. So part one, we're gonna talk about inhibin action. Again, I'm having trouble with the pointer. So again, the model is that inhibin will bind to one type II receptor with high affinity in the presence of this beta glycan, and this is work from Wiley-Vail's lab. And this was all based on in vitro observation, so we went ahead and made a conditional knockout where we knocked out beta glycan specifically in the gonadotrope in mice. And the prediction was that if we were to get rid of that beta glycan, then inhibin wouldn't be able to out-compete activin, and so you would basically have an unfettered activin signaling, which should lead to, let's see if this arrow gets bigger, more FSH beta expression. That would lead then to an increase in FSH production, which would enable the pituitary to drive more ovarian follicle development. You get more follicles, more follicles will ovulate more oocytes, and you might enhance fertility. So that was the a priori prediction. And that's pretty much what happened. So we published this a few years ago. This is showing you large antral follicles in these knockout mice relative to controls, and you can see they have more antral follicles. These COCs are cumulus oocyte complexes, and this reflects the fact that these animals will now ovulate more eggs per cycle. And again, as predicted, they have enhanced fertility. The surprise, however, came when we measured FSH levels, which did not differ between the genotypes, and that we looked all across the estrus cycle. I'm just showing you a couple stages here, the morning of pro-estrus, and then at the time of the so-called primary and secondary FSH surges. And at both time points, when FSH was elevated, we didn't see a genotype difference. Now, to this day, we don't actually have an explanation of how we have an ovarian phenotype that looks like enhanced FSH, but we don't have an increase in FSH levels. Some of you will know that FSH is a glycoprotein, and that different glycoforms have different activity. Our chair here knows that probably better than anyone else in the room. And so we're speculating that there's probably a more active form of FSH that's released in this model, but as our chair also tells me, that's gonna be tough to measure in the mouse. We went on to ask the question, are these mice even still sensitive to inhibin? Because we knocked out this inhibin co-receptor, and it turns out that they are still sensitive to inhibin. And the way that we assess that is that we treated them with an anti-inhibin serum, and if we just focus on the black here, this is within subject's design. We took a blood sample, then we injected the animals with the anti-inhibin serum, and you can see that there's the expected increase in FSH in the controls. What was unexpected is that we also saw an equal increase in red here in the conditional knockout. So in the absence of this all-important co-receptor, in fact, we still had inhibin action. So then we started to look at this in more detail, and we took pituitaries out of the control in the knockout mice, and we asked about their sensitivity to both inhibin A and inhibin B. And this is where we got probably the most interesting insights. So if you just focus on the left here first, this is looking at expression of the FSH beta subunit in these pituitary cultures. As we increase concentrations of inhibin A from the left to the right, and in the black line in controls, you see the classic inhibin response, this dose-dependent suppression of FSH beta. And if we knock out beta-glycan, we pretty much blocked that response, which was the a priori prediction, because this is an important co-receptor for inhibin A. What's interesting, however, is when we do the same experiment with inhibin B, you can see that the response in the controls and the knockouts are the same. So what this told us is that beta-glycan was essential for inhibin A action, but not for inhibin B action. And so this led us to look for a new, or an alternative co-receptor for inhibin B. And as we recently reported at the end of last year, we discovered this protein called TGFBR3L, which is a mouthful. It stands for the TGF beta type three receptor-like. So we could even call it beta-glycan-like. I'm gonna call it R3L, just for simplicity. And the important thing to note here is that beta-glycan is this large transmembrane protein, multi-domains, and the juxtamembrane domain is something called the zona pellucida, or ZPC domain. And this is where inhibin A binds. And the R3L, you can see, is a much smaller protein, which also has a zona pellucida domain that has about 33% sequence identity with this ZPC domain of beta-glycan, which is why it's called TGFBR3-like, and why we suspected that this might be involved as an inhibin B co-receptor. So here are some data that we collected collaboratively with one of our audience members, Fredrik Ruff-Zimoski and Stuart Silfen in New York. And what this is showing is single-nucleus RNA sequencing from mouse on the top and human on the bottom. I think most people are now accustomed to seeing these TSNE plots, or maybe UMAP plots, which show you different, in this case, different cell types. In orange here are the gonadotropes in the two species. And if you can see the purple here, you can actually see that this R3L is only expressed in the gonadotropes in both mouse and in human. This is also true in rat, if you do similar analyses. And when we look across mouse tissues, and they're also published, or online data in human, it really looks like this is only expressed in the pituitary. And then within the pituitary, it's only expressed in the gonadotrope, which is the classic inhibiting target. Now, these are unpublished work. I just wanted to give you a sense. We're looking into what's restricting the expression of R3L in the gonadotrope. Maybe we'll focus on the right here first. We made SF1, which is a transcription factor that's specific to the gonadotrope. We made gonadotrope-specific SF1 knockouts. And this is showing you in females and males that when we knock out SF1, this R3L expression in the pituitary disappears. And again, in collaboration with the folks in New York, we did single-nucleus attack sequencing. This is the R3L locus. This is exon one. This is where the promoter is. You can see in controlled gonadotropes that the gene is open. And in the absence of SF1, you have a compacted locus, in particular the promoter, which likely explains the loss of expression of R3L. So just to show you some evidence about how R3L functions as an inhibin-B-specific co-receptor, we made stable cell lines. These are CHO cells that overexpress either nothing, R3L, or beta-glycan. We transfect in a SMAD responsive reporter. This is an Activin responsive reporter. We treat them with Activin, and we set the maximal response to 100%. And then we titrate in either inhibin-B or inhibin-A. And what you can see is that these cells with the absence of the co-receptors are insensitive to inhibin-B, but both beta-glycan and blue, and importantly R3L and red, confer sensitivity to inhibin-B that otherwise doesn't exist. In the case of inhibin-A, you can see that the cells actually have some inherent response to inhibin-A, probably because of low-level beta-glycan expression. When we overexpress beta-glycan, we shift the curve to the left, so we've made them more sensitive to inhibin-A, but you can see that R3L does not enhance the inhibin-A response. We also did binding experiments, and we see that R3L binds inhibin-B with high affinity, but not inhibin-A. So we went on to make knockout mice using CRISPR, makes life much easier. I think I've lost the pointer. No, here it's back. That's not my time up, is it? So this is just to show that we knocked out exon-2. There's still a low level of expression. This is work done in Cynthia Andoniado's lab in the UK. There's some message left, but this can't encode a functional protein. And just to go quickly through the phenotype of the animals, comparing wild type to knockout, we see, I probably shouldn't say this, but there's an increase in natural follicle number. It's not statistically significant, so I guess that means it didn't happen. And then there's an increase in number of corporaludia, indicating more ovulations. In fact, there are more eggs ovulated, more implantation sites in the uterus. And just like the beta-glycan conditional knockouts, these mice have enhanced fertility. We looked at FSH again. In this case, you can see a trend in this cohort of mice that there's an increase in serum FSH. I'll show you data in a subsequent cohort. There is a significant but modest increase in FSH, which is driven by a significant but modest increase in FSH beta. Again, in this cohort, not statistically significant. And this is just showing you by qPCR that that R3L message is quite low. And no effect on SF1, which I just told you is important for regulation of R3L. Interestingly, these animals also had some sustained inhibin sensitivity. So same setup that I described before, but the data are presented somewhat differently with wild type on the left and knockout on the right. So it's a within-subjects design where we take a blood sample before and after treating with an anti-inhibin serum. And again, in the wild type, you see the predicted increase in FSH levels, which we also see in the conditional knockout. So at this point, we suspected that beta-glycan was probably compensating when we knocked out this R3L. And that proved to be the case. So this is a little bit complicated, but we're comparing three genotypes here. Heterozygotes for the R3L knockout. Knockouts for R3L. And here is where you're seeing now a statistically significant increase in FSH. And FSH content, sorry, this is in serum. This is in the pituitary. And this is the FSH beta message. But when we knock out R3L globally and beta-glycan specifically in the gonadotrope, that's when we see the highest levels of FSH in all of these measures. Serum, pituitary content, FSH beta. And as a result of these high levels of FSH, you see this massive stimulation of the ovary. The ovary is about four times larger in the double knockout than in the single knockout. That's quantified here. You can see that they ovulate a lot more eggs in natural cycles than do these double knockouts, sorry, the single knockouts, which already have enhanced ovulation. You can see one animal ovulated 80 eggs. You can see there's an increase in the number of antral follicles, the number of corporal lutea. But these animals do not, the females do not produce live young. We haven't completely characterized the fertility defect, but it turns out that they do get pregnant, there's implantation, and around day 10 and a half, they start to, the pregnancy starts to fail. And we're trying to figure out what the cause is. Okay, so finally we went back to the pituitary culture system. I showed you this before, where when you knock out beta-glycan, there's still sensitivity to inhibit B, but not A. And now we looked at the pituitaries of the double knockouts, and here we lose sensitivity to both inhibit A and inhibit B. So we think that the combination of these two receptors is required for inhibit B action, but with R3L being a bona fide specific inhibit B co-receptor. So that's what's indicated in this first summary here. And what I showed you is that disruption of inhibit B action through R3L modestly, but physiologically significantly increases FSH levels that enhances fertility. If we block all inhibit action by knocking out both co-receptors, we get very high FSH levels. Over-stimulation of the ovary and infertility, but again, they do get pregnant, but there's pregnancy failure. Okay, so let's move to part two, and that took me longer than I thought, so I'll have to pick it up. So for many years, we and others have been looking at how activins regulate the transcription of the FSH beta subunit. Worked for many labs, including our own, have worked out this model where activin B binds to one of two activin type two receptors called 2A and 2B. You get recruitment of one of two type one receptors called the activin type 1B and C receptors. Sorry, I know this is gonna start to be alphabet soup. The type one receptors phosphorylate SMAD3, which partners with SMAD4. They move into the nucleus, and they can bind either directly to the DNA, the FSH beta promoter on their own, or importantly, in partnership with FOXL2, which many people think is only in the ovary. It's not. And so our mode of doing things is to work out things in cell culture, and then go and start to knock everything out. And so here's some papers here where we showed that knockouts of SMAD3, SMAD4, and FOXL2 lead to isolated FSH deficiency. So that part of the model holds up. Gautier-Xiong, a former PhD student in the lab, then went to look at the type two receptors. He found that the type 2A was most important, but if he knocked out both receptors, you can see here in females that they basically don't make FSH. They're infertile, and if we look at their ovaries, you can see follicles that grow up to the early antral stage, and not beyond. Because, as I like to say, they're waiting for an FSH signal that never comes. So we were just doing our due diligence, and we just kind of wanted to, sorry, finish all of this up, and say, okay, which one of those type 1 receptors is important? Is that type 1B or type 1C? So there are seven type 1 receptors in the TGF beta family. I'm not gonna go through all the names. For convenience, we like to call them the ALCs, or active end receptor-like kinases. For active end B, which is our candidate ligand, the two type 1 receptors that have been shown to mediate its activity are ALK4 and ALK7, or ACVR1B and C. We knew we could eliminate many members of this list based on who's expressed in the gonadotrope. We also had made some previous knockouts, and there were really only two type 1 receptors that were possible as mediators of the signaling, one of which was ALK4, the classic active end type 1 receptor, and then there was also the TGF beta type 1 receptor called ALK5. We weren't really that interested in this because we had shown before that these cells, the gonadotropes don't express the type 2 receptor for TGF beta, so we figured they were just TGF beta insensitive. So our surprise came, however, when we knocked out this ALK4, and we saw no effect on FSH levels in males or in females. That was a real shocker. And even more surprisingly, when we knocked out ALK5, which I told you initially we weren't interested in, that's when we saw a suppression of FSH in both sexes. This was really unexpected. Luisina Ungaro, a research associate in the lab who did this work, then went on to make double knockouts, and that's really where we saw pretty much the absence of FSH in fertility in the females. You can see the thread-like uteri's, small ovaries, small testes in the males. These are essentially FSH deficient animals. So we went back to this question, well, how could it be active in B? Because active in B, it uses those type 2 receptors I described, but it does not bind to ALK5, and it does not signal through ALK5. So we revisited some old data from Marty Matsuk where they had knocked out the beta B subunit. These are mice that don't make active in B, and again, if active in B is important, FSH levels should be low in these animals, but in fact, it's exactly the opposite. So if you look at females here, you can see the FSH levels are elevated when the beta B subunit is knocked out, and that's as a result of an increase in the expression of the FSH beta subunit. We see the same thing, but to a lesser extent in males, but males have higher FSH to begin with. We went on to make floxed inhibin beta B mice so we could knock out beta B specifically in gonadotropes to test this autocrine-paracrine model, and you can see that there's no effect on FSH production. In fact, the difference that we see here disappears here, and so actually, we think the increases in FSH that we see in the global knockout reflect a loss of inhibin B negative feedback. So okay, we know a TGF beta ligand has to be involved. It doesn't look like it's active in B, and so we went back and looked at the family and asked, well, who else might be having this action on FSH, and there were some criteria that whatever the ligand was would have to satisfy, would have to be antagonized by inhibins, which we've talked about, follystatins, which are proteins that classically bind and bio-neutralize the activins, but they can bind and bio-neutralize some other members of the family as well. I showed you data that these two type II receptors are essential, so whatever member of the family is important has to use these type II receptors, and I just showed you that the member of the family that's important has to use this ALK5, and to a lesser extent, ALK4, and really, when you look at the whole family, there are only two ligands that satisfy all these criteria. They're known as growth differentiation factor 11 and myostatin, or GDF8. Now, we didn't think about myostatin at first because it's really muscle-specific. That's where people think about myostatin being. So we focused on GDF11 because from the sequencing data, we actually saw some GDF11 expression in gonadotropes. So we got Phlox GDF11 mice. I think you get a sense we knock a lot of stuff out. So we knocked out GDF11 in the gonadotrope, and in males, there was no effect on FSH. Similarly, no effect in females. So then we thought maybe there's compensation by activin B in the absence of GDF11. So we made double knockouts, and again here, no effect on FSH in males or in females. We knocked out furin, which is a protease that's important for activation of many TGF-beta ligands. No effect on FSH. So then we started to think, well, maybe the ligand isn't even coming from the gonadotrope. So what we did is we got some antibodies, bioneutralizing antibodies from, sorry, George, does that mean I have two minutes until five minutes left, or does that mean I have two minutes until 30 minutes? Two minutes until five minutes. Okay, thank you. So Acceleron had made some antibodies that bioneutralize myostatin in GDF11. These molecules are about 90% similar in sequence, so it's hard to get antibodies that discriminate between them. But they do have one that inhibits myostatin specifically. We treated, in this case, male mice with these bioneutralizing antibodies. And you can see that within two days, we really had a robust suppression of FSH in serum when we treated with both of these antibodies. If you look at circulating levels of GDF8, which is myostatin here on the left, versus GDF11, please note the break in the axis. You can see that there are very high levels of GDF8 in circulation and very low levels of GDF11. And also based on the antibody results, it really looks like myostatin may be the main driver of FSH production, at least in the mouse. So Luisina went on to do some additional experiments where she did a dose response. So these are single injections with the antibodies. You can see that at all three of these doses, she sort of maxes out this response. So about 75% reduction. So there's probably something else that's contributing. But most of the stimulation of FSH seems to be by one or both of these ligands, myostatin and GDF11. She followed the mice out for four weeks. You can see it's a reversible response in a dose-dependent fashion. So at the lowest dose, that's when FSH seems to come back fastest. So we then went on to look at the myostatin global knockout mice. These are some samples and then mice that we got from Seijin Li. And you can see that in the case of males, global knockouts, you can see significantly lower levels of FSH in the knockouts relative to wild type. This translates into a decrease in the size of their testis, which is similar to what you see in the FSH beta knockouts. Seijin also provided us with plasma from muscle-specific myostatin knockouts. He also used a Cree driver that was specific to skeletal muscle. Sorry, skeletal muscle. And you can see here, too, there was a significant decrease in FSH levels, although not to the same extent as in the global knockout. So that might reflect that not all myostatin was knocked out or that there is myostatin from extra muscle sources. When they measure myostatin levels in circulation, you can see there still is some myostatin left, and that may help explain why FSH is not suppressed to the same extent here as here. Finally, we treated females with the antibodies as well. So either with IgG and Activin A, bioneutralizing antibody, again, the myostatin GDF11 antibody or just the myostatin antibody. You can see these two antibodies, but not the Activin A antibodies suppress FSH. And this then translates into a decrease in the number of follicles that are ovulated that cycle. So if we revisit our HPG axis and the role of activins and inhibins therein, we think we've pretty much conclusively shown that Activin B is not an important regulator of FSH in vivo, at least in the mouse, and that a main driver, I won't say the main driver, but a main driver seems to be myostatin coming from the muscle, which I don't think any of us would have anticipated. We certainly didn't. So if we go back to the overall summaries, we now see that endocrine myostatin rather than autocrine paracrine Activin B seems to be a major regulator of FSH production, again, in mice. And we think this establishes a novel endocrine axis between the muscle and the pituitary, at least we don't know of other examples. I showed you in the first part that this R3L is an inhibin B-specific co-receptor. And we're particularly interested in this given where it's selectively expressed and that it only seems to mediate the actions of inhibin B. We think it might be an interesting drug target to block inhibin B binding to this co-receptor. Anticipate would lead to increases in FSH which may have pro-fertility effects. So with that, I tried to mention people as I went through. Emily Boulay, a PhD student, led our R3L project. Luisina Ongaro, a research associate, led the myostatin project. We've had lots of fruitful collaborations. I mentioned Cynthia and Frederique who I think both are in the audience. And our funding sources are mainly CIHR and Canada. So thank you for your attention. I think I've left only a couple minutes for questions. Thank you, Dan, this is open for questions. Hi, Dan, great to meet you. Nick Webster, San Diego. So, I mean, this is really provocative, obviously. I was hoping it would be. Obviously, there's a lot of interest in sort of the anti-myostatins and sarcopenia and other things. Now, this raises important issues. So sort of looking at it the other way, you've been looking at the reproduction, but in any of your knockout models and things, did you look at muscle effects? We never did. So you're asking if maybe FSH would- Yeah, going the other way. Or is it through a different receptor in the muscle than in the- Yeah, so the simple answer is we haven't looked at it. We have been looking at extragonadal effects of FSH based on Mona Zaidi's work. So our focus has been mainly on the tissues that he's identified as potential off-targets of FSH, the bone and the fat. So I just don't know yet. Yeah, thanks. All right, Pat, you get to give me, tell me how old I look. Oh, you know how old you are. Yeah. Pat Chappelle, Horry State. That was amazing, really interesting. I guess my question is, I don't know if you've gone this far, but obviously there are metabolic deficiencies, right, that cause all sorts of muscle issues. Is it known how myostatin looks in those that then also are associated with deficiencies in reproduction? Right, so a great, great question. Again, don't have a good answer for you. There are models, as you said, especially in mouse models of muscle wasting, which is where a lot of people have interest in myostatin. I think we'd like to get our hands on those mice or at least on serum from those mice to see if there are effects on FSH. We would predict that there will be. I kept emphasizing that the results we're seeing here, we're only seeing in mice, and I do want to emphasize that myostatin is very high in circulation in the mouse. That's not true across species. In fact, in the rat, it's GDF11 that's high in circulation. So our hypothesis is that when we do some of these studies in the rat, we're gonna see that GDF11 is important. In human, so in mouse, myostatin really is the key regulator of muscle mass, essentially. In human, it's myostatin and activin A. So actually, we're also open to the fact that maybe humans, sorry, a peripheral activin may be important in regulating. So I think we might have to look at more than just myostatin. In the mouse, that'll be the focus, but in other species, we might have to look at other ligands. But I think the important concept I want to get across here is that we have to stop thinking about autocrine, paracrine regulation and open our minds up to the fact that maybe an endocrine TGF beta ligand may be a main driver of FSH production. Thanks. Okay, I think we've run out of time. I'm sorry. To move on. Thank you, Dan. Sure, thanks. Okay, our next speaker is Dr. Livio Casarini, who's Associate Professor of Molecular and Translational Endocrinology in the Department of Biomedical, Metabolic and Neurosciences at the University of Modena in Modena, Italy. Okay. Thank you for the introduction and thanks to the organizing committee for having invited me. Well, I work in an endocrinology unit, in a basic endocrinology unit, where I study the cell response to the gonadotropins. Therefore, the story that I will tell you today takes place during the human menstrual cycle that I'm sure you very well know. It is represented here and especially in this slide, you can see the recruitment of a pool of pre-antral follicles that becomes antral during the last stage of the folliculogenesis before the volation. This stage is the so-called follicular stage. It is an FSH-dependent stage and FSH, in this case, induce proliferative signal to support the growth and the selection of one single dominant follicles until the ovulation. The story is completed by other actors, such as the estradiol that is produced under the stimulus induced by FSH. And what happened later is that LH supports the production of progesterone to allow the pregnancy. But this is another story. Let's focus the attention on the FSH-dependent stage. Looking at this picture, I was always interested in understanding why and how is selected only one single dominant follicles in the human species. So I think this is a very interesting topic and by reading in the literature, it is well described as a process that is induced by the declining level of FSH during the follicular stage. In this view, only the single follicle that is able to resist with the lower amount of FSH becomes the dominant. This is the concept provided by the literature and it is a very common concept. For instance, you can read the same concept also on Wikipedia. It is not important if the words are so tiny and you can't read it. It's just a concept that is important. But if you want to be more technical, you can open a textbook of endocrinology where it is described that the FSH is required for the antrum formation and the prevulatory follicle growth. Of course, the estrogens are required to support the dominant follicles, but what happened later is that the atresia that is a pro-apoptotic process characterizing the follicles that are not dominant is induced by the declining levels of FSH. What is very interesting in reading in other pages of the same book is that it's written that more or less the role of sex hormones in stimulating the prevulatory follicles should be similar in rats and in humans. Yes, you have understood well. Most of our knowledge in the dominance comes from rodent models, but rodents are not humans. First of all, they are, for instance, a multi-ovulatory species, so I may think that the process for selecting the dominant follicles is not the same than in humans. So I was stimulated to rethink to this issue on how the dominant follicles may be selected. So I have decided to not focus the attention on the FSH level rate, on the FSH receptor and the later receptor level during this stage, and with this view we may divide the follicular stage in three different steps. The first one is the secondary to early antral transition where the pool of follicles is recruited. At this stage, the number of FSH receptor expressed in those follicles is very low. What happened later is that FSH receptor level increase and increasing the selection of the dominant follicle comes together. Later, the FSH receptor in the dominant follicles is replaced by the LH receptor, which is a different receptor. So in my mind, it means that the FSH receptor may have a pro-apoptotic potential that is covered in vivo by other hormones or growth factors, but with some specific in vitro experiment, we may uncover this pro-apoptotic potential. And this is the sense of our first experiment performed about seven or eight years ago in which we have transfected granulosa cell lines with plasmid coding for the FSH receptor or the LH receptor. What happened in this case is that under an antibiotic selection, we may see that both the FSH receptor and LH receptor stable transfecta start to grow after one month. However, the FSH receptor expressing cells start to die. It seems that the FSH receptor cannot be maintained in the cells for a long time. It is not for the LH receptor transfecta cells. We have still in culture after several years those cells, so the LH receptor seems to not have the same properties of the FSH receptor. But if we want to deepen the analysis of those cells and exactly in the period in which FSH receptor transfecta start to die, we may see that they are characterized by very high cyclic NP levels, intracellular levels. And at the same time, we have evaluated the activation of caspase-3 as well as the progesterone production. So in these cells, the FSH receptor induced at the basal level the production of progesterone and at the same time, it killed the cells via caspase-3 activation. These are two different kind of signals that comes together in the cells. This concept is not relatively new. It was discovered thanks to a group that characterized a partial compartmentalization of the cyclic NP signaling in the cells by microscopy techniques. But briefly, FSH receptor that is coupled to the G alpha S protein induced the cyclic NP and this second messenger should be the responsible of the activation of steroidogenesis, which is important to produce progesterone and estradiol. But at the same time, through another pathway, the caspase-3 is activated and killed the cells. Well, taken together, these data make me think that FSH receptor may have a dual role. The first one is accepted during the secondary to early antral transition when it is expressed at the very low levels. What happened in this condition is that the receptor activates preferentially on a survival signal or anti-apoptotic signal thanks to the activation of molecules such as ERK1-2, AKT. The explanation of this is in that graph. That graph showed the coupling affinity of the receptor and several actors such as G alpha S, GQ, GE coupling. When the receptor is expressed at the very low levels, there is not a preferential coupling with the G alpha S protein. It means that there are several proteins that compete against each other to couple the receptor and the resulting signaling is the preferentially mediated by GQ or GE. That is an inhibitory signaling and therefore proliferating. What happened under FSH receptor overexpression is that the preferential coupling of the receptor is with the G alpha S protein and it induce downstream the cyclic MP increase, the steroidogenesis, and also the apoptosis. We have strengthened the results by incubating some cells with a cyclic MP analog that cannot be metabolized by the cells. What happened is that the cyclic MP analog over time decreased the cell viability. But we may rescue the cells by incubating them with estradiol. So on one side in the ovary is important to activate the cyclic MP to produce the estradiol and support the dominant oocyte growth. But at the same time, the cyclic MP induced maybe the death of the non-dominant follicles. And this is exactly the sense of this slide. We have the FSH acting through its receptor expressed at high level in non-dominant follicles which may produce the estradiol to support the dominant oocyte growth. But the question could be, why exactly that follicle? Which kind of characteristics should have the dominant follicles to not be killed like it would be an athletic follicles? It should have something different. To answer to this question, we have to introduce a new actor in our presentation. It is the so-called G-protein estrogen receptor. The acronym is GPR. I have met this guy times ago in a meeting where a nice speaker explained that this receptor is expressed in the ovary as well as in other tissues. It is coupled to an inhibitory complex which displayed some proteins such as MAGUC-ACP5. This inhibitory complex avoid that the GPR activate the G-alpha-S protein. And therefore, the signaling of this receptor is mainly proliferative. I mean, this proliferation is activated through calcium increase, ERK1-2 signaling, AKT signaling, it also depend. It is also a tissue-specific signaling, of course, but it is not a cyclic NP-dependent signaling. In our mind, what could happen in the ovary is that in a certain way, GPR may interact with the FSH receptor and modulate its signaling. I was very excited when one of my colleague at the same university of me, she's named Francesca Fanelli, is a structural modelist, have predicted that FSH receptor and GPR has a certain grade of affinity and they may form likely heterodimers. Especially those heterodimers seems to be a lower thanks to the contact between the seventh and the sixth transmembrane stretches of one receptor with the sixth and the seventh of the other one. Of course, we know that the class A GPCR are not so prone to form heterodimers and it is a very debated in the literature issues, so we have decided to strengthen the data, adding more experiment. In this experiment, we have evaluated the colocalization of the two receptor in a transfectal cell line. In the other one, we have performed breadth experiment in which we have produced a very nice binding core, I may say, to certify the FSH receptor and GPR interaction. And finally, and thanks to Eileen Hanieloglu to have provided this nice picture, we have demonstrated also by the PALM, super-resolution microscopy, the presence of associating receptors, GPR and FSH receptor in the cell surface. Okay, at this point, we were sure to have demonstrated that heterodimerization, but one of our reviewers told us it is not enough for method to Merlot. You need one more method. You have to add one more analysis named proximity ligation assay. At the time, I didn't know about this method. It is based on two antibodies that target two specific proteins. These two antibodies are labeled with two DNA fragment and by adding something like PCR mix, you may see a reaction that produce dotted points on the cells if the two targets are in close proximity. And this is exactly what we have seen now. As you can see, the arrow indicates the dotted spot. Okay, at this point, we were more or less sure that these two receptor may for heteromers. And what we have thought is that the inhibitory complex related to the GPR upon interaction with the FSH receptor may in some way block the G-alpha-S protein associated to the FSH receptor. This is a breath experiment revealing indeed the decrease of the signal produced by the interaction between the G-alpha-S protein and the FSH receptor, it is the blue lines, that decrease together with the increase of the expression level of GPR in the same cell. And it is not with a control receptor. So our theory seems to be true. Our theory seems to be true, but even more interestingly, we have thought that as well as GPR, also FSH receptor in this condition may activate other signaling pathways that are different than that mediated by the cyclic AMP. For instance, the AKT pathway that could be mediated by the beta-gamma proteins unit. This is exactly what we have demonstrated. In this picture, you can see that the FSH receptor-expressing cells, when they are treated by FSH, respond by increasing the cyclic AMP level, both in the presence or in the absence of estrogens. But when you co-transfect the cells with plasmid coding, both the receptors, the FSH alone is not able to induce the cyclic AMP increase. It means that the GPER blocks the FSH-dependent cyclic AMP production. Even more interestingly, as you can see in this western blotting, oh, sorry. I would like to activate the pointer. Anyway, it's not important. You can see in the AKT bands in the lane number four, which is the molecule phosphorylated upon FSH treatment of cells co-expressing the FSH receptor and the GPER. And it is not for the cell-expressing, only the FSH receptor. But please note that this is the beta-gamma-dependent phosphorylation of AKT because it is depleted by using the specific inhibitor named gallein. What happened downstream is that the AKT phosphorylation is related to an increase of cell viability. And in our mind, it means that the FSH receptor, upon dimerization with the GPER, may deviate its cell signaling from a pro-apoptotic signaling towards a proliferative signaling. This is exactly what happened in the dominant follicles that compared to the athletic follicles, which are characterized by a high FSH receptor expression and cyclic P activation, they die. But this is not in the dominant one, where the dimers of these two receptors may deviate the FSH signaling towards proliferative signaling. At this point, I could entertain you with a lot of boring experiments that are controls, but I would prefer to simply provide you the main concept of this experiment. As you can see here, we have produced a mutant GPER. Again, thanks to a bioinformatic prediction, we have detected the amino acid that on the GPER contact specific amino acid at the FSH receptor stretches. We have muted those amino acid into alanine. And in this way, we have produced a GPER unable to form heterodimers with the FSH receptor. Of course, we have performed also a functional characterization. Under this condition, when the cell co-express these two receptors, FSH perfectly activate the cyclic AMP, and the cell die. Another control experiment were provided by our colleagues in Poland. The professor Adolfo Ebermuller was the director of this experiment. He has produced a knockout cell model in order that the inhibitory complex linked to the GPER was disrupted. Even in this condition, the FSH receptor and GPER may form heterodimers, but the G-alpha S-proteins related to the FSH receptor was free to activate the denilicyclase, and therefore the cyclic AMP upon FSH treatment. This experiment were performed in cell lines, but we have also experimented in human primary granulosa cells, in which we have inhibited with small interfering RNA probes the production of GPER, and in this case, what happened is that the FSH potentiates the activation of cyclic AMP, the steroidogenic potential, but also kill the cells more easily. Well, this are in vitro experiment, but in the next slide, that is the last one of my presentation, I will show you some data that should provide a more physiological weight of this experiment. First of all, we have characterized the presence of FSH receptor and GPER in the ovary. As you can see, the immunostaining and western blotting, but most importantly, we have characterized the gene expression level of these two receptor in human primary granulosa cells obtained from women undergoing cytoretrieval for assisted reproduction. We have divided in two group those women. The first group, represented by the green line, is the group of the so-called normal responder women, that are women that, upon the FSH treatment in vivo, respond by producing a huge amount, a relatively good amount of oocyte that we have identified as a number of 10 or more oocyte. The sub-responder women are women that produce a low amount of oocyte in response to the FSH. In this case, they were four or less oocyte. So, in our mind, in the sub-responder women, what happened is that the GPER cannot counterbalance the pro-apoptotic signaling mediated by FSH receptor, and what happened in that context is that the FSH, in a certain way, kill some of the follicles or otherwise, in any case, the yield, the follicle yield, is not optimal. By plotting the FSH receptor gene expression level against, in y-axis, the GPER gene expression level, as you can see, there is a very positive correlation between the FSH receptor and GPER levels in the normal responder, but this is not in the sub-responder, where the increase of FSH receptor gene expression does not correspond to an increase in the GPER expression levels, and for us, it means that the theory is true. Enough level of GPER is required to counterbalance the pro-apoptotic signaling mediated by FSH. This experiment match our other data that we have investigated, such as that one, which express in x-axis the ratio between the estradiol and the FSH as a measure of the ability of estradiol of counterbalance the pro-apoptotic of FSH. Well, this data is plotted against the total or site number, and again, the data is the same. In poor responder, there is not a correlation. To conclude, I think that we are near to demonstrate that we may provide a new view of the dominant selection. In this case, FSH receptor should kill the athletic follicles via a cyclic NP increase, but it is a signaling that is required to produce the estradiol to feed the oocyte group. At the same time, what happened in the dominant follicle is that it is rescued by the heteromer formation between GPER and FSH receptor. Before to conclude, I would like to thanks our collaborators, especially Professor Haile Langellaoglu of the Imperial College in London, Adolfo Rivero-Muller from Poland, Francesca Fanelli, and especially Eric Reiter and Francesco De Pasquale from INRE in Tur. Again, also my group, the director is Professor Manuela Simoni, an endocrinologist, but also the two girls that provided a large part of this experiment, Elia Paradiso and Clara Lazzaretti. And thank you for your attention. Thank you. For questions, please come to the microphone, identify yourself, and ask your question. Hi, Angus McNichol, UAMS. That was a really nice talk. I was curious, you showed us FSH receptor levels during the estrous cycle. What does the GPER levels look like? And I guess what I'm getting at is, do you think it's an availability of GPER to then form this inhibitory heteromer, or do you think the heterodimer is regulated? Well, first of all, the GPER level during the menstrual cycle are not so known. There is only a few of paper describing with some immunostaining that the GPER expression level should increase during the follicular stage, but I'm not pretty sure it is exactly the same. We have to investigate this. Of course, it is difficult to prove the existence of heteromers in vivo, and this is the reason why we have to use some mechanistic experiment with cell lines to demonstrate this. But I think that it may be true in vivo also. Steve Franks from Imperial College London. Lido, thank you for that talk, and that sort of shattered my ideas of physiology of follicle development. But what you're suggesting is that that GPER-FSH receptor complex occurs just in the dominant follicle, but not in the subsidiary follicles. But really, to establish that, you'd need to be able to look at antero-follicles from human. Good question. This is what we are trying to do. We are trying to, in a collateral project, we are trying to obtain follicles that are, after the EVF cycle, remain little compared to the other one. And in our mind, they could summarize anarthritic follicles in nature. I'm not pretty sure that it is true. It is an approximation, of course. But we have to evaluate better in that context if exactly it happens that the athletic follicles have a lower amount of GPER. I am pretty sure, because the data are anyway precise. But let's see. We have not the data now. That is a good thing. Just a quick other question. Can you remind me, does GPER form complexes with the LH receptor? Yeah. It does, okay. It does. It is another project. And I may anticipate you that in that case, it does not inhibit the cyclic NP, but modulates other pathways such as the AP1 pathway. Then that may be important in polycystic ovary syndrome. Yeah, yeah. Thank you. Janet Hall, NIH. I wanted to ask where the source of the granulosa cells, I mean, the source of the cells from the women, were they post-HCG or before? And do you know if that makes a difference? Yeah, good question. This is a limitation of the study, indeed. Because we have obtained the granulosa cells from women after the ACG stimulation. So they are such as luteinized cells. Yeah, yeah. It means that have about 100 fold higher level of LH receptor expressed in the cell surface compared to the FSH receptor. This is not the best model, I know. You can do natural cycle aspiration. So it is possible. I mean, it's, you know, you have to set up for it. But it is possible to do it. And I'm just wondering, do you think that that could have had effect on anything that you found? Do you think it's totally independent of LH receptor activation? Yeah, the problem is that in the clinics that provide our cell not easily can obtain other kind of material. You don't get as many cells, which is hard. What, please? I have not understood. Yeah, you don't get as many cells, which is hard. Yeah. Hello, Rodolfo Rey, Buenos Aires Children's Hospital in Argentina. I have two questions. When all the results you showed us were devoid of estrogen exposure, weren't it? Yeah. No estrogens? No, no, no, no. We have no estrogens in the cell culture medium. Okay, so you hypothesize that this is an estrogen independent effect? Yeah, indeed. Because the GPER works even without the estrogen. Works even without the estrogen. This is an interesting issue because certain authors describe the GPER as, again, as an orphan receptor because not in all the cell model it respond to the estrogen stimulation. And I'm not pretty sure that it is the same in the ovary, but I think my experience that the most potent effects in the ovary is due, the estrogen dependent effect, is due to the nuclear receptor more than the GPER. And the second question is, do you think there is a place for any topographical issue, given that my understanding, which can be wrong, is that the FSH receptor is mainly placed on the outer cell membrane, where the GPER is mainly placed in the endoplasmic reticulum. Yeah, indeed. Exactly, the GPER is located mainly in the endoplasmic reticulum. But we have other data, not published in the paper shown today because they are very new, demonstrated that the GPER may retain in the intracellular level, also other GPCR that form a complex with that receptor. Okay, time for one last quick question. Thanks, George. Hi, Joni Jorgensen, University of Wisconsin in Madison. I'm just wondering if you can postulate if there is some sort of communication between the non-dominant follicles that tell this particular follicle to be the dominant. Is there some sort of signal that they might be sending that allows them to allow that heterodimer to form? I'm not understood very well, can you repeat? Sure, so the follicles that are recruited that don't make it, do they send some kind of signal to select that one dominant follicle? Yeah, good point. I don't know which kind of signal pre-programmed at follicles to become dominant. This is what we are trying to understand as a future experiment, but this is a good point. I think that the dominant follicle among the pool is pre-programmed to become the dominant one. I don't know, I don't know how, but maybe there is some epigenetic process, I don't know. Okay, thank you, Dr. Casserini. It's time to move on. So our last speaker today is Dr. Kim Jonas, who's a senior lecturer in reproductive physiology in the Department of Women and Children's Health at King's College, London. The title of her talk is The Luteinizing Hormone Receptor Knockout Mouse as a Tool to Probe the In Vivo Actions of Gonadotropic Hormones Slash Receptors in Females. Thanks, George, and thanks to the organizers for inviting me. I'm sorry for such a mouthful of a title for the talk, but I'm gonna talk to you today about a study that we recently published where we've, I suppose, repurposed the LH receptor knockout mouse model and a few other knockout, a few other kind of transgenic models that we've used along the way to try and understand aspects of LH receptor physiology and the wider kind of gonadotropin hormone receptor physiology, and ask some questions about how they work. Okay, so I have no financial disclosures, and here's the QR code should anyone want to ask questions via that method. Okay, so we've had quite a lot about the gonadotropin hormones and receptors, so I've literally just put a couple of bullet points in terms of fundamentals. And so they're glycoprotein hormones, primarily follicle-stimulated hormone, luteinizing hormone, and the omission here is chorionic gonadotropin, which I'm also going to talk about. They mediate their functions via the cognate class A GPCRs through FSH receptor for FSH and for HCG and LH through the LH receptor. They're primarily, when we think about kind of dogmatic views, thought to act via the GS signaling pathway, but we know that this is quite a simplistic overview of how they work and that they can couple to additional G proteins and also G protein-independent pathways to signal, and they're essential for reproduction. And we know this from a plethora of research that has happened, that has been reported, with functions elucidated via genetically modified mouse models that kind of sprung up over the late 90s and early noughties and human mutations, naturally occurring mutations, that have been discovered. Okay, so I'm going to, as the title of the talk suggests, I'm going to talk to you a little bit about the LH receptor knockout mouse model or the LERCO mouse model as it is colloquially known. So on the left here, we have the histology of the ovary, and we can see within the LH receptor knockout mouse, we've got arrested follicles, at the large antral phase and follicles of all other flavors kind of leading up to that. And here we have the wild type as a comparison. When you look at the gross morphology of dissected reproductive tract, we can see hyperplastic uteri and small ovaries in comparison to the wild type on the right. Now, when we drill down and we try and kind of overcome that ovulation block within the LH receptor knockout mouse and stimulate with prime with PMSG and mimic ovulation by giving HCG, we can see in the wild type that we have multiple corpus uterium that form showing ovulation occurs, but the LH receptor knockout mouse is resistant to that. So we have an ovulation block within the LH receptor mouse model, perhaps unsurprisingly, when you look at serum LH and FSH within the LH receptor knockout, we get elevation in LH and FSH to try and drive the ovary to produce steroid hormones and kind of as, I suppose, a symptom of that lack of steroid hormone feedback to the pituitary. So here's an overview of the LH receptor knockout mouse. Obviously, it's been kind of around for the last 20 years, so I'm sure I'm preaching to the converted in terms of what the phenotype of that mouse model actually is. But there have been a number of kind of questions that have come up over the last maybe 10 to 15 years that kind of really got us thinking about how the receptors are working and also questioning whether these are the only receptors that these hormones work through. And we're gonna talk about kind of some of the work that we've done to address that looking primarily at the ovarian phenotype to look at some questions. Okay, so we're at the point where we thought, okay, can we use the LH receptor knockout mouse to kind of repurpose it to address these fundamental questions in gonadotropin receptor and gonadotropin hormone radiated reproductive physiology? And so the study questions that we addressed were the first of looking at the role of LH receptor dimerization, and Olivia has kind of set the scene for talking about GPCR dimerization and if that has a role in directing ovarian function and female fertility. So we reported a number of years ago on the male, but we never addressed the female. And so that was our first question. Our second was talking about HCG and whether it can activate alternative receptors to the LH receptor. There's a little bit of data out there that perhaps suggests that it can. And so we were really looking at it from an ovarian function and that kind of fundamental role that you would associate with the LH receptor and whether it can work in a different way. And then recently, very recently, our group reported that the FSH receptor, constitutive activation of the FSH receptor can actually overcome the hypergonadal phenotype of males and restore fertility of the LH receptor knockout mice through constitutive activation. And so we wanted to ask that same question of the females and can constitutive activation of the FSH receptor overcome that ovulation block within the LH receptor knockout mice. And so at a basic level, we have three different kind of approaches that we use to answer the three different questions. And so our first was using, but so always we're gonna use the base of the LH receptor knockout mouse model. So that's our background kind of strain, if you like. And so in our first model, we co-expressed an LH receptor ligand binding deficient receptor with a signaling deficient receptor, which I'll show you in a second. We have evidence that they functionally cooperate and work through dimerization or oligomerization to ask the question about the role of LH receptor dimerization in directing female reproductive physiology. Our next was a simple kind of co-expression, or not co-expression, the LH receptor knockout with overexpression of HCG beta of the HCG beta subunit to address whether HCG is acting via different receptors. And then our third was utilizing a constitutively active FSH receptor mouse model to see whether this could overcome the ovulation block of the LH receptor mouse model. So we're going to start with the first one. And we are, by way of background, we're interested in this because, I suppose, and those of you that have seen me speak before, I apologize, this will look like a familiar slide. The GPCRs, I think, until 20 to 30 years ago, this was kind of a dogmatic idea that they act as monomers with a single receptor, binding a single hormone, activating a single pathway. As Livio has kind of elegantly described, and there's lots of literature in the GPCR world that supports this, that GPCRs actually exist as, can exist as homodimers or homoologamers, or in different flavors, as Livio has described, and some of our work kind of supports as well, as heteromers kind of associating with different receptors. And this has been shown to be an efficient and important way to diversify signaling in terms of specificity, selectivity, and amplitude. And some of our work is kind of looking at the roles that this has in specifying specific physiological events as well. So we initially started with a question, okay, so we have this and we know that it happens in a cell culture dish, but what happens in terms of reproductive physiology and is this a relevant way that these receptors work in vivo? And so the work by my colleague Adolfo Rivera-Muller, along with us, he so generated a receptor mutant, the LH receptor B minus, as we call it, that can bind hormone, so it was unable to bind LH or HCG for that matter. And then we generated a signaling deficient receptor, quite a brutal mutation as it was lacking two of the transmembrane domains, and so it can bind hormone, but it couldn't signal. And then when you co-express these receptors, they functionally cooperate to recapitulate the ability to bind hormone and the ability to signal through functional cooperation. And just as a proof of principle here, when you isolate lydic cells, you can see HCG-dependent kind of recapitulation of cyclic AMP production here in that purple line, okay. So we get recapitulation of LH or HCG-dependent cyclic AMP production in a cell culture dish and isolated cells, but what happens in the in vivo scenario? And I should say that these are batch transgenic animals as well. Okay, so when we look at the LH receptor lockout mouse, now these are the males that we have here. You can see the hypergonadal phenotype with small testis. When we co-express these receptors, we see recapitulation of full testosterone production and gonadal function. From a grace morphology perspective, it's exactly the same as the wild type, and by way of control, when you singly express these receptors, we don't actually see any recapitulation of function and it really resembles what the LH receptor knockout mouse is. So we know that for the male system, that functional cooperation or LH receptor dimerization is sufficient to activate and signal and is required for fertility, and so we went on to ask that question for the female, and the answer is it's complex, I think. So when we have the functional cooperation mutants co-expressed, which is this third panel here, it essentially resembles the LH receptor knockout, and so we're not seeing that recapitulation of signaling that we were seeing within the male scenario. When we look at the day of vaginal opening, so we have delayed puberty in the LH receptor knockout in the gray, and we see that delay in puberty still perpetuated. When we look at serum LH, serum LH is still elevated, showing that that lack of steroidal production from the ovary and that increased drive from the pituitary to try and stimulate ovarian steroidogenesis, and when we look at side chain cleavage, they're low, and essentially it resembles what the LH receptor knockout mouse is. And so we kind of concluded that functional cooperation was insufficient to rescue the ovulatory function of the LH receptor knockout mice, but we wanted to know why, because it works so elegantly for the males, why not the females? And so I think it's a bit of a complex mix of things. So firstly, when we compare the receptor expression levels to the wild type, we see about a fifth of the receptor expression level, okay? So we see really low levels of the expression, suggesting that we've got just very low levels of functional receptor. Now, one of the observations that we also made was looking at the relative ratio of the binding deficient to the signaling deficient receptors, and we did this because we know from our in vitro studies that the ratio of the two receptors is really important for recapitulating and promoting signaling, okay? And so we saw quite distinctly different ratios of the receptors between, in each of the individual mice that we assessed, suggesting that this may actually be one of the things that's underlying why we're not seeing sufficient signaling within these animals. And so we think it's in part the lack of a dynamic LH receptor expression that we know is required for ovulation, but also the ratio metric composition of these two receptors. And if we think about that a little bit more, and we think about kind of the functional roles that the LH receptor has within the ovary, and we compare that to males. Within the ovary, the LH receptor has a lot more to do. So if you think about just even in terms of the compartmental expression that the LH receptor has to do, and the functional roles within those compartments, so within the theca we've got androgen production, within granulosa cell when we get LH receptors actually induced in that gearing up for ovulation. We've got ovulation and luteinization, and then we've got the luteal cell function that the LH receptor has to do in terms of being that kind of powerhouse for progesterone production. Now within a human scenario, we've also got kind of what the LH receptor needs to do within pregnancy as well, so it's pretty complex. Whereas within a Leydig cell, it really is, you know, it's there, it's expressed, it produces androgen. So we're looking at quite a complex system within the female. And when we think about this a little bit more in terms of the signaling as well, and so turning back to some of our cell culture generated data, and looking at comparing kind of the wild type with the signaling and binding deficient co-expressed receptors, we know that when we add in HCG or LH, and we look at cyclic AMP dependent signaling in terms of GS signaling, that we get comparable levels of cyclic AMP signaling. Now when we look at GQ signaling, it's a different story. And with the functional cooperation mutants, this binding a signal deficiency, so we have impaired LH dependent signaling through both inositol phosphate and calcium when we measure those two different measures as GQ signaling. So really kind of showing, at least in the in vitro culture system, that functional cooperation isn't allowing GQ signaling or full GQ signaling to occur. And why might this be important? Well, this work from Mario Scolia a while back kind of showed in the conditional granulosa cell GQ knockouts that there was impairment in the number of ovulated oocytes. So that LH dependent GQ signaling is really important for ovulation. And so this is one of the mechanisms that we're also pondering, as well as it being low LH receptor number, we're also wondering whether that composite of signal transduction pathways that the receptor needs to activate are not able to be done via this model. Okay, so to summarize this, we have shown that the functional complementation approach fails to produce sufficient LH receptor expression. This may, in turn, have implications for the dynamic regulation of the LH receptor expression within granulosa cells. We are postulating, potentially, that we may have impaired GQ signaling and also, you know, this has implications for potential heteromerization with the FSH receptor, which we've also shown is important in additional studies for GQ signaling, LH dependent GQ signaling. And we've got some current studies underway, which are a little bit too premature to kind of show you at the moment, but to actually investigate the physiological regulation of LH receptor dimerization, but using a slightly different method to what we've used here. Okay, so we're now going to move on to HCGE and its potential, I suppose, alternative receptors that it can potentially work through, focusing on ovarian function. Okay, so thinking about kind of where that question has come from and why we might be interested in that, there have been a couple of reports, well, there are a few more than a couple, I'm just kind of highlighting a couple of these here. The first is that from Philippa Saunders' lab showed that in uterine natural killer cells, uterine natural killer cells, the co-expression of HCG with the mannose receptor, and they did a number of kind of functional analysis that suggested that HCG stimulated the proliferation of these uterine natural killer cells in a mannose receptor dependent way, and this was postulated to be important within receptivity and implantation. The second study, and perhaps, I suppose, a little more controversial, was in these aortic ring assay showing that hyperglycosylated HCG could stimulate angiogenesis, and it was postulated via the TGF-beta2 receptor that this worked through. And I say it's controversial because there was some kind of report saying that it may have actually been contamination of HCG with TGF-beta2 in follow-up studies, but nonetheless, this data existed and was out there. So we wanted to kind of ask the question about whether there were additional non-LH receptor pathways that HCG might be working through, and we focused on the ovary for this study. And so we turned to the HCG over-expressing mice, so HCG-beta2 over-expressing mice, so the over-expressing is in inverted commas because obviously they don't endogenously produce HCG. And these mice typically have between 10 and 40 times the circulating concentrations of what LH would be, so they're high levels of HCG. And typically, they present here—sorry, is that working? With— Your finger's covering it. Oh, thank you, George. With larger ovaries, they are infertile, and they have luteinized kind of cysts that develop, and hemorrhagic cysts. At the bottom here, we can see the uteri here, which are also enlarged within these animals. They have—although they have normal levels of estradiol, their progesterone levels are really high, so the ratio of estrogen to progesterone is very, very high, again, supporting this highly luteinized phenotype that we see within these mice—within the mouse ovaries. And so what we wanted to ask is, if the receptor is knocked out, are there alternative pathways that modulate ovarian function? And the bottom line is, not within the ovary. So, just to orientate you here, so we've got the ovarian histology kind of sections. We have the LH receptor knockout here, where we have follicles arrested at the large anterole stage, and we have a phenocopy, essentially, with the LH receptor knockout and co-expressing knockout and co-expressing with the HCG-beta animals. When we look at ovarian weights, the ovarian weights and the uteri weights are the same between the LERCO and the LERCO HCG-beta animals, and when we look at aromatase expression, it's similar between the knockouts and the HCG-beta animals as well. So, the ovarian phenotypes are phenocopies. The mice are arrested at diestrus, as you might expect, and the day of vaginal opening kind of as a measure of puberty is the same between the knockout and the HCG-beta animals. Again, so again, supporting overexpression of HCG, at least within this system, failed to rescue the phenotype of the LH receptor knockouts. So, to summarize this, we don't see any kind of phenotypic change, at least within the ovary of these animals, so suggesting that the LH substrates are working exclusively via the LH receptor within the ovary. I suppose the caveat to that is that we can't rule out species differences in findings or the extragonadal roles, and that's something that we have in the tissue, but we haven't looked at specifically. So, moving to that third and final kind of part of the talk and looking at constitutive activation of the FSH receptor in the background of the LH receptor knockout. And we were interested in looking at this, as I kind of alluded to earlier, because when you have the constitutive activating FSH receptor expressed within the Satoli cells, I should say, because it's under the AMH promoter, what you see is restoration of spermatogenesis in the LH receptor knockout mouse. Okay, and this is androgen independent. We still see high levels of serum LH and low levels of androgen within these mice. Okay, so we see recapitulation of spermatogenesis. So, we wanted to know what happens in the females. Are we seeing any restoration of ovulatory function within these animals? We, sorry, I should probably tell you a little bit about the phenotype of the constitutively active FSH receptor mice before I move on to that question. And so, again, for the females that just have the constitutive activation of the FSH receptor we see hemorrhagic cysts and we see premature follicle depletion of these animals. In terms of the breast tissue, we see enhanced ductal formation and breast budding. So, now we're going to ask the question, can the enhanced activity overcome the ovulation block of the LH receptor knockouts? Now, interestingly, if I orientate you here in terms of if we look at the gross histology first, we can see some progression here in terms of the ovary size and the uteri as well. We see some increase in progression of the large anterole follicles with separation of the cumulus oocyte complex. And in terms of the breast tissue, we also see enhanced ductal budding in comparison to the LH coanimals. So, we are seeing some enhancement of the phenotype, which we speculate is probably via enhanced oestrogen production within these animals. We see increased uterine weight and we also see enhanced or advanced puberty that's comparable to the wild type within these animals. However, although we see these advancements, we don't see the presence of corpus oetium within these animals, suggesting that they're not ovulating, but also they're acyclic as well. So, again, the mice have persistent diaestres and they remain anovulatory. So, to summarize, although the attainment of puberty resembles the wild type and we do get some progression of the ovarian phenotype and some breast budding, we're not seeing any restoration of ovulation within these animals. And so, what does this tell us about gonadotropin hormones? Ultimately, the LH receptor is required for the final stages of follicle maturation and ovulation. It's absolutely critical and it doesn't happen if it's not present. But there are some outstanding questions that still arise. And so, I think, really, we haven't really answered the question about LH receptor dimerization. I think the mouse model that we were using wasn't quite right for addressing that. So, we still really need to understand its role within modulating ovarian function. Still some questions about the extragonadal roles of gonadotropin hormones and their receptors. There's a lot of data. There's a couple of posters here alone that have been speaking about that. And, obviously, I think, as Dan alluded to, it's quite a hot topic, particularly within the FSH receptor world. And then roles of the heteromers as well for LH receptor, FSH receptor, as Livio kind of suggested, again, with the FSH receptor in GPA2. So, just to acknowledge the other people that did the work. You saw their photos kind of pop up at different points and our funding. So, thank you for listening. So, we're just about at the end. We have time for a quick question. Thanks. That's great. Maybe I'll just ask about the last model. With the constitutively active FSH receptor, is that expressed in all granulosa cells? Yes. So, it's the AMH promoter. So, it comes on quite early during follicular genesis. Because, as you well know, LH receptor is really only in the outer layer. I know. And so, do you think that that might contribute? Possibly. Possibly. I suppose that the question came really because with the male model, obviously, FSH receptor is only in the Satodi cells. But it's still overcoming the testosterone block, essentially, and still managing to mediate somatogenesis. So, as the LH receptor is expressed in those latter stages of follicular genesis in the granulosa cells. So, it's asking the question, but in a slightly different way. I'm not sure the model is nuanced enough to really fully understand. But it's, I suppose, it's doing, we're getting that enhanced cyclic AMP, whether it's detrimental or not. We're kind of unsure about that. But, yeah. And, George, can I ask one quick one? So, the point mutation that you have in the receptor, it makes the receptor constitutively active, but does it have any biases in its signaling? Does it fully recapitulate? Hasn't been looked at. The only pathway it's been looked at is cyclic AMP. Okay. Thanks. But it's a good question. Okay. Thank you, Dr. Jonas. And let's thank all the speakers for this interesting session. That was certainly worth staying for.
Video Summary
The video content consists of two separate studies presented by Dr. Livio Casarini and an additional study presented by another researcher. <br /><br />In Dr. Casarini's study, he challenges the notion that declining levels of FSH during the follicular stage lead to the selection of a single dominant follicle. He suggests that the FSH receptor may have a pro-apoptotic potential, which is masked by other hormones or growth factors in vivo but can be revealed in vitro. Through his experiments on granulosa cell lines, he found that FSH receptor-expressing cells died after one month, had high levels of cyclic NP and activated caspase-3, and produced progesterone. This implies that the FSH receptor has a dual role in promoting progesterone production and inducing apoptosis. He also introduces the G-protein coupled estrogen receptor (GPR) as a potential regulator of the FSH receptor in the ovary. This suggests that the selection of dominant follicles may involve interactions between the FSH receptor, cyclic NP, and GPR.<br /><br />The first study presented by another researcher examines the dimerization of LH receptors and its impact on ovarian function. Using a mouse model, the researchers found that functional cooperation between ligand-binding deficient and signaling deficient LH receptors was insufficient to rescue ovulatory function in females. Low expression levels of the receptors and impaired GQ signaling may contribute to this. <br /><br />The second study investigates alternative receptors through which HCG can act. Overexpressing HCG in a mouse model did not rescue the ovarian phenotype of LH receptor knockout mice, suggesting that HCG primarily operates through the LH receptor in the ovary.<br /><br />The final study explores the potential use of constitutively active FSH receptors to restore ovulation in LH receptor knockout mice. While these mice showed advances in puberty and ovarian phenotype, ovulation was not restored.<br /><br />Overall, these studies emphasize the crucial role of the LH receptor in follicle maturation and ovulation, and suggest limited involvement of dimerization and alternative receptors in female reproductive physiology.
Keywords
Dr. Livio Casarini
FSH receptor
follicular stage
dominant follicle
pro-apoptotic potential
granulosa cell lines
cyclic NP
activated caspase-3
progesterone production
G-protein coupled estrogen receptor
ovarian function
LH receptors
ovulation
EndoCareers
|
Contact Us
|
Privacy Policy
|
Terms of Use
CONNECT WITH US
© 2021 Copyright Endocrine Society. All rights reserved.
2055 L Street NW, Suite 600 | Washington, DC 20036
202.971.3636 | 888.363.6274
×