Medical Conference 2012 - Hugh Taylor, MD

Medical Conference 2012 - Hugh Taylor, MD

Endometriosis Foundation of America
Medical Conference – 2012
Genetics, Epigenetics and Stem Cells
Hugh Taylor

First, I am honored to be here. I want to thank Dr. Seckin, not only for bestowing this tremendous honor on me but also for founding this fabulous society that he and Padma put together that I think will be a big boon to the field and to women.

Today I am going to talk to you about a few different topics, some of the research we have done over the last several years on endometriosis. I hope you will find it interesting, some exciting new findings, both in genetics, epigenetics and the role of stem cells in endometriosis.

First, what is the pathogenesis of endometriosis, an age old question that is in every introductory textbook on obstetrics and gynecology. We will learn about the theories for the origin of endometriosis. The most commonly accepted theory for the origin of endometriosis is retrograde menstruation, or Sampson’s Theory. But we have heard of theories of embryonic rests and colemic metaplasia. Clearly, as others have pointed out this morning there is certainly a familial tendency to develop endometriosis so there is a genetic component, and we will talk about that today. And, we will also talk about the role of stem cells and endometriosis, something that we have been working on for the last several years.

Why do we think though that most endometriosis comes from retrograde menstruation, consistent with Sampson’s Theory? Well the endometriosis tends to most commonly be found in the dependent portions of the pelvis where you would expect it to settle as a result of gravity. We see retrograde menstruation in almost all women so we know those cells get there. We see a very high incidence of endometriosis in women with obstruction of their outflow tract. If the cervix is somehow constricted more retrograde flow, and if you have no cervical patency you will almost inevitably get endometriosis. So we know that enough bulk of flow in a retrograde fashion certainly does lead to endometriosis. Also the tubes are often patent even some of the worst cases of endometriosis where we see a lot of adhesions, remarkably the tubes are often still patent allowing that retrograde flow. The risk factors that we heard about earlier this morning frequent menstruation, early menarche and probably most importantly the animal models that Dr. D’Hooghe and others have championed that even placing normal endometrium in the peritoneal cavity can lead to endometriosis. You do not need to have defective endometrium to begin with to get endometriosis.

Yet, I do not think Sampson’s Theory really explains all of the endometriosis we see. It certainly does not explain endometriosis outside of the peritoneal cavity. How does endometriosis get to those areas like the brain or the lungs where those of us who treat a lot of patients with endometriosis have certainly seen it? Nor does it explain endometriosis after hysterectomy, nor does it explain those rare reports of endometriosis in men, typically men being treated for prostate cancer and receiving estrogens; rare, but certainly that does not come from retrograde menstruation. So there must obviously be other etiologies for endometriosis.

First I would like to talk to you about our stem cell theory for the origin of endometriosis. First, what is a stem cell? I think many of you in the room are probably already familiar with the concept of stem cells. It is a relatively undifferentiated cell that cannot only replicate itself but that can give rise to more mature cell types. You know when most cells divide the two daughter cells are identical to the parent cell. But a stem cell replicates itself in undifferentiated fashion and then gives rise to these other progeny that can differentiate into more mature cell types. And we have these very undifferentiated stem cells that self-renew that are relatively quiescent, do not multiple much and can reconstitute tissues. They give rise to local progenitor stem cells that can give rise then to all of the specialized tissue in a particular organ system and they go through this phase of transient amplifying cells that are, again, much more restricted and not in this undifferentiated state. So there is a hierarchy of stem cells.

And the endometrium we have known, even before there was a concept of stem cells, that this highly regenerative tissue that undergoes complete renewal in each menstrual cycle probably has some sort of cell in the basalis layer of the endometrium that can give rise to a new endometrium every menstrual cycle. Again, even before we had a concept of stem cells we all had some theory about these regenerative cells in the basalis layer of the endometrium. Indeed, some of Caroline Gargett’s work who was the honoree here last year identified some of these progenitor stem cells that are in the endometrium and can regenerate the endometrium in each menstrual cycle or estrous cycle in a mouse. So it is probable that most endometriosis that arises through retrograde menstruation comes from these progenitor cells, these cells in the endometrium when they are shed in a retrograde fashion, and then can regenerate an entire endometrium or entire endometriosis lesion within the peritoneal cavity. But over the last 20 years we know that there are multipotent stem cells that are found in multiple tissues. They have a vast ability to differentiate into a huge number of cell types. They are multipotent. They are rare but some of them are found in the circulation, so they can move around the body and relocate to new places.

So the first thing we asked is could these stem cells from other locations potentially engraft the uterus? Could these be the cells that can repair and restore the uterus in the event of injury or repopulate the endometrium in the estrous cycle or menstrual cycle?  The first place we looked was the bone marrow. And the bone marrow is known to have a large number of these stem cells; not only stem cells that give rise to blood cells, the hematopoeitic stem cells that we commonly think of, but also mesenchymal stem cells that has the ability to give rise to a vast number of mature cell types including fat, muscle and epithelial cells for example. We reasoned that this might be a source that endometrial stem cells could be derived from. If bone marrow can give rise to all these different sorts of cells certainly they should be able to give rise to endometrium, which has this tremendous need for regeneration. So we asked whether bone marrow derived cells could repopulate the endometrium, first in a mouse model and here you can see we looked at Y chromosome fish. There is fluorescent in cytohybridization. The red dots are Y chromosome. We took male bone marrow and transplanted it into female mice and then looked for Y chromosome cells in the endometrium. “A” here is a control male tissue and you can see most of these cells express Y chromosome or show signal for Y chromosome. “B” is the uterine tissue, an endometrium from a mouse that got a female bone marrow transplant as a control. No Y chromosome signal. But here in “C” and “D” are the mice that got transplanted with male bone marrow, these are endometrial sections. Here is the endometrial lumen, here is the endometrial epithelium, and here is a Y chromosome bearing cell in the endometrial epithelium of this uterus after bone marrow transplant. Here is ___ stromal cell.

Now, after bone marrow transplant you would expect all the white blood cells of course to be derived from the bone marrow donor. Just to make sure we were not seeing some of the white blood cells that would naturally migrate in the uterus in each cycle. We also looked at CD45, which is a marker for all white blood cells and we looked at some other specific epithelial stromal cells markers. Again here is our male as a control, here is a white blood cell. This is not what we were looking for or this is not what we would expect in these. This green is showing CD45 a marker of leukocytes and it has got a Y chromosome. But here, in this panel, you can see the cell that has a Y chromosome. It is not green so it is not a white blood cell and it expresses cytokeratin which is a marker of epithelial cells. So that is an endometrial epithelial cell that came from that bone marrow. It is a male cell that is now in the endometrium.

Well, that is a mouse model. Could this happen in humans? So we asked if women who had had bone marrow transplants also could have some of these stem cells that eventually are incorporated in their endometrium. We looked at just a small number of women who had had a bone marrow transplant with a single HLA antigen mismatch that allowed us to identify the source or origin of any cell. They were reproductive age women, they had all had chemotherapy and total body irradiation followed by bone marrow transplant. And this is what we saw. The brown here is staining for the mismatched HLA antigen, that of the bone marrow donor. You can see that some of these cells have intercalated here in this epithelial lining, the blue or the endogenous cells, and some looked like they have the mismatched HLA antigen from the bone marrow donor. We looked in the stroma and here we serial stained for the CD45 to look at leukocytes and also the mismatched HLA antigen. These red cells here are the ones that are standing for CD45. Those are the leukocytes that you expect to be all of bone marrow donor origin of course. Somebody had a bone marrow transplant. The blue cells are the cytoplasm of those cells, are not staining. It is just the nucleus you see there. Those cells are the endogenous stromal cells. But the brown cells that the arrows point to, these are the cells standing for the mismatched HLA antigen of the bone marrow donor, they are not white cells, those are the stromal cells that are of bone marrow donor origin. Again, in humans we see stem cells going to endometrium. Stem cells are recruited to the uterus for repair and regeneration of the endometrium.

But of course we really want to know could stem cells contribute to disease, such as endometriosis? Again we created a mouse model of endometriosis. We sutured pieces of the uterus into the peritoneal cavity to make experimental endometriosis in mice. We used two different strains of mice; wild type mice and LacZ transgenic mice. And the LacZ gene we could just use here as a marker again to tell us the origin of any given cell. The mice had a hysterectomy so none of the cells that we still have migrating here were from the uterus. It could not be from retrograde menstruation, it could not be from hematogenous __ lymphatic spread of endometrial cells. We looked at then the expression of this LacZ gene in our endometriosis implants.

So to set this up, this is our wild type control mouse endometriosis. No staining for this LacZ gene. Here is the LacZ transgenic, every cell stains brown, and so those are the LacZ transgenics. When we transplant the wild type endometrium into the peritoneal cavity of the LacZ transgenics we get something that look like this – wild type, and here is the demarcation where you have the border between the endometriosis that we have created and the peritoneal cavity. So, what did we find? Well, after some time we started to see cells that were incorporated into the endometrium, endometrial cells that were LacZ expressing cells. So these cells came from a source other than the uterus. These are stem cells that differentiate here into stromal cells. This again is looking for immunologic staining of the LacZ Beta-galactosidase protein that is encoded by the LacZ gene. Just to make sure this was not an artifact of our immunochemistry we looked for enzymatic activity of that gene. The Beta-galactosidase activity turns a certain substrate blue. And here you can see this is the uterine lumen, this is an epithelial cell and this is a stromal cell. These cells come into this endometriosis, are not from a uterus, remember these mice have had a hysterectomy, these cells are migrating into the endometriosis and driving endometriosis.

This is a novel origin of endometriosis. Stem cells contribute to murine endometriosis. Ectopic stem cell differentiation can cause endometriosis and likely accounts for the endometriosis outside the peritoneal cavity, and perhaps endometriosis after hysterectomy. So even when we do a very complete job of excising endometriosis or even a hysterectomy, how do we still get endometriosis? Well, this may be one mechanism. More than that, I think this is a novel mechanism of disease – ectopic transdifferentiation of stem cells may indeed lead to other diseases. Endometriosis may only be the first example.

Just so you do not think this stem cell migration is all a bad thing, the fact that the endometrium and the uterus have these multipotent stem cells we can use for various therapeutic purposes. I will just go over this briefly as an aside to show some of the benefits of these stem cells. We can sort these cells, and I will not go over all the markers that we use here, but we can use facts sorting to pull out the stem cells from the endometrium. We can use various protocols that we essentially borrow from the embryonic stem cell world and we can differentiate these cells into all sorts of different mature cell types. Here we make chondrocytes cartilage cells. Here we actually got them to differentiate into neurons. Whether there is potential to differentiate into neurons may play a role in the nerve fiber scene in endometriosis patients’ endometrium is still a question. But they look like neurons, they connect and synapse with the surrounding cells. And when we create a Parkinson’s disease model in mice we got them to produce tyrosine hydroxylase, which is the rate limiting enzyme involved in dopamine production and we create Parkinson’s disease in mice. We transplant these cells and versus the control, we can help rescue dopamine production in these mice. So here are the un-lesioned animals. Here is what happens when we give them this medication that causes Parkinson’s disease. We transplant our endometrial derived stem cells into these mice and their dopamine production goes back up helping to partially rescue Parkinson’s disease in this animal model. These stem cells from the uterus have tremendous therapeutic potential.

Finally, more recently we were able to differentiate these cells into insulin producing cells that resembled pancreatic beta cells. Here is our culture. The green here is staining for insulin. Our control is up here. Here is our insulin production after beta cell differentiation and not only do these cells make insulin but in culture when we treat them with glucose they make more insulin. They are glucose responsive insulin production resembling exactly what we would want to see from a pancreatic beta cell. Here are the undifferentiated cells, do not make insulin, low glucose they make a little high glucose they make more. When we transplanted them into mice under the kidney capsule you can see that several of these cells continued to produce insulin and when we create a diabetes model in mice – here are glucose levels. These are untreated. Here are controls that did not receive stem cells. They received undifferentiated endometrial cells. And here are the mice that received stem cells from the endometrium that we differentiated into insulin producing cells. Their glucose control is better.

More dramatically, those were fasting levels. We think we probably evened out more of the peaks of glucose that we see after meals. More dramatically these mice looked dramatically different. The mice that were treated with our stem cells were not dehydrated. They were active. The mice that were not treated with stem cells but were treated with control endometrial cells that were not differentiated into insulin producing cells lost weight, became dehydrated, most of them developed cataracts and they were lethargic. You could easily tell from across the room; the good, the bad and the ugly of stem cells. The stem cells are there to help repair the uterus and regenerate the endometrium in each menstrual cycle. They can repopulate the endometrium, repair the uterus after injury but they also cause disease. They can contribute to endometriosis, again, accounting for some of those more unusual forms of endometriosis. But they also have tremendous potential as a new and easily accessible source of immunologically matched stem cells for women that can be used for tissue engineering and regenerative medicine.

Let’s shift gears a little bit now and talk to you about some of our findings that we just published a couple of weeks ago on a genetic basis for endometriosis. So, if all women have retrograde menstruation, and all women have these stem cells, why do only some women get endometriosis? Why do we see an increased risk of endometriosis if someone has a mother or sister with endometriosis? Clearly there is a genetic component. And many, including some people in this room, have been working on this. One of the best candidate genes is an endometriosis, perhaps quasi endometrioses, this is the KRAS gene. Another group several years ago published a paper showing that an activating mutation of an oncogenic KRAS gene leads to endometriosis in mice. This is really the only de novo endometriosis model in mice. We can transplant endometrium into the peritoneal cavity and create experimental endometriosis but mice with this genetic defect developed spontaneous endometriosis.

Is this a novel genetic etiology of endometriosis? Well, a lot of people starting looking then for mutations in the KRAS gene in women with endometriosis. Despite an extensive search and several publications no one has found one. But we hypothesize that there may be a defect, not in the KRAS gene itself but maybe in some of the regulators of KRAS expression. Interestingly, KRAS is regulated by a micro RNA termed Let-7, Let-7a in particular. For those of you who are not familiar with Micro RNAs these are small areas of the genome that do not encode protein. They are transcribed. They are cleaved. They are exported from the nucleus. They are then small, about 20 to 22 nucleotide single stranded RNA pieces that are incorporated in this risk complex. They bind to other genes where they have complementary base pairs and usually in the non-coding region, often in the 3’UTR of a gene. When these small RNAs bind they usually either lead to decreased ability of the RNA or block translation. So you get less of the gene produced. These Micro RNAs are very common and they regulate genes.

The KRAS gene has a binding site for this Let-7 Micro RNA and it is 3’UTR, not in the coding region of the gene but in this area that controls RNA stability and translation. When Let-7 binds it blocks translation of the KRAS. You get less KRAS protein. We wanted to look to see if there were potential mutations in this Let-7 binding region. Some polymorphisms in this region had been described in humans. We screened about 150 women with endometriosis for polymorphisms in this Let-7 binding site in the 3’UTR of the KRAS gene. We found that about 30 percent of the women with endometriosis; now these are my patients with endometriosis – it is a referral practice where commonly women who failed treatment with oral contraceptives have been referred from other gynecologists, so maybe women with more severe endometriosis – but about a third of them almost had this KRAS variant as opposed to only about five percent in a large diverse population. Idhaliz Flores is in the audience she was a collaborator with me on this study and many of the patients were hers – not just my referral population, mine and Idhaliz’s population.

What happens when they lose this binding site? Again, here is the Let-7 binding site. When you mutate that binding site the Let-7 does not bind. You get an A to G mutation. Let7a can no longer bind. We looked to see if that did indeed result in increased KRAS expression. When you have the loss of that Let-7 binding site we compared either normal women without the KRAS mutation to endometriosis patients who had a little bit higher KRAS expression. But we looked specifically at the endometriosis patients with the KRAS mutation, the expression of KRAS RNA was dramatically increased and indeed same thing, the KRAS protein was significantly increased.

To make sure that this really was this mutation that was regulating that and not something else about the cells from these women with endometriosis we took out just that 3’UTR of the KRAS gene. We made a genetic construct where we could put it in another cell type and just see what happens to KRAS expression or a reporter gene that is linked to that 3’ region in association with either the Let-7 or endogenously without the Let-7. Again, we have a reporter gene, a gene that just sends out a signal that lets us know the influence of this mutation or not mutated site in a cell type unrelated to endometriosis. So we know this is not just some other artifact of the endometriosis cells themselves but that it is specifically this particular site that is causing the change in expression of KRAS. When we look at the wild type KRAS 3’UTR in this assay fairly low expression. When we put the variant in with the mutation the signal goes way up showing that loss of that signal really drives gene expression up higher. When we now add a new Micro RNA that is designed specifically to bind the mutated site to mimic what the Let-7 was doing on the original site, now on the mutated site, we drive the expression of this KRAS back down. A control siRNA that does not specifically bind that site, does not do it. It looks like that mutation really is driving KRAS expression.

We also looked at what happens then to all the Let-7 micro RNAs. I told you Let-7a normally binds there but there is a whole family of Let-7 RNAs shown here and in women with this mutation they all had much lower expression of all of the Let-7 micro RNAs. Now Let-7 does not just bind KRAS. It binds a whole lot of things, and regulates a lot of different genes. When these micro RNAs bind they generally down regulate genes. When Let-7 is lower in women with this mutation it is no longer stopping the expression of many genes that are involved in mitotic signaling, not only RAS but genes like MYC. It is no longer stopping things involved in cell cycle like cyclin dependent kinases and cyclins and other genes that are involved in cell adhesion and migration. There may be more effects other than just those based on KRAS.

We looked indeed at how these cells behaved. We looked at BrdU incorporation which is a measure of how quickly they proliferate. The ones with the mutation proliferate faster. We looked at their ability to invade extra cellular matrix – how invasive are these cells? Those from endometriosis are a little more invasive but those with the mutation are the most invasive. It does seem to change the behavior of these cells. Finally, we took cells from women, with or without the mutation, and we transplanted them under the kidney capsule of immunodeficient mice. They both formed lesions that look like endometriosis but those were the mutation expressed gene called PCNA a little higher which is a gene involved in proliferation. It tells us these cells are proliferating faster and they had lower expression of progesterone receptor. Progesterone resistance as you know is a characteristic of some women with endometriosis.

This is the only clearly identified so far genetic cause of endometriosis. Knowing a gene responsible for endometriosis may allow us to do some sort of risk assessment especially in families where we know that this gene is the etiology of endometriosis. Again, it is not found in all women with endometriosis so it will not be a general screening for endometriosis. But in those who we know have KRAS related endometriosis it may allow us to screen family members and assess risk. It may be that different pathways are affected in this particular mutation. So it may allow us to individualize or personalize treatment. Then we may find that these particular variances of endometriosis respond differently than others. It also provides novel targets for treatment and right now we are doing experiments looking at altering some of the pathways that I showed you here to see if we can develop novel treatments for endometriosis.

 Other groups have done studies looking for genetic etiology of endometriosis and again, some of the people in this room have been involved in these studies. Some of the larger screening studies have been the Genome-wide Association or GWA studies that have tried to equate or correlate single nuclear type polymorphisms or those normal genetic variants that we all have with disease. One of the best studies, a large international consortium, identified one candidate gene that had an odds ratio of about 1.2, almost 1.4 for severe disease, was a SNP that was located upstream of the HOXA10 gene. Now those of you in the audience who know me know I have been working for the last 20 years on the HOXA10 gene, so this was music to my ears when I heard that this gene that we had associated with endometriosis may really be genetically associated with endometriosis as well. For those of you who are not familiar with this, this is some of our work we showed during the menstrual cycle in the uterus the expression of this gene goes up, this is early/mid proliferative phase, early/mid and late secretory phase, it goes up around the time of implantation. When we finally got a good antibody to it you can see it is the glands that start expressing this right in the middle of the window of implantation. It is the glandular expression that goes on that accounts for that up tick at the time of implantation. If you knock this gene out in a mouse they are infertile. The uterus is defective, implantation will not occur, embryos are fine if you take them out and put them into a surrogate, they do just fine. But if you put even a normal embryo in that uterus it will not implant. It is necessary for fertility.

We showed some time ago that in the endometrium of women with endometriosis you do not get that same cyclic increase in the expression of this gene. The blue shows the women…across the X axis is the menstrual cycle, the blue shows an increase across the menstrual cycle that I showed you earlier, the red shows the women with endometriosis. You fail to get that up regulation at the window of implantation, which I think accounts for some of the infertility we see.

Why is this gene expressed at a lower level? We turned to some of the animal models of endometriosis, both mouse and non-human primate. When we created experimental endometriosis we put endometrium in the peritoneal cavity and we looked back at the expression of several genes that are important for implantation in the uterus – they are affected. They did not have an abnormal endometrium to begin with; these were mice that were completely normal. But just placing normal endometrium in their peritoneal cavity changes the gene expression in the uterus including that of HOXA10, the gene that I showed you earlier.

In baboons, some work we did in collaboration with Asgi Fazleabas. We can transplant endometrium into the peritoneal cavity and get lesions that look just like human endometriosis. You follow out the expression of this HOXA10 gene and it goes down. It takes a while, not months, it took about a year before we saw a significant drop. Of course most of our patients have probably had endometriosis for years before we diagnose it. But it looks like endometriosis is having an effect on the endometrium. You do not have to have a genetic abnormality or abnormal endometrium to have an effect.

Transplantation of even normal endometrium gives rise to endometriosis and that in turn affects gene expression in the uterus. How does that happen?

Now I want to talk about the third topic that I promised I would get to in my title, Epigenetic Alterations in Endometriosis. It is not just genetic. Changes in DNA other than simple mutations can influence gene expression. We now know that DNA is very highly modified. CpG based pairs or cytosines in the CG based pairs can be methylated. Histones, which package the DNA can be methylated, acetylated, phosphorylated – many covalent modifications change the packaging of DNA and change gene expression. The HOXA10 gene that I showed you has a lot of these areas that have multiple CpG based pairs clustered together so-called CpG islands. We wondered if in the presence of endometriosis these might be methylated and that might explain the lower expression of this gene.

Well, indeed we looked in our mouse model and a particular sum CpG based pairs in the promoter region were very highly methylated compared to the controls. It looked like that methylation was indeed suppressing the expression of this gene in endometriosis. In our baboon model, the top boxes are the promoter region of this gene, so-called F1. This top block here are those disease free animals that were our controls that did not get endometriosis. Here are the animals with endometriosis. The promoter region of this gene was the only area that was methylated differentially and it was much more highly methylated in the baboons after creation of endometriosis. It shows that transplantation of even normal endometrium into the peritoneal cavity leads to methylation, decreased gene expression and _____ group in Shanghai has also shown that this same methylation pattern is seen in humans with endometriosis.  

So, increased methylation represses HOX expression. This expression, you can get there probably a couple of different ways. Anybody with endometriosis is going to have some methylation epigenic repression. But based on the genetic studies I showed you earlier maybe there are also some people that are born with the genetic propensity to have lower HOX expression, or an altered regulatory region of this gene may be one other etiology of low expression of this gene. This gene is necessary for normal differentiation in the endometrium, lower in women with endometriosis. Multiple different ways to get there but probably very similar end results.

Let me end by turning to our clinical treatments. Can we treat infertility associated with endometriosis and how do we treat it? Well, medically I think we all agree that there is really no role for medical therapy for endometriosis. Multiple studies that have looked at various drugs for treating endometriosis related infertility have shown no significant improvement. We do not use various medications for infertility treatment. They may work well for pain. How about surgical therapy? Well, there have been a couple of large prospective randomized trials, the larger being the endocan study. Indeed, surgery does improve fertility. This looks at weeks after randomization to either surgery for ablating and resecting endometriosis versus a diagnostic laparoscopy and the pregnancy rate goes up considerably when we actually remove the endometriosis rather than just look at it. But you will notice that the control group had about a 2.5 pregnancy rate per cycle and yes we can double the pregnancy rate to about five percent per cycle, but it does not go back to normal. Why not? Why do our treatments not work better? How does endometriosis affect fertility? Well it may be that we can be a great surgeon and remove all of the endometriosis. But based on the methylation that I showed you, based on some of the epigenetic changes that I showed you, you remove all the endometriosis, you may have helped prevent pain, but the changes in the uterus are left. These epigenetic changes are very slow to turn over. It may be that they do in time but they are not turning over rapidly. So, even when we do a perfect job with our surgery we may be left with some residual endometrial epigenetic changes and maybe not up to full cure. So, even when we take all the endometriosis out it could come from stem cells, even when we do a hysterectomy it could come from stem cells and find its way back into the peritoneal cavity or elsewhere. Even when we take all the endometriosis out of peritoneal cavity we may be left with some residual epigenetic defects that can still cause infertility.

I will summarize by saying endometriosis probably is not just one disease, it is a syndrome characterized by not only genetic susceptibilities and genetic forms of the disease but also cellular. It may be related to stem cells and epigenetic alterations that result in this ectopic localization of endometrium and defects also in the eutopic endometrium.

Endometriosis certainly has a genetic predisposition and we think we have found the first cause, genetic cause, of endometriosis in this KRAS variant. Stem cells contribute to both endometrium and endometriosis and the endometriosis induces certain irreversible epigenetic changes in the endometrium. It may make this disease not fully curable. We can get rid of the lesions but there may be some residual effects left behind.

I will summarize by saying endometriosis – most of it probably comes from retrograde menstruation but it also can come from stem cells. Everyone has both of these so some people probably have a genetic predisposition; either this KRAS/Let-7 binding site or maybe a HOXA10 mutation that was found in the GWA study. Those probably have lesions that are more prone to proliferation invasion. They may be more likely when they have retrograde menstruation to develop endometriosis. And anyone with endometriosis then develops a defective endometrium and the epigenetic alterations may very well mimic some of the same genetic defects that some people are born with.

Of course the environment can influence these, the medications we use, some of the exposures that were mentioned earlier may all impact on this process and that tells us also that we have the ability to modify the process and perhaps improve things for our patients.

I would like to end there. I want to thank some of my collaborators at our Yale Center for Endometriosis where we do much of this research, some of the people in my lab who have been involved in the endometriosis research. There are some other people working in the lab that are working on other projects so do not feel slighted that you are not mentioned there Leo. And some of my collaborators that have done some of the animal studies with me, and the NIH of course for financial support. Thank you.