Transcript of Fernando Fierro’s Video Presentation

Hello. My name is Fernando Fierro. I’m going to share my screen now. I hope this works, it’s a little strange to be talking to myself, essentially. I really hope we get to meet in person sometime soon. I’m looking forward to that.

Slide 1: I will be talking in the next fifteen-twenty minutes about our work in my laboratory about how we’ve been trying to understand skeletal disorders that are so common to patients with Snyder-Robinson Syndrome.

Slide 2: The basic cell biology & maintenance of the bone is a dynamic process. Our bones have a turnover of approximately every ten years. We completely change the bone and this is due to balance of two cell types that have basically opposing function. One are these osteoclasts that resorb bone – they take bone out and they are derived from what’s called hematopoietic lineage. Hematopoietic stem cells give rise to these cells that take bone out. On the other hand, of course we need cells that are able to precipitate bone – these are the osteoblasts that eventually get further differentiated into osteocytes which are cells that are completely embedded in this precipitated calcium all around them. They come from a different lineage called the skeletal stem cells.

When we started to work with this condition Snyder-Robinson Syndrome, it was very difficult to ask really if the skeletal stem cells or the osteoclasts have some kind of misfunction. So what we do is model the disease by taking cells out from the bone marrow that [] to what we call here mesenchymal stem cells (MSC). This is a little bit confusing what I’m trying to explain, but these cells are really isolated in culture that have the ability to strongly differentiate into bone. And we’re going to go back to this in the future but basically it’s a surrogate culture model of what are the skeletal stem cells because the truth is that skeletal stem cells, outside the body, quickly change and they become indistinguishable to these cells that are defined by in vitro properties–by properties they have in the dish. This is what we call mesenchymal stromal cells or multipotent stromal cells – the name usually keeping the acronym MSC.

Slide 3: As you know many patients with SRS have very fragile bones– many have reported to have a bone fractured from just falling. That’s the underlying question of my laboratory: what’s going on? Why is it that these bones are so fragile?

Five years ago now, there was publish research – they took cells from the bone marrow of a patient with SRS, or I believe two patients actually, and they found suspected (because you know Snyder-Robinson Syndrome has to do with this mutation in spermine synthase gene) elevated levels of spermidine (the precursor) and very little of the product of the enzyme spermine in the osteoblasts. And they also took these cells that I told you about earlier that can expand in culture from bone marrow and what we called mesenchymal stem cells (MSC), and they showed that those cells actually showed a very deficient differentiation into the osteogenic lineage. The bone of course is normally white, but here [see picture] we stained the bone with something called alizarin red to make it look red, in order to confirm that it’s true calcium phosphates precipitates and you see that was deficient. But the problem with this study is that it was done with very few patients (one or two) and the cells are so rare and so few that its very hard to study them on a molecular level to understand what’s the underlying problem.

Slide 4: So, what we did was to model the disease – this is probably familiar to most of you—it’s called the polyamine pathway. Here is the enzyme SMS [see picture] that converts spermidine into spermine. What we did to model the disease is we took MSCs from healthy donors, which are bone marrow aspirates that are commercially available (so basically we have an endless supply of those cells). And to model the disease we simply silenced the gene SMS (spermine synthase). By silencing we used this genetic engineering approach that the total SMS levels were dramatically decreased here and mRNA level and protein level. The first thing that caught our attention—remember, we’re studying the cells in culture now, and modeling the disease by either keeping the gene normal (with the healthy cells) or turning the gene off. The advantage of this too is of course that we are comparing the cells from the same person, just with or without the gene. So you see there are multiple advantages to modeling the disease in this way. The main one, for me, is actually reproducing [?] its just the fact that we can do this 10, 20 times if we need to with MSC from different people and always find what’s the consistency here. This is one of those features: the cells clearly had remarkable loss in proliferation potential. This was the first thing that was apparent. Look here [see picture]: these are what normal MSCs look like in a culture, and we have way less if the cells were not expressing SMS or showed reduced levels of it. That’s published work, and I will still tell you more about this publication.

Slide 5: To do so, we’re interested in what happens with the bone formation of these cells – the ability to differentiate into osteoblasts and osteoclasts. I quickly need to review with you how that process works. It starts with proliferation of the stem cells and then there are the [?] of certain markers that indicate that cells are maturing toward osteoblasts. And finally they become these mineralizing cells that also have unique markers that are expressing in the late stages when the cells are actively precipitating calcium.

Slide 6: As you can see here, actually the early markers of differentiation were not effected. You don’t need to remember the names of this, but having silenced the SMS gene did not effect that original commitment of the stem cells towards the osteogenic lineage. But later, in the stages of maturation, we noticed a strong drop in a marker called bone sialoprotein (BSP). That reflected in D [see picture] with a decrease in the ability to mineralize bone- to precipitate calcium. As you can see here, this is again the red staining and here is the [?] of it. So this is very encouraging because it’s basically looking a lot like the data that had been published before with just one or two cases but with the advantage that the two samples derived from SRS patients.

To test if this was also the case in [?] (in an organism) because this is all created in a culture dish, we used the following model [see picture of mouse]. We took the MSCs that were either silencing SMS or not, and we implanted them into a scaffold and this scaffold was then implanted subcutaneously into immunodeficient mice. These mice are called NSG? And they have the ability to tolerate the human cells without rejecting them because they are deficient in their own immune system. So we can study human cells in a mouse thanks to their impaired immune system. The scaffolds with the cells are simply placed subcutaneously and they create bone there. It’s an ectopic bone formation assay. Then we measured by microCT – this is basically a x-ray based method- the bone density after a certain time you can see here that indeed having silenced SMS impaired bone formation in the mouse. We also did a statistical analysis but as you can see here the difference is not that obvious. At least here when we quantify the density of the bone, we could tell that silencing the gene had impaired that osteogenic differentiation.

Slide 7: To summarize the findings from the previous slide, silencing the SMS gene in MSCs had inhibited cell proliferation and osteogenic differentiation and the key question is, of course, what’s the underlying cause?

To find the underlying cause, we did two different approaches to look at the cells on a more molecular level. On the one hand, we looked at the gene expression profile of the cells and what turned out here is that surprisingly there were a little bit over 1000 genes differentially expressed because of having silenced SMS. That’s a little bit of an overwhelming number because its very hard to nail down which precise pathway or genes are effected- the truth is that many genes were changed. This is just a few examples of genes that we confirmed by our method called Real Time PCR. Of course, SMS had to be silenced, we measured that – but also some other genes of the polyamine pathway were differentially expressed perhaps as compensatory a/in? mechanism. Genes of course associated to cell proliferation were changed – that was not entirely surprising. These two genes here, CDK1 and CCND1 are genes that are high in proliferating cells and are now down regulated and this [the CDKN1 gene] is an inhibitor of [?] it marks inhibition of proliferation and that was indeed upregulated.

We also noticed some differential expression in genes associated to glucose metabolism and iron metabolism. This glucose metabolism is interesting because when we looked at the metabolites (MS) entered – here’s what we found.

We measured about 500 of these using something called mass spectrometry and the problem with metabolites is that most of them are just compounds that we don’t even have a name for, so it’s really hard to understand what’s their role. Out of these 564 measured metabolites, only 138 had actual names and could be associated to a specific metabolic pathway. And out of these 138, 33 increased and 12 decreased in MSCs with out SMS or with reduced SMS as compared to the control MSCs.

Slide 8: Here is a cartoon description of what we saw at the level of metabolites. You can see that spermine could not be detected in either of the conditions (both the control and the shSMS). We would have expected of course spermine to be reduced in shSMS, but we couldn’t really detect it in any of them, so it tells us that metabolite is in low levels to be detected. But, as expected, this green square indicates that spermidine was found increased, and interestingly, putrescine – the precursor of spermidine- was actually decreased, so perhaps there is the cell trying to use some compensatory mechanism to avoid this excess of spermidine. Here in the red dotted line is Ornithine. We used MSCs from four different donors – separated, without combining any of this, and all four of them showed consistently a reduction of Ornithine. But the magnitude of that reduction was variable, so statistically speaking, that difference was not significant. So the dotted line represents something that was consistent but on a quantitative level not statistically significant. The ones with continuous lines (putrescine and spermidine) were statistically significant.

And then, the other metabolites – many of them are associated with this glucose metabolism. Glucose is increased, this phosphoenolpyruvate (PEP) was also part increased, and citrate. As soon as you go into this [?] cycle you see that many of these metabolites were found decreased in the MSCs where SMS had been silenced. That brought us to think that maybe there was something wrong with the mitochondrias.

Slide 9: This here again are just measurements of glucose in the culture media of the cells. Silencing SMS, as you can see, made more glucose in the media. It tells us that the cells are not able to consume so much glucose and typical from an impaired crib cycle – I’m going to go back for a second—this pathway [pointing to shSMS image] when it’s blocked, what happens is that the cells cannot do crib cycle [goes back to previous slide] – this happens to the mitochondrias. When mitochondrias are defective or something is wrong with this metabolic pathway, cells simply do something called glycolysis that is anaerobic- a product of that glycolysis is something called lactate that gets secreted because it’s a product the cells cannot handle, so they just secrete it out into the media.

So you see that although the cells were indeed consuming less glucose (the glucose, remember was in the media not inside the cells) [back to slide 9], they also had increase secretion of lactate. So that’s really consistent with impaired mitochondria function.

Then we used this transmission electron microscopy—you can see here the mitochondria. They are the organelles here with the typical structures. What we found in the cells that have silenced SMS is that these mitochondria looked very large and fusing to each other. This is typical of mitochondria that aren’t working well.

We don’t have much time to go into the details of this but essentially what we are seeing is that having silenced SMS had effected many genes and metabolites and that much of this could actually correlate with impaired mitochondrial function. So the big question of course is what can we do about this? How can we move forward?

Slide 10: I think we could envision three approaches to overcome these obstacles.

One would be to use a pharmacological approach to improve the bone mineralization. I believe this is the current treatment for patients with SRS who have severe osteoporosis. They probably use bisphosphonates, or parathyroid hormone (PTH). Those are two common treatment approaches for patients who have osteoporosis, but they have their limitations. The main limitation of these pharmacological treatments is they work primarily by inhibiting the osteoclasts. I’m going to go back to that—it’s those cells involved with bone resorption, so you are stopping the resorption of bone. But an anabolic therapy – something where you are adding new mineralized bone to make bones become dense and as strong as they should normally be is [?] when using pharmacological approaches.

Another alternative – and this of course would be a little bit less common – would be gene therapy. Why don’t we think of a mechanism to replace the damaged gene with a healthy version of the gene? But the main problem I see with this approach is how do you deliver such a repair? You can’t go easily to all of the cells of the skeleton of a person with, for example, an antivirus to try to correct the gene. It seems impractical in terms of how to guide that.

What we have been exploring is a third option: cell replacement therapy. For that, I am going to go quickly back to this cartoon from earlier.

Slide 11: and remind you that what we are saying is that these cells are not really the skeletal stem cells what we really know is that these cells skeletal stem cells [?] without SMS are showing deficient mineralization, deficient differentiation here towards this osteoblast cells. This is the current mechanism [see picture] we’re using probably for SRS patients. You’re trying to inhibit this from these cells (osteoclasts) to have too many of them, or at least to reduce their activity so there’s less bone resorption. And clearly need more of these cells (osteoblasts)- and for that, what we envisioned was the idea of taking skeletal stem cell from a healthy donor OR taking skeletal stem cells from an SRS patient and then correct the gene and introduce those back, and now with the correct gene they should also do normal bone formation. This is possible because of very recent work on the skeletal stem cell—this cell type seems so obvious that we always knew it would exist because of course osteoblasts have to be replenished by some cell type in order to skeleton. But the truth is, they have been extremely difficult to find markers to isolate them. But now we do have this as of very recently.

Slide 12: I don’t the time to go into all of the details of this but essentially we know now that skeletal stem cells are characterized as cells that do not express this [] marker (CD45-) and are double positive for CD51+ and CD200+ — this small population of cells here. This is cells isolated from healthy donors, bone marrow aspirate. What we know is that if you take the skeletal stem cells (SSCs) and put them under normal culture conditions they become indistinguishable from MSCs. But I need to stress the notion that SSCs and MSCs are not the same. MSCs [goes back to previous slide- slide 11] are a rather heterogenous cell population that has a fairly even ability to differentiate to its bone under normal conditions. But in vivo, in the body of a person, they seem to correspond to at least three distinct cell types: the skeletal stem cells (SSCs), cells we associate with bone marrow stroma (cells supporting hematopoietic stem cells), and pericytes or perivascular cells. MSCs can be virtually isolated from any tissue of your body – it doesn’t have to be bone marrow—so basically telling you not all those cells are skeletal stem cells- SSCs are only cells that would recite in the skeleton and are in charge of course in maintaining your bone and responding when there’s a fracture.

[back to slide 12]: So SSCs I’m going to show you very quickly have very unique features that MSCs do not have. When I mean skeletal stem cells I mean the cells purified directly from a patient within hours and then immediately examined without ever putting them in culture (because that culture condition is what is changing the cells making them become something else from what they were when you really had them in your body).

Slide 13: This is one of their characteristics. We injected the SSCs (only 1200 cells which is very low for normal cell therapies) into our mouse – the immunodeficient mouse I mentioned before, but where we had caused a fracture in the femur. We injected the cells and as you can see here, the injection of just these few 1200 cells was enough to promote a robust improvement in bone mineralization. That’s also shown here: the red areas are parts of new bone and you’ll see it’s all increase with having injected these cells. I’m not going to go into detail of this, but basically the SSCs were actively participating in the deposit of new calcium- this is called []. This implies that the cells are actively engaged in the repair of bone by replacing precisely those missing osteoblasts. This is ten days after [see picture] you see lots of the cells in the bone [] are now red and colocalizing with this green mark of the new bone that is being added to the fracture site.

Slide 14: And, something that we really did not expect, is that if you inject these cells intravenously, the cells actually [] directly to the bone marrow- they go back to where they belong, they know where they belong. And you can see that here—this is an image of trabecular bone so there is a little bit of marrow and bone and lots of our labeled red cells—this is a pontification of that. Most of the cells were very nicely localizing the bone. This is something MSCs would not do. If you inject MSCs, unfortunately, a large majority of cells would get stuck and lungs and never make it to where they really belong with the bone or bone marrow.

Slide 15: So these two notions that the cells are a potent affect to promote bone repair in that they can home, made us think that maybe we could think of a therapy to improve the bone of SRS patients by injecting or implanting healthy skeletal stem cells. That’s based on the two observations I just mentioned.

We thought of investigating this with two different approaches. On the one hand, we used mouse SSCs so we could inject them into mice who are lacking spermine synthase. This is done in a conditional KO (cKO) Mouse and I’m going to explain that more in a second. The other approach is with human SSCs, But we can’t inject normal mice with human cells because they would reject them, they would clear from the system very quickly. So to overcome that, we would inject the human SSCs into immune deficient mice that are not missing SMS, they have normal SMS expression, but to model the osteoporosis we do an ovariectomy. The ovariectomy will mimic menopause and that estrogen deficiency is enough to cause osteoporosis, loss of bone density.

Slide 16: Back to the first approach I mentioned. We used conditional knock out mouse (cKO)- that means we take mice that the parents don’t have anything strange, they look completely normal, and we cross them and now these pups that are mice that have flanked SMS and the expression of something called Cre – this is a bunch of mouse genetics. The idea is that if we give these mice something called tamoxifen it will cause the depletion of an [] of the SMS gene basically causing mutation at the time point we can choose – not from birth. Then we waited for two months, hoping that during that time the mice would start to show a decline in bone density and then we injected the skeletal stem cells. We injected very few because these cells are extremely rare so we didn’t have many to inject. We waited another two months and at that point we euthanized the mice and looked for what happened with the bone density. During this time, we were curious that this mouse has been minimally characterized so far, so we wanted to see what happens to these mice.

Slide 17: We looked a little bit into behavioral features of them and one thing we noticed within this short time frame is that one out of the 9 cKO mice seemed deaf. Two of the 9 mice showed excessive shaking. This resembles what had been done previously by the group of Tony Pegg on their mice that I hope we can talk about at the conference in more detail. We used this method called rotarod to look at how much the mice can coordinate themselves and maintain themselves on a spinning wheel – we did not see significant difference, but as you compare here: a normal mouse showed an improvement where the time they would hold to the wheel without falling would increase over time (during the days after OTH), that is a measure of learning. The cKO mice did not show this same improvement. That suggests perhaps the cKO have intellectual disability.

The other phenotype that was very obvious is there was a strong stop in gaining weight of these mice. The mice that were silenced conditionally SMS made the mice not gain weight as the normally would have done. This looks very much like what Tony Pegg’s lab had found as well.

This is very preliminary and it’s just early stage, but it does seem like silencing SMS in these mice has an effect on the mice biology. Slide 18: However, the bone density, at least at the time points that we looked, was not decreased. You can see here, we measured vertebra and we measured femur and there was variation but not a clear decrease. And here, there might be a little bit of a trend becoming lower, but this difference still was not significant. The most disappointing part of this experiment was in fact that when we inject SSCs into the conditional knock out mouse (remember we had very few SSCs we only injected 3 mice), they were clearly not improving bone density — so they were either looking just as low or perhaps even looking a little lower. The main challenge in this experiment is that we could not see what we expected as a loss in bone density. Another challenge was the dose that we used of SSCs was too low to see a recovery.

Remember that the cells that are injected systemically and are spreading throughout the whole body to improve bones through the entire skeleton.

So we are currently repeating this experiment and we are looking for longer time points. We are primarily focused on seeing if there is a loss of mineral density in the mice where we have the cKO mice. The other suggestion from the polyamigos is that we might be silencing the gene too late in development. Perhaps we should be injecting tamoxifen to turn off the gene at birth, instead of at 2 months old.

Slide 19: The other model was a little more promising where we could inject the human SSCs. We had many more cells to inject. We caused the ovariectomy that caused estrogen deficiency. The mice then had a month and a half to lose bone density. Then we injected the cells and waited for another month and a half, then measured bone density. Here this is just vertebra, but you can see that the ovariectomy caused a significant drop in minerals in the bone density. With the skeletal stem cells, its not a significant improvement, but at least it has a trend towards the right direction. We are encouraged about these results. We want to repeat them and use a higher dose of SSCs. We have recently found methods to amplify the cells a little bit at least before they transform into something else and that’s what we’re doing right now.

Slide 20: This is the end of my presentation. Here are the key future experiments we want to do. First, and what we’re working on currently, repeating experiments in both SMS cKO an OVX mice, with special emphasis on characterizing those mice in terms of their bone biology. Another thing we would like to do now that we can purify SSCs, is to silence SMS there — Just like we did with the MSCs and culture—we can do that very quickly: take off the cells, silence the gene, and then immediately transport them back into the immune deficient mice. That could be very informative to see if it’s the lack of SMS in the SSCs that’s associated with the loss of bone density. Finally, an even bigger step back to understand the biology of SRS bone biology, would be to study the cellularity of the bone marrow of SRS patients. Now we have awesome technology called Single Cell RNA Sequencing, so we can look at what really happens to all the different cell types of the bone or bone marrow of them and see if maybe there’s an accumulation of one type or loss of certain cell type. This is just an example- we are doing this for another condition called fibrodysplasia- these are three different patients and we can, based on their gene expression, determine what are the cell types we are interrogating and this is actually very informative to understand what cell types/what in each of these cell types is going wrong because of the mutations in SMS.

Slide 21: with that I would like to finish here. This is my current lab. I would especially like to thank my grad student Bryan Le because he has been leading most of these studies. I also want to thank the Snyder-Robinson Foundation because they have been funding this research. I stop here and I hope we get to talk about this for our in-person meeting. Thank you. Bye.