 All right, it's eight o'clock. We may have some people trickling in, but we'll go ahead and get started. This morning, we're lucky to have Dr. Ying Bing Fu, who's gonna be talking to us about age-related macular degeneration. He's originally from China, but completed his PhD training at Michigan State. And he's been here at the Moran since 2007. So it sounds like he's gonna be talking about a mouse model of age-related macular degeneration and possible implications. His role here at the Moran is he's currently the director for model development at Translational Medicine here at the Moran Center. So thanks, Dr. Fu, for coming and speaking with us this morning. Grand Ramp presentation at Moran. And mostly I work in the research side, but I also feel that deep in my heart, I'm connected to the clinical side. I still remember when I first came here for a job interview. I asked Robert specifically that I wanted my office to be on the clinical side so that at least physically I'm in the territory of the doctors and the residents. Today I'm gonna talk about our research work on age-related macular degeneration. Most of the results I'm gonna present today were published in August. And I also added many new results we obtained very recently. So as all of you know, that AMD is the leading cause of irreversible blinding. So AMD affects the region is the macular, very critical area, responsible for high-acuity vision such as reading and driving. And that's why it's so important. If you zoom in this area, so any damage occurs, all of us AMD can be separated into dry AMD and wet AMD and start to lose this information. So it all comes from majority of AMD and one of the most common early signs of dry AMD is frozen. Yano is deposited between brooks membrane and R. So on the other hand, the wet AMD is very serious for more AMD. And the healthy macular, I just mentioned there's three layers, photoreceptor, the corn photoreceptor, RPEs. And you see the cord vascular is a very nice, nice underneath. But if you look at the wet AMD, what happened is the cord vessels they break the RPE of brooks membrane and invade in the sub-regional space because the vessels tend to be very fragile and then you flew it not and cause serious damage to the mac. So there's a variant of what AMD is so-called polypoidal cord vascular naphthalene, which was first described by Columbia physician, Yano at 1982 as a cause of recurrent hemorrhagic and excessive heat-saving retinal detachment. And there are two key features on indocyanin-green angiography or ICGA. One on that was a branching of normal cord vessels, the other polypoidal vascular dilation. I will come back to this later. In terms of demography to the younger age so currently the pathophysiological mechanism for PCB download. So there are two major regions on the chromosome that can account to the majority of genetic risk for AMD. One is on the chromosome node side, one Q3, there's a well-known company in the factory edge region, a number of labs. Especially Dr. Hickerman's lab has done major contribution in this discovery. On the other region is on chromosome 10, that's centered around two nearby genes when high-temperature requirement effect A1 or HTI1, the other is arms two, no, 387715. So what do we know about the two genes? The HTI1 is highly conserved all the way from human to equinox, even after a family of Hitchcock, seven proteins involved in many functions such as protein quality control, there are growth differentiation and apoptosis. So in humans, we have HTI1234, you can see the 134 are very similar, you'll be quintessely expressed. HTI2 is mainly in mitochondria. One common feature is that they have a system like close domain as a causal type protein inhibitor domain suggesting that enzyme may be self-inhibiting and also has the insulin growth effect of binding domain and they have a signal sequence that are known as proteins to be secreted out of cells. So many functions of HTI1 would be implicated in a number of diseases such as cancer, your muscular disorder or Alzheimer's disease, arthritis and aging. Just reminding you, this is a genetic association studies, HTI1 has been linked to both PC and CMAs and also other types are in contrast to HTI1, very little is known about arms two. Other than that is a primary specific made in old world monkeys with a proposed function in mitochondria but even that is in this field right now. So it's not entirely clear whether it's really equal. Chromosome, one reason pretty much most people agree that the complement factor H is very important role. There's a great deal of controversy surrounding the two competing candidates. There are three major hypotheses. The first one is that increased protein level HTI1 is responsible for increased risk for AMD. The second major hypothesis is that increased level arms two is responsible for the increased risk of AMD. The third one is combination of both. So there's a binary model basically claimed both increase HTI1 and you can always see it's getting the reason for this great deal of controversy is that the two genes are very closely judged on. In other words, all those SNPs are in tight linkage this equilibrium makes genetic association studies very difficult to distinguish. Early studies, the studies on the genetic association studies I should say that that has been replicated in almost all ethnic groups in the world. So the genetic association studies obviously very strong on the contribution of chromosome 10 region to AMD risk. But however, afterwards, a number of labs trying to determine the functional involvement of the chromosome 10 region get a lot of difficulties. For example, the deniesion insertion polymorphism as I mentioned would show the decrease in spread your wonder which one is more important. And there's another polymorphism that's been called an early stop for the R38F associated with a protective or non-risk type was showed by a number of labs that in direct conflict with the idea that increase in labor arms too is associated with increased risk of AMD. And finally, the very recent paper published by Web of Group in Germany the strongest supporter of arms too they concede that currently unimaginably immediate pathogenic effect of the risk associated variance at a chromosome 10 region. So when we come to this project we think okay, the genetic study is quite difficult and we decided to take a functional approach by using animal models. So the idea is quite straightforward. We decided to transgenic spread human FTI1 in mouse RP and we asked a simple question will this transgenic mice exhibit any AMD phenotypes at all? The reason we can do that is that a human and a mouse FTI1 are highly conserved. Both of them have 480, I mean actually it's 91% identical. Even though the domains are almost the same, totally domain, very similar PDG domain, inhibitor domain. So what we predicted is that the human FTI1 when we put it into the mouse or interact with whatever the endogenous interaction partners of the mouse FTI1. So we generate the transgenic mice and we started characterizing them first looking at their expressions. So they looked at the message and ever by real-time RPPCR you can see the FTI1 was specifically expressed in the mouse RP not in the retina, liver, heart or skins. And even the western blood, the proteins we can detect in the transgenic. We're also trying to quantify the expression level of human FTI1 in the mouse RP. So this is by using the purified human recombinant in chemistry, trying to locate the human's consistent with the FTI1 being a protein that may be secreted out of the cells into the actual synonyms matrix. Okay, so now we have generated transgenic mice. Important questions don't as mice have any AMD like phenotypes. So we use a number of techniques. The in vivo imaging is actually exactly the same used by our pharmacists to examine AMD patients. Fluorescent, angiography, ICGA, founders image, OCD, even the machines, the hydra-broke spectrometer thing as used for patients diagnosis. And of course, with the mouse model, the added or the bound disease, we can do a pretty good histology to look at the architectures, okay? This is just a reminder of that. In the human mouse, there are two vascular beds, the cord vessels and inner retina vessels. So the FA is because the emission in the green florescence cannot penetrate the RP. The ICGA is because the infrared emissions, they can penetrate through the RP, you will see, both the retina vessels and the cord vessels, where the OCD, of course, give you in vivo cord sections. With the mouse model, as you can see, we, the machine, the actual analysis to focus from the anterior side to the posterior side. But the bottom line is both the FA and the ICGS are normal as you expected. So then when we started examining the transgenic mice, the FA is pretty normal, I will say. So the, such as the retina vessels, okay. But when we start to look at the ICGS, even starting from the anterior side, they start to see emission, you know. And we're focusing more, more toward posterior side, when we start seeing more emission. So I still remember the student Alex Jones, one of his first Guinness results. He was so excited. He ran out of the vivarium, trying to find me. And I don't know what it is, because with my limited knowledge at that time, or I was thinking AMD is the same way, which is the same region of FA, but a wide FA is normal and ICGS, I saw in these regions. So I called up Paul and Banna and asked them what happened and both of them said that resemble the PCV in human. So this is a composite ICG angiography, which show you almost like a three dimensional structures. You can see the wire types are normal structures, but in transgenic to the wide distribution of emission, they're just, right. So there's a more severe phenotype. Let's also show you another picture, show you really stunning resolution on the vessels, you can see more just reminding you the mouse eye is tiny compared with human eye. So to get this kind of resolution with mouse eye is quite challenging. The student did outstanding job. Okay, so when we started zooming on the bigger nations, if you look carefully, it's not really a bigger nation per se, but it consists of many smaller dots or it looks like really grip, cluster like structure on the branch. So it's very typical for polypoidal nations within the human PCV are patient on ICG. Also, we look at some of the normal porous vessels, branching network, you can see there's no structure of emission in a very similar, in other words, the mouse exactly this phenotype on ICG will fit very well with two key definitions of the human PCV. From that copy, you can see why type is pretty normal on the transgenic mass with all the, they locate actually same location, most of them on ICG emission, suggesting it's probably exudates from the corid vessels that's in the human PCV patients. Okay, so more recently, although with the, or those mice actually say pigment in the background, those on the black six background, very popular background. So with the ICG, you can see the emissions, but here you cannot see very well the vessel structures of the coroid. I mean, most of the vessels in the retina are much more clean on the mouse. So to get around this problem, we decided to take another approach is to bring all those mice into the CD1 is our final background, which will give you much better resolution on the corid vessels. So this one, because we focus on it, we are pretty confident majority of them are actually coroid vessels. And then in the wire type, it shows very normal structure. Okay, so when we start to see the transgenic mass, it's very interesting, we show a couple of them, you can see the way they're much more clear, giving a much better resolution on the black six background. So while still in the process of the anonymized name, also in combination with EMs to see the ultra structures. We also found there's a gradient phenotype variation. Some of the transgenic mouse has very weak phenotype with a few missions. Some of them have a moderate phenotype, some of them have very severe phenotype, you can see there's many missions. We also found there's a gradient variation in terms of phenotype progression. This is a mouse at two months of age. You can see the mission, when we examine three months later, there's not a huge change, I would say, pretty present, but if you look at the other one, it's seven months of age, but three months later, there are far more missions, it's kind of pretty active. And we also found there's a gradient, the influence of genetic backgrounds on the phenotypes. We examined at that time about 114 transgenic mice. Out of them, 67 showed PCV phenotypes with the other 47 denotes, so we call them PCV negative. If we took the PCV positive mice bred with the wild type, Black 6, out of them, majority of them, 78% developed PCV. However, if we took the PCV negative mice and bred with Black 6, majority of them don't have phenotypes. Only 28% developed PCV. So, pretty obvious, a genetic background has an influence on it. In the beginning, I was thinking, maybe the expression level of HTR1 is different in those different mice, but it actually turns out not the case. So you can see in both PCV positive and negative mice have the exact same level, suggesting other factors inside the mouse modulating HTR1 activity. And that's actually pretty interesting when I'm trying to determine what are those factors. So I will come back to this on the internet. The other question is, most of those mice, when we examine them in adulthood, the question is, because the HTR1 transient expression turned down very early on, the question is, do those mice have normal coiled vessels in the beginning? Maybe it's developmental issue. Maybe all those vessels were scrubbed in the very early stage, how do we know? Right, so just to remind you that, for mouse eyes, they don't open their eyes until about two weeks age. So about P14 days. And even that age for mouse, it's kind of difficult to imagine them. So again, we use the image of the mice on our final background, it's much easier. So the P14, wire type, you can say a picture not the best one because the mouse eyes are pretty, pretty smart. Just open their eyes. But we can tell not normal wire type transient mice, we don't see a huge differences with the image in the cup of man. But a few days later, about P19 days, you can see the wire types still normal. But in the transient mice, so based on this, it's such that those mice have normal cord vessels to begin with. But nature with constant accumulation, secretion of X-ray from RPs, they start to have these problems. So we are trying to now, doing EMs on histology, trying to confirm it. We also did all cities, kind of locate the nations, but based on FAs, ICG, we can guess, not just not keeping the cord vessels, but it's in the case. So for physicians that we remind you, so it's probably needed different from one or two of the human or cities. But anyway, so there's ICG, it's mostly located in the beginning. Okay, so with the mouse, we can do systematically look for this histology. And this is transient mice. One thing is actually is really very striking. We always say, two of red blood cells accumulated between the RP and the cord, very rarely seen in the wire type. And we think those were due to leakage from compromised cord vessels. And if we cut ultrasonic plastic sections, about 200 nanometers, you'll see a cluster of very single cord vessels and structure on the cord. There's a cross-section of the artery, you can see. Very nice new wall, very linear, it is doing all the layers of the transient mice, look at the artery, look how thin it is. Okay, so there are also great similarities between the transient mice and the human PCV in terms of histopathology. You can see the single vessel will show you transient mice and the cord vessels and in human PCV patients, you can always see also those very single cord vessels. The other thing is, when you look at retinal vascular journey, this is by Flamon standing with isonectin. So you can see, largely the transient mice are pretty normal vascular on the retinal vessel. This is consistent with the imaging, which I showed you that earlier, the FA is quite normal and also that effect with immunohistochemistry we have shown at TR1 was secreted out from the physical side of RT for the cord. We don't see much of the secretion to the other side, which is to the retinal. So I should say that most of the transient mice, we only have a classical CNV cases because the FA is quite normal but in some of the very old ones, they develop something looks like occult CNV, you can see that on the more than 11 months old on the transient mice, you can see it's speckled hypofluorescence and also this is ICGA. We're also trying to look for whether that's coronation between PCV severities and occult type CNVs. Turns out they don't have really a direct coronation. For example, because some of them have very severe different factors control the development of the two types or subtypes or where type M. Another very prominent feature of the transient mice is the degradation of the nut. Let me show you on the wire. That Brooks member really transient mice very striking that it's a very selective degradation of the quantified the integrity of the EL. Let me just show you that. The definition of the integrity of the EL is that the total length of the EL divided by total length of Brooks member. In other words, if it's 100%, that means no gap whatsoever. So in a wire type, it's pretty good, it's about 95%. The transient mice decrease significantly to about 64% bigger gaps. And the PCV neck is also much better. So this is pretty exciting to us because earlier work by the treatment team shows very nicely that in both early MB and the DMV that decrease thickness and the integrity of the mechanoelastic layer. You can see here compare with the controls. So the idea is that layer is very important or perhaps as a protective barrier to prevent DMV formations. And we think maybe one reason or mechanism for this breakdown of the EL in this region is due to actual activity. So far I have shown you that in the transient mice there are two prominent features. One is the degradation of the correlated vessels. The other is the degradation of the EL of the Brooks membrane. For both of them, a very straightforward explanation is due to the actuality's activity of HTI-1. However, it was not shown before the HTI-1 as inactivity, so we did this in vitro acid. Purified becoming the HTI-1 and then used the resin, quaint resin labeled inactin control with which is the pancreatic inactivity. You can see the HTI-1 indeed has inactivity for 30 times more. By the way, Dr. Hegeman's team also have the same result. We also come back to examine the transient mice because if that's the case, we should see some degraded inactin removal. Which is really the basic building block for the insoluble inactin nature. So you can see that increased degradation in that. Because the basal activity of human HTI-1 is so low, the question is that sufficient to explain the severe phenotype we observed. Just reminding you that HTI-1 is quite complex molecule as quite a few domains. It has been shown previously that both the N-terminal is causal type inhibitor domain and the C-terminal PTZ domain regulates HTI-1 activity. It kind of inhibits their activity. And let me explain why the basal activity of HTI-1 is quite low. And also the removal of inhibitor domain or binding of PTZ domain can activate HTI-1 protease activity by more than three-fourth with a combined effect of nine-fourth, okay. So what we think what happened is that because the basal activity is so low, there must be some other actocentral matrix proteins involved in activating HTI-1 and producing the PTZ phenotypes. That may explain why the variations of phenotypes we observed. We also explained why the Asians has higher prevalence with PTZ while the Caucasians has lower. So in addition to the two main features I just show you, we also observed photoreceptor degeneration and RT atrophies in the transgenic mice you can see in the wire types is the arterial structure very nice of the transgenic mice. On the other hand, with the PCV necromyte, they have a transient but they are pretty normal. When we look in their histology and arterial structure, they also play a very significant role in the progression or in the CME development. So we also explained the VEGF expression in the transgenic mice to see that in both PCV and necromyte, there's increased VEGF levels also by immunocompetible chemistry to show in the RT. Because it's increasing both the PCV past the necromyte, what we think is that increase at chance in the PEP while still trying to determine the mechanism for this. However, we think that VEGF and VEGF itself is more sufficient to induce PCVs. Although it's critical for PCV as in the pathogenesis. I may explain the relatively less effective treatment of anti-VEGF for PCVs in comparison with very effective treatment. So let me summarize what I have talked to you so far. What we conclude is that we show that by increasing the expression level of HGRI in the mouse, it's sufficient to cause PCV phenotype. And because the degradation in the group's membrane and also the inhibition of VEGF is also a significant risk factor for CNVs. So more recently, based on this mechanism and also by the mouse model, we're trying to develop some preventive and treatment strategy for both PCV and CNV by targeting the HGRI molecule. So the idea is to take this very potent peptide inhibitor that was first developed by Michael Hermann's group at Germany. So this is published more recently in 2011. So what they did is they crystallized the catalytic domain of human bacteria one. It's in complex with inhibitors. The challenge is because remember in the mouse model we developed the mouse occurred in the core vessels. The challenge is how to deliver the inhibitor in the posterior side of the core. The way we're trying to figure out this out is to use nanoparticles as a sustained release in the brain and by intervascular injection. So I want to go through all the new features, but quite a few studies to show that by this way you can have the RP take up the nanoparticle pretty nice and you can have sustained release anywhere between weeks to months, okay? So the nanoparticle was prepared by John Rainier at the chemistry department and we did a transmission, you have to look at that much in that complexity. So you can see almost like a hole in the meter, but the average diameter is about 80 nanometer. And then we did intervascular injections. The nanoparticle encapsulating there's a kumarine dye from the resin dye. So we can visualize it as a control. Four days after injection, you can see very nicely they go up to the junction between RP and our segment or the photoreceptor layers. And this is there quite preliminary with this data recently. So we inject of your transgegromite with the PCV phenotype. Before injection with the ICDA, the control is I would inject it with nanoparticles plus the kumarine dye, but the other I would inject it with nanoparticles plus the peptide inhibitors and the kumarine dye. So you can see with the control injection, we don't see much change after 15 days, but with the inhibitors, we can see a decrease of the lesions. So we still need to replicate it with more mice, but it's kind of encouraging, very encouraging to us. Just to remind you currently the treatment for PCV is quite limited. So there's, you can do, I think as far as I know, you can do thermodynamic treatment and also in combination with anti-VAGF, but in most cases that you can reduce the exotations but the lesions most cases they remain very, very resistant. And also in this, I think it's pretty difficult rescue because what we asked is actually in the artificial structure as a severe degeneration on the core of the vessel wars. So not only we have to stop the constant insult by the FTR1, but also we ask the vessels to repair itself. The other way we are currently doing now is that trying to prevent, so that is to inject inhibitors before the mouse develop any PCV phenotypes to see how much we can actually be made on set of phenotypes. So the implication is that maybe by doing so, ways in the future, if people have carried the polymorphism in the concentration region, maybe we can give them inhibitors to prevent this So the implications I just show you that what we think is the analysis activity of FTR1 seems to play a very prominent role in AMD. We also show you a neckly pathological mechanism for PCV development. I also go through that we are trying to develop some preventive and treatment strategies by using the mouse models and to target the FTR1 molecules. We also think that our results may explain why the genetic variance and the chromatin region are preferentially associated with the wet type AMD. And also may explain why the progression of wet AMD is strongly correlated with increase especially in the serum. You can detect that degraded in nothing pretty well. Maybe one factor is due to FTR1. So I will not go through all of them, but this is really time to address the mechanisms. Preventive therapy I already mentioned by the one exciting project that we are doing now is trying to generate a primary models of AMD really based on the success that on the mouse models and this is done in collaboration with a number of net Dr. Hickerman, Paul, and Sandra's lab. The reason is for the simple reason that you cannot really give out a true AMD model with the mouse because they don't have a mechanism. So if we can develop a really primate AMD model in terms of both study the mechanism and treatment it is, I think is a huge step forward. So most of the work was done by very talented student, Alex Jones, post-doc, Sandeep Kumar, and other people in the lab. Robomark's lab has really been instrumental in the EM studies. Bala Nino has been very helpful on the available imaging. Dr. Hickerman, Paul and Sandra are involved in the primate project and Kandah has been involved in the early phase of the project. I don't have time to talk about child care but she provided us the HTI one local device. Thank you, I'll be happy to answer your question. Transcending mice with design is to really express very specific only to the art but the other mouse model we are trying to develop is kind of with systemic expression of HTI one. That's an excellent step. Right, that's an excellent survival read. We don't know what happened. So we want to do this the other ways to get around this with the inducible promoters. So we can get around the developmental stage we can turn it down. Yeah, that's the next one. We have tried a few experiments but the students just get recently so I don't have time to do it but it's very exciting that if we took those mice and put on oxidative stress, so for example, one thing we do is we took those given them very bright light cause night in deals of oxidative stress. And what happened is those PCV-inactin mice quote unquote we saw started to develop PCVs. So pretty clear when you're giving them oxidative stress you make the worse. And then for the treatment, the efficient supplement. So we haven't looked, so we saw on artificial structures we haven't been able to do nothing but if you're, we did, I think we did a lesson to show that we do see increased degradation in lasting in the PCV positive but not in the PCV negative. Maybe not a direct answer. But let's go point the other way where I'm going to do microwave analysis to compare the gene expressions of multiple genes between H1 activity. If that's a case, that means in some cases or maybe we need to make by designing specific factors in H1 activity. The photoreceptor are pretty okay in the PCV negative. Yeah, they are pretty okay. About 80 to 100 nanometers. Oh, no. I think I still show a transmission EM. It's about 80 nanometers. It's reducing some of the injections. So they can always reach to the, I think that other people show that around 100, 150 nanometers. It's a good question. So when we generate transgenic mice it's on a mixed background. It's on the CBA, BlackSix or whatever used by the transgenic core facility. And in the gold standard in the theory you have a back cause with BlackSix for 10 generations. I mean we DNAs, I mean we obviously we're trying to cross them but so all the phenotype we show is in the early stage. One, five, three, four, five generations. So we know it starts from mixed backgrounds and then what we want to know is last month so he's very keen to develop the transgenic mice in a pure background. So the idea that I talk with the transgenic core facility with Susan, he said it's possible either on the pure BlackSix or CD1 backgrounds. It is possible. So it's much easier to inject to a pure background with the pseudo-martyrs than start with 10 generation brilliant ticks forever. Takes very long time. Yeah, that's a good question. But we start from the mixed backgrounds. There's a very little, as far as you can tell there are not many studies to address why the Asians and the Caucasians have different incidents of PCVs. And definitely it's a genetic factor but people don't know. We're trying to do this in the microwave as it's on the mice to say maybe give us some hint. If we find some interesting genes, we can go back to the human genetics. We're trying to develop some S-H-R-N-A's either by Plasmids or use virus. To develop a PCV negative one is more challenging because so far when we put either BlackSix or CD1 background, we're giving it my own pure background, so it's a pure PCV positive set. Oh, we like PCV negative, you know what I'm doing here. Yes, definitely.