Hugh Taylor, MD - Endometriosis and stem cells
Advancing the Science and Surgery of Endometriosis
Monday and Tuesday, April 19, 2016
The Union Club, New York
I just want to thank Dr. Seckin and the Foundation for inviting me here today. I want to thank also a few people in the audience who work with me in the laboratory. It was work that I will be talking about today. Thank you for coming.
Today I am going to talk to you about endometriosis, particularly some of the work we have done in the last few years about the role of stem cells and endometriosis. The message I want to leave you with, as your last speaker for this conference is that endometriosis is a systemic disease. What we see when we do surgery is only part of the story. Endometriosis really has a more diffuse presence. I will show you that through some of the work we have done on stem cells.
I think all of us know that the endometrium is a tissue that has this dramatic turnover. Every month with menses the entire endometrium sheds, it regrows from the basalis layer. It is no surprise that this tissue would have some stem cells that regenerate that endometrium in each monthly menstrual cycle. Indeed, our group and other groups have found stem cells that really regenerate the stromal cells, the epithelial cells, the endothelial cells, the vasculature undergoes a complete regeneration in each monthly estrus cycle. These we consider progenitor stem cells, stem cells that are already programmed to develop on to endometrial cells. But a few years ago we wondered whether other types of stem cells might also contribute to the endometrium. Bone marrow was a rich source of multipotent stem cells. These mesenchymal stem cells in the bone marrow were distinct from the hematopoietic bone marrow stem cells that give rise to the blood cells. But there are also these mesenchymal stem cells that have been shown to differentiate into all sorts of different types of cells, some of which are shown here and contribute to repair of some organ systems. Well, we thought that if these cells could give rise to so many other different cell types certainly with a rapid turnover we have seen in the endometrium they might very well be able to give rise to endometrium and contribute to endometrial repair and regeneration.
We did some simple experiments in mice and in humans to look to see if indeed stem cells from bone marrow are elsewhere and could contribute to endometrium. Here we transplanted male bone marrow into female mice and very simply looked to see if we could find endometrial cells with a Y chromosome. Very simple experiment and indeed we did. In panel A, the control on the upper left this is looking at Y chromosome by Y chromosome fish with little red dots in the nucleus of each shell are the Y chromosome and that is a male control and you can see most of the cells have a Y chromosome. About 85 percent of the cells we could detect a Y chromosome depending on where you slice through the cell you may miss it so whatever numbers we get are probably a little bit of an underestimate of the total number of Y chromosome bearing cells.
B is a female to female transplant. We looked at the uterus just as a control. You can see there are no Y chromosome bearing cells there. But in C and D you can see Y chromosome bearing cells after transplantation of male bone marrow into females looking at the uterus and see on the bottom left that cell is an endometrial epithelial cell. The black stripe is the lumen of the uterus and the epithelium is on either side. And in D in the lower right is an endometrial stromal cell that is derived from, again, the male bone marrow that was transplanted into this mouse.
I will not go through it all here today but we looked at various markers to be sure that these truly are cells that are turning into endometrial cells that are transdifferentiated as we say. For example we looked at cytokeratin, which is a marker of epithelial cells. You can see the yellow there is a cytokeratin and that cell on the right with the Y chromosome signal also expresses cytokeratin. So it is an endometrial epithelial cell that is derived from the bone marrow transplant. On the bottom left you can see a cell here that is colored green that is a leukocyte we screened for something called CB45, which is a marker of all white blood cells and we excluded those cells. We expect after a bone marrow transplant all the leukocytes will certainly be bone marrow derived but here we show clearly these cells are endometrial stromal and epithelial cells and again, derived from bone marrow.
And we looked in women as well. We looked in women who had undergone bone marrow transplant many years before biopsy and had some bleeding, clinical indication for a biopsy. These women had all undergone chemotherapy, total body irradiation and bone marrow transplant in the old fashioned way. In the old days we actually took bone marrow rather than peripherally mobilized themselves and transplanted that. And we looked for women who had got a bone marrow transplant with a single HLA antigen mismatch. We used the mismatched HLA antigen to identify the origin of any cell. This is the endometrium of a woman with a bone marrow transplant about ten years prior to our biopsy and the mismatched HLA antigen is stained in brown. The HLA antigen of the bone marrow donor. You can see in the epithelium on the left those brown cells intercalated into the epithelial layer are bone marrow donor origin, and on the right stromal cells. The arrow heads pointing to the blue nucleus of the endogenous stromal cells they do not stain for the mismatch HLA antigen. The brown cells with the arrows pointing to them are the cells that express the mismatched HLA antigen are bone marrow derived. And the red cells are leukocytes again that we would expect be bone marrow derived in anybody undergoing a bone marrow transplant. We know that leukocytes do migrate in and out of the endometrium in each menstrual cycle.
How functional is this? These cells do go to the endometrium they are not in very large numbers but can they really make a functional difference? Can they really play a clinically relevant role? Or is this just a curious phenomenon? Well here we took mice and we created Asherman’s Syndrome. So we injured the lining of the uterus, we did it a very reproducible way, we put a needle into the lumen of the uterus and scraped the uterus and showed that it resulted in some damage to the endometrium. In this strain of mice cumulative pregnancy rate is pretty close to 100 percent over several months. When we created Asherman’s Syndrome in the blue bar there we knocked that pregnancy rate down. It was about 30 percent. Yet, when we gave them a massive amount of bone marrow derived stem cells at the time of injury to help facilitate repair of the endometrium you can see the pregnancy rate is about 90 percent. So we were really able to rescue these mice and treat the Asherman’s Syndrome showing that these stem cells really do have an important functional role in the uterus.
How about endometriosis? What role do they have in the disease that brings us all here today? We think most endometriosis arises through retrograde menstruation, and I think it does. I am not trying to say that it does not. You do? You agree? Well I think we all agree that is how most of it arises. But certainly that is a source of endometriosis but what about endometriosis that we see outside of the peritoneal cavity in areas where you cannot reach through retrograde menstruation. What about endometriosis that we occasionally see in the lung, the brain, other places? What about in the old days when men who had prostate cancer were treated with high doses of estrogen? Some of them developed endometriosis. That is certainly not coming from their uterus. It is not hematogenous or lymphatic spread of endometrial cells. There really must be another source and we reason that maybe stem cells could be the source.
We did some simple experiments in mice. I will not go through this in a lot of detail with you but here we created endometriosis in mice. The brown cells on the left, the blue on the right are markers of stem cells from bone marrow that migrated into the endometriosis and continued to propagate the endometriosis. It plays an important role in endometriosis. We think that this probably accounts for some of the endometriosis that we see outside of the peritoneal cavity but it also continues to contribute on a regular basis to endometriosis in the peritoneal cavity as well. But I think it accounts probably for the majority of disease in those unusual locations remote from the peritoneal cavity, brain, lungs, where we occasionally find endometriosis.
What recruits these stem cells to the uterus? We have started to understand some of the molecular biology in some of the molecules that are involved. To make a long story short CXCL12 is a cytokine, also known as a stromal derived factor 1 (SDF1). Here we look at what we call a chemotactic index. We put the stem cells from bone marrow in one chamber and we watch their migration through a gel to see if they are attracted to something we put on the other side of that gel in the chamber. We set our control we just put the normal cell culture media on the far left. You can see we arbitrarily set that as 1 the natural propensity of these cells to migrate. But when we put them in media of cells that were obtained from human endometrial stromal cells you can see there is about an eight-fold increase in their migration. So these stem cells want to migrate through something that is secreted by endometrial stromal cells. When we treat them with AMD3100, which specifically blocks that CXCL12 receptor, the receptor is called CXCR4, it does responsive fashion, we can block that migration. That proves that CXCL12 is really necessary for that migration. As a matter of fact, if you just put CXCL12 in the well they will migrate to that so it is necessary and sufficient to attract these stem cells to, in this case, endometrial cells. Indeed the receptor at the bottom left shows the receptors expressed on bone marrow cells and here on the right in the center it is human endometrial cells express the cytokine, the CXCL12.
Interestingly it is expressed at much higher levels than endometriosis. In the top graph you can see SDF1. A sham is the normal endometrial expression, endometriosis in a mouse model left during induction massive levels of this cytokine that attracts the stem cells. If you remove the ovaries it drops down. It is estrogen dependent but still far higher than normal endometrium would. Again, you can even detect that protein. This is quantified below. But you can even detect that protein in the serum. The top one is a mouse and you can see even though very high levels are made locally there is enough produced and secreted into the bloodstream that you can actually detect higher levels in the circulation of a mouse or even in women with endometriosis shown on the bottom. It is not a lot higher because it is diluted as it gets into the bloodstream but it is the very, very high local concentrations that serve as a concentration gradient and attract the cells specifically to the endometriosis.
We looked again at endometriosis and how they attracted bone marrow derived stem cells. We asked for the treatment of endometriosis, medical treatment, here we used a serum, could reduce the traction of these bone marrow cells, the engraftment of these cells into the endometriosis and indeed it did. This is a percent of those white chromosome bearing bone marrow cells. You can see it is pretty dramatically decreased when we use medical therapy. So we think that one mechanism of action of some of our hormonal therapies is to actually block these stem cells that continue to feed these lesions.
But we saw something very different when we looked in the uterus. Now, in animal with endometriosis compared to those that did not have endometriosis, this is without treatment, they attract far fewer stem cells to their uterus. The stem cells are preferentially attracted to that endometriosis. Again, we showed you they produce much higher levels of the SDF1, which attracts the stem cells. There are a limited number of these stem cells in the circulation. We believe that this is essentially a competition for a limited pool of circulating stem cells that the endometriosis out competes the uterus and may be one of the mechanisms where we do not get adequate uterine repair endometrial regeneration and why we may see decreased implantation in women with endometriosis.
What happens when we treat the endometriosis? Well, looking at the uterus here, this is back of the uterus, when we treat the endometriosis we see we get a restored stem cell flux back into the uterus; so medical treatments may help to repair the endometriosis.
This just summarizes that data. On the left is no endometriosis, no treatment; yellow is the animal without endometriosis but with the treatment it had no effect on stem cell recruitment to the uterus; the red is you create endometriosis in those animals, the uterus does not get the stem cells you treat the endometriosis medically and the stem cell recruitment goes way back up. We have not done this surgically but I would imagine you would see the same thing with surgical treatment of endometriosis as well.
It is a novel mechanism of how endometriosis affects the uterus and fertility. We like to think of the endometriosis as a stem cell sponge. It competes with the uterus for this limited supply of stem cells that are important for uterine repair. The other thing we started to ask if these cells are circulating to the endometriosis, and some of these stem cells persist in the endometriosis, could they also then be leaving the endometriosis? Could there by a bi-directional flow? Indeed, here we looked at – we had tag cells and a mouse, we tagged them with a fluorescent protein DsRed and we looked in the circulation, we sorted these cells by facts from the blood and we could detect these cells in the circulation. These fluorescent cells that came from transplanted endometriosis in our mouse model. When we looked at various markers to try and characterize these cells again we sorted them for DsRed so they are all cells that came from the endometriosis that we transplanted into these mice. They have stem cell markers. They do not have hematopoietic stem cells markers, they have mesenchymal stem cell markers and they do not have the markers of mature endometrial cells. The endometriosis continues to give off more stem cells. They also have that CXCR4 which is the receptor for that SDF1 that recruits those cells.
It looks like endometriosis not only absorbs stem cells but it is giving off stem cells. Where do they end up? Well the first place we looked was the uterus. The example we had was in tumors. Often you will find that cells from a metastatic tumor can actually travel back to the primary tumor. There is some communication there. We wondered if the same thing was not happening with endometriosis? Indeed, in this light you really cannot see it well but the bottom line is by facts, by looking for the DS-Red protein or by immunofluorescence we were able to see these cells going back into the uterus. When they reach the uterus, here we used GFP in this one, it is another fluorescent protein to mark the cells we transplanted, they express markers of epithelial cells, such as here Wnt 7. These cells are at least some of them go back and become endometrial epithelial cells. But what I want to point out here, let me see if the pointer works, again the red is showing cytokeratin which is a marker of epithelial cells. Here the green is showing these cells that have migrated from the endometriosis back to the uterus. It is an epithelial cell but notice it is not in the gland, it is outside of the gland. Same thing here, Wnt 7 is another epithelial marker. This green cell expresses that Wnt 7 but is not incorporated into the gland. Same thing here, again expresses cytokeratin it is an epithelial cell but it is not in the gland. They are misincorporated into the uterus. They go back to the uterus but it is a defective incorporation and interferes with some of the communication between epithelial and stromal cells that again, probably contributes to infertility.
If these cells can go back to the uterus we started to wonder where else could they go and Elham, who is the audience here, actually did the work I am going to show you next. We started to ask if these cells from the endometriosis could go anywhere other than the uterus? And could this perhaps explain, if they do, some of the systemic manifestations that we see with endometriosis? Endometriosis patients complain of a lot more than just pelvic pain and infertility. There is immune dysfunction, there is differential pain sensitization, all sorts of other systemic manifestations of endometriosis. Well, Elham looked in lung, spleen, liver, brain and bottom line is in those that had endometriosis these little blue dots here show some of these fluorescent cells that came from the endometriosis incorporated into these tissues. If we look here the number of cells, it is fairly small the percentage is tiny, but they are there in the spleen, brain, liver and lung, at fairly higher levels in the liver and lung than the brain, which is even higher than the spleen.
So every organ she looked at and in every individual mouse that she looked at in each of these organs, right, they all had, they all had some cells from the endometriosis that were incorporated into these tissues. I will show you that here. Here is some of her work that shows in the endometriosis model the DsRed cells. They are not leukocytes, the CD45 is green that is leukocyte but you have separate DsRed cells here in the lung, the spleen, the liver and in the brain. They are all incorporated in all these tissues. Endometriosis is doing a lot more than just causing the disease that we see in the pelvis.
So I will summarize by saying again I think most endometriosis does come from retrograde menstruation, out the fallopian tubes and implants in the peritoneal cavity. But it also can come from bone marrow, bone marrow contributes to the uterus, it contributes to endometriosis, and I put this arrow much larger here because of the endometriosis, because of its inflammatory nature and high production of this CXCL12 out competes and attracts more of the stem cells in the uterus, so the endometriosis can come from bone marrow. But once you have endometriosis it gives off stem cells that can go back into the uterus but it can go to all sorts of other places as well. That paper describing this has not been published yet but is under review right now and we hope to have that out soon.
I think cell trafficking, particularly stem cell trafficking is really a central feature of endometriosis. Stem cell recruitment for repair of the uterus is inhibited by the endometriosis. Circulating stem cells are inappropriately incorporated both into endometriosis, into the endometrium and as I have shown you lots of other places as well. Exactly what damage that does or whether that contributes to the systemic manifestations of endometriosis is something still under investigation.
I will just conclude by saying endometriosis really is a complex chronic systemic disease. What we see when we operate on these patients with endometriosis, although very important, and probably the source of most of their pain, is not the whole story.