Scientist Stories: Michael Rosenblum, Exploring Regulatory T-Cells

This is episode 28 of the Immunology Podcast, exploring regulatory T cells with Dr. Michael Rosenbaum. Hey everyone, this is Dr. Jason Goldsmith and Dr. Brenda Rapp. Welcome back to the Immunology Podcast, where we have conversations with immunologists. The Immunology Podcast is brought to you by Stem Cell Technologies, a global biotechnology company supporting life science research and fostering communication and collaboration in science. Today we have Dr. Michael Rosenblum from the University of California, San Francisco on the podcast to talk about his research understanding how immune responses are regulated in tissues and how this knowledge can be explored for therapeutic benefit. In other words, it's Tregs and Brenda was very, very, very happy.

Yes.

We've also got our usual roundup of recent highlights in immunology news coming up. But first—

Stem Cell Technologies would like to introduce you to immune regulation news. A free weekly newsletter brought to you by the Stem Cell Science News Program. Covering research on the regulation, suppression, and modulation of the immune system, Immune Regulation News keeps readers current with the latest news, research, policy events, and jobs relevant to the immunology community. Subscribe for free at www.immuneregulationnews.com. Hey, Jason.

Hey, Brenda. So good news, because we've had a 1-year anniversary, they're celebrating our awesomeness and there's a swag contest going on. So people can go and subscribe to the Immunology Podcast newsletter at www.immunologypodcast.com/newsletter. And if you do it by May 15th, you could enter and win a Goldie wireless Bluetooth speaker. So they're giving away swag. Go and subscribe, folks. And if you don't win, you know, the newsletter lets you guys, lets everyone else know about our awesomeness and all the things we're up to.

For sure. I mean, you want to stay updated with the latest. So do subscribe. And I wonder, can I, if I subscribe, do I get into the, into the draw? Can I win this thing?

That's a good question. We haven't gotten these Bluetooth speakers, you know.

That would be nice. I would definitely post a picture of myself with it on Twitter, like enjoying it on the beach or something. I'm just saying.

Good point. We could do like the show where in the world has a Bluetooth speaker been? That could be a good one.

I love that.

You'd probably win that contest though. You could just like, you know, start taking a train and hit 15 countries before you know it.

Exactly. By the time you are in the next state, I've already been to Belgium, Paris, you know, Rome and back.

Yeah, no, I'm on the East Coast, so at least I can get to states quicker. When I was on the West Coast, all day to get to another state.

Yeah, there you go. And then we can, we can move on with today's science update.

Yeah, I think we pitched a COVID no-hitter, didn't we?

Yeah. I mean, I'm not sure if there are less papers coming out. Probably not. But I just, I feel like also looking at other— there's so much research coming, going on.

Are our COVID immune cells exhausted, Brenda?

Oh, gosh. Such a bad joke.

I know.

You know what? I was exhausted. Your jokes are exhausting, Jason. Just, why don't you just start with, uh, what, just, just, you know, just hit me with the science, please.

All right. Well, in my old age, I have trouble with my memory, so we'll discuss tissue-resident memory CD8 T cells and how they cooperate with CD4 T cells to drive compartmentalized immunopathology in the CNS. By Elena Vincenti is the first author. Last author is Neron Merkle. It came out in Science Translational Medicine on the 13th of April. This paper is actually pretty hard to describe. I had to read it a few times because it gets into kind of areas I don't know super well and some very interesting mouse models. So, so tissue-resonant memory cells have been implicated in pathology of autoimmune diseases. But how they work in the central nervous system hasn't been fully sorted out. So what they did is they created an inducible system in mice that with this tamoxifen inducible, uh, let me get the name of this gene. It's a glial fibrillinic acidic protein. So, and there's a promoter with this and then this LCMV glycoprotein. And so it makes this viral fusion setup where it secretes a glial protein upon induction. So, and this construct is attenuated compared to native regular LCMV, which is this neurovirus. And so they use this as a model of infection and then reinfection specifically. So an immune-induced and then a secondary effect later.. And so what they can do is they can infect the mice with LCMV-GFP or the LCMV or Delta GP infection, wait 6 weeks, and then they give TAM and then assess whether infection with this virus induces a secondary response., and I'm not doing this justice very well, but basically what they're showing is these T memory cells who have been trained from, or these, you know, resident memory cells, right? The T resident memory cells, TRMs, that have been trained by this virus to respond to this protein, then will re-respond after you have the self induce its own protein. So this neoantigen, this neo-self antigen, and that then drives inflammation and drives the downstream effects of motor neuron function loss. And so they spend some time demonstrating this system and it's CD8 T cell independent. And then they go and show that when the system's activated, it's— they're polyfunctional and they pull and they look— orchestrate local inflammation upon encountering the cognate self-antigen in the CNS, right? So because you— this virus— so the tamoxifen induces the cognate self-astrogen in the astrocytes and then causes inflammation there, which then causes brain damage., and so they show then, and they show this does require the resting CD8 T resident cells. If you deplete CD4 helper cells, this doesn't happen. And they also then show, interestingly, that if you do RNA-seq, it's not a clonal expansion. Effect. It's not like some single clone of a CD8 memory cell somewhere else comes and does the damage. It's recruited and then new cells activated. It's multiclonal inflammation upon repeat. So the memory cells are sitting there remembering, but then when it's time for the party to start again, it's new cells being recruited. And then they also show that there's, there's something like this in MS lesions, which we kind of already were having a sense of from other work that we've covered here before. So it's interesting in that it's showing that these memory cells, which are then baked from the initial infection, are driving things. Remember we learned that, well, I think CMV is a cause, they think root cause of MS, they're starting to get at. And so maybe it's the memory in the memory cells of the CMV, which has some cross neoantigen presentation that's driving the disease. And so how are you going to fix something like MS if you can't get rid of T, resident memory T cells. So how can you target those resident memory T cells with the neo, the bad neo self-antigen you shouldn't have? That's why I thought this was interesting. It's identifying a cell type that may be really dominant in the root cause of the pathology.

So that was cool. So they're calling, when they say resident tissue resident memory, what is the tissue exactly?

So this is the brain, the brain in this case, because this is all brain inflammation, but these memory cells exist elsewhere. Okay, so these cells, they, they sit there, and then when they see new neoantigen induced by the tamoxifen from the expressible system, they go, oh, now I need to attack this, and that causes the neural inflammation.

So they actually infiltrate the brain parenchyma, so to say, and then they act, they So they find these cells that are artificially expressing the glycoprotein that they've been primed against during a real actual infection. But then you're expressing this target protein with this construct that they have. And that's, that's why they see these cells as targets because they are artificially expressing this. Yep.

Yep. Exactly.

Okay.

And then if you think about what we know about CMV and MS, it starts lining up. Do they try anything about like preventing the recruitment of cells from like lymph nodes? CD4 helper cells are necessary for the function.

All right. But they're, they're done coming from, they're not being recruited from outside the brain.

They're already there. Right. There were no, so the memory cells are there. The CD4 and everything else are being recruited from everywhere.

Oh, okay.

So the CD8s are the ones that stay? CDA resident cells are. The other ones that come and join the party, the cytotoxics, everything else, are coming in.

All right, so there's more. So it's just adding to our understanding of, yeah, how autoimmune diseases are sustained in, in this, in, in this kind of situations. It's very interesting. Oh, these T cells, they're nice until they're bad. You know, but well, um, let me move on with my first paper of the day, which I thought was very interesting, and is titled Prolonged Viral Suppression with Anti-HIV-1 Antibody Therapy, uh, for, for HIV. Uh, first author Christian Gabler from the lab of Mikkel Nielsen-Slyke at Rockefeller University. And it was really cool because these are the results of a Phase 1 open-label randomized clinical trial in which they test the efficacy or this— mostly is the safety and the feasibility of treating patients with chronic HIV infection instead of with retroviral and antiretroviral therapy, with anti-HIV-1 antibodies. So more specifically, these broadly neutralizing antibodies that have been already identified that target glycoproteins of the HIV virus, of the HI virus. And they actually are really comparing and looking what are— what is the promise of this therapy in patients, which is very interesting because it would give patients with HIV an alternative therapy option, which might be interesting under certain circumstances. So basically, they have two different antibodies that have been previously characterized. And so they do this as a bitherapy. So it's already been shown that monotherapies can be problematic due to escape from the virus against that particular antigen is lost. So they are focusing on therapies with two antibodies. And they basically have a group of people that are— in which retroviral therapy is stopped for 20+ weeks. And they receive up to 7 injections of these antibodies, and they are followed throughout the whole study. And then they are, after the last dose, they are left without antiretroviral therapy until, in most cases, until they see rebound of viral titers. And so that's when they, in most cases, they need to restart antiretroviral therapy again. But what they do see is that in the cohort of patients that are treated, that are taken off ART, and they're treated with these antibodies, 70%, 13 out of the 17, it's a small trial, partly unfortunately because of COVID that has some issues with recruitment and with So they had to settle for a smaller group than originally expected. But well, at least they got to do this. 13 out of the 17 patients that were in this regimen maintained the viral suppression for at least— so the 20 weeks of the— well, while the ART was discontinued. So without antiretroviral therapy, they kept the viral suppression for 20 weeks. 5 months. And most of the cases, they also kept it for longer. So the median time to rebound was another extra 8.5 weeks. So these patients did not have any therapy for several weeks and still kept their viral titers low. And there was even 2 patients which, even after 48 weeks since the start of the treatment, so it would be 28 weeks after the last antibody therapy, they were still virally suppressed. And then apparently one of the patients was even virally suppressed after 2 years without detectable antiretroviral therapy treatment, which was, I think, was very, very interesting. So I think this is a therapy is very, very cool to see that this could be a potential actual viable therapy. And also, it has some interesting implications to what is— to the characteristics of what would be the viral reservoir of HIV, because HIV infects, amongst other lymphocytes, well, mainly lymphocytes. And even in patients that are treated with antiretroviral therapy, they have these latent reservoirs of genome, of viral genomes integrated in T cells. So that's why often when you stop the therapy, this just, these viruses get activated and they just start producing virus again and you can detect it in the plasma. What was very cool about treating patients with this antiviral therapy, with this antibody therapy, is that there was a reduction in the amount of fully functional, fully integrated viral genomes on this— on reservoir cells, mostly CD4 T cells, compared to people that just had antiretroviral therapy. So having this immunotherapeutic pressure on cells had an impact on the persistence of the kind of the productive viral integrations, which you don't see when you only treat with ARTs. And they was— unfortunately, the, the study was too short and there weren't enough patients for this to be really statistically robust, but they did see a tendency towards a reduction in the prevalence of intact proviruses within the cells from the blood of these patients. I thought that was really cool because maybe if you treat them long enough, you can have a substantial reduction in these intact proviruses, which might actually increase the time before relapse after therapy. So maybe we're going to have a therapeutic alternative to HIV treatment. That will be nice. However, of course, this is very specific for a type of virus, HIV-1. So it's a little bit less widely applicable, but I think it's a very interesting, very interesting approach.

Well, and you can make antibodies that float in the system for a while, which appeals to some people.

That's pretty cool.

Yeah. Are they probably going to then add it to combo therapy, though?

Is that what it looks like with the data? Yeah. So they do suggest that maybe you can have it as an extra to ART, but then maybe by having this both therapies, you can still affect the provirus reservoir. At the same time. So maybe if you give it long enough, they must actually reduce the proviral load even. So very interesting. They've been working on this for a long time. It's nice to see that they have this really cool data out, and I hope that they keep researching it.

That's awesome. All right, well, um, I don't have a good segue because HIV is all about T helper cells and this is about none of that, but that's okay. It's still about T cells, which you love, but NK cells too. Okay. Also in Science Translational Medicine, also on April 13th, we have NK cells limit therapeutic vaccine-induced CD8 T cell immunity in a PD-L1-dependent manner. Uh, first author here is Mariana O'Diniz. And last author is Mala K. Mani. So this is one that I think is really interesting. It's not about COVID but it's about vaccines. It's about hepatitis B. So if someone's chronically infected with hepatitis B and you give them hepatitis B vaccine, it doesn't work very well to cure them of hepatitis B. And this is because of T cell exhaustion. So this makes vaccines kind of suck for people already with an infection, right? So that's unfortunate. So they're trying to understand the mechanisms behind this and they do a, they look at cells that regulate CD8 T cell responses and lo and behold, NK cells regulate CD8 T cell responses. And of course, they do it through PD-L1. So PD-L1 on liver NK cells suppress PD-1 CD-high CD8 T cells within hepatitis B, within the liver, and suppress the hepatitis B response to vaccination. That's the conclusion here. So how do they go about this? They do some science. No. So they, um, use a multivariant vaccine with a chimpanzee adenoviral vector including hepatitis B core polymerase and surface antigens that's highly immunogenic, and it's being tested in phase 2 trials in humans with chronic hepatitis B. They use a mouse model, they infect mice They then harvest liver and spleens from said mice, and then they look at ex vivo analysis of CD8 T cells specific for the restricted epitope of the antigen, surface antigen, and look at the cell's response to the surface antigen, right? And see how well you're doing. So if they then deplete NK cell, NK1s, or NK cells, so that NK cells in humans can regulate T cells directed against HBV and chronic infection, but they didn't know if it was affecting the vaccine. So the general principle is known, but not the effect on vaccine. So if you do an antibody depletion of NK cells, boom, you have better immune response to the vaccine. And then they do some, and it really, you deplete before immunization. If you deplete after immunization, you don't get anything. It's get rid of the NK cells before immunization, after the infection, before you immunize. That's what restores vaccine functionality. And so that's neat. They show that induction PD-1 could be a relevant receptor for NK signal regulation. They do some sequencing to show this, and then they do a PD-1 blockade instead of NK cell depletion and show that PD-1 blockade causes the same effect as NK cell depletion in one or the other. And if you do both, you don't get any more. And how it works is PD-1 blockade enhances helper function of cytokine-activated NK cells. So these NK cells with this PD-1 activation, which is by default, you have less of the helper cell function, you have less immune cell response. So basically the idea here being if you can, if you were to give someone's chronically infected, you want to make them immune, right? You want to get a better immune response. Give them NK1 blockade or PD-L1 inhibitor, right, along with their, you know, right before their vaccine, like the day before, then you're going to have a better vaccine response. It's going to overcome the exhaustion from their chronic infection.

Do they speak about other cells in the liver that express PD-L1, or is this really just mostly NKs?

So this— so there's other cells that do it, but she's— they're talking about specifically that these NK cells are the culprit for suppressing their signaling suppresses helper cell function post-vaccination. Okay, that's quite interesting. So they're acting as a bad actor in the setting of a chronic infection.

Okay, so if you add anti-PD-L1 antibodies or you blockade that, then you can really allow a proper, a proper priming of the T cell, of the CD8 T cells. Exactly. Okay, very interesting. I wonder if that applies to other vaccinations.

I don't know, but it'd be interesting to look at, especially for chronic vaccines, for things that you have and you're trying to vaccinate to stop.

Yeah, indeed. All right, so I do have a segue to my paper, uh, last paper of the day. Uh, it also— it's also about, well, T cells, in this case CAR T cells. And the paper is titled CAR T cell killing requires the interferon gamma receptor pathway in solid but not liquid tumors, was published in Nature. Also the previous, I think, I don't think I mentioned the previous paper as well. First author Rebecca Larsen from the lab of Marcella Maus at Harvard. And in this paper, they're looking into trying to understand why is it that CAR T-cell therapy has such limited efficacy against solid tumors. And so they are focusing on one particular tumor type, glioblastoma, which is particularly— has a particularly low efficacy with CAR T-cells. And they tried, kind of set out to find some genes or some pathway that would improve or change the sensitivity of tumors to CAR T cells. So they have a system in which they have U87 glioblastoma cells that are expressing EGFR, and they have CAR T cells. Recognizing EGFR. That's their system. And so what they do is they have basically a co-culture and they measure the toxicity of these T cells. While after doing genome— so they do a genome-wide CRISPR knockout screen. So they have this CRISPR library in which they can target 19,000 genes. On this U87 glioblastoma cell line. And they look into kind of the— after co-culture with EGFR-targeting CAR-T cells, what are the tumor cells that are left? And they try to see which are the pathways that are underrepresented in the leftover cells, which would suggest— or overrepresented in the leftover cells, which would suggest a change in sensitivity towards CAR T-cell therapy. And the— in the major kind of hits they have, of course, the first one is EGFR. So loss of the target reduces the efficacy of the CAR T-cells. But then they have a bunch of genes related to interferon gamma signaling, which includes the JAK kinase 2, the interferon gamma receptor 1 and 2. And basically when they do kind of this pathway analysis, they see basically interferon gamma signaling being reduced in surviving tumor cells. So, which is— at the beginning they find this very curious because usually interferon gamma signaling in tumor cells is associated with mostly upregulation of MHC and therefore upregulation of antigen presentation. So it's very— this is very critical towards resistance of tumors against traditional TCR-directed T cell killing, which, because you need MHC molecules presenting antigens for T cells to recognize tumor cells. But this is not the case for CAR T cells because they have a CAR, so they don't need MHC anyway. So there must be something else that interferon gamma is doing to the tumor cells that is affecting their sensitivity towards CAR T cell treatment. So, they do this. So, in between, they validate this with— they do interferon gamma receptor 1 knockouts, and they do JAK2 knockouts, and they also do interferon gamma blockades. And they always see kind of the same tendency that interfering with interferon gamma signaling reduces the sensitivity of the cells towards CAR T-cell killing. They also check the other CARs, including one with a CD28 signaling domain. They also add— test a JAK inhibitor, ruxolitinib. Again, they also see if they pretreat tumor cells with this inhibitor, they see again a similar effect. And so they really kind of conclude that this interferon gamma signaling into the tumor cells is needed for the CAR T cells to engage with the target cells. They compare different solid tumors, and they have some pancreatic tumors and ovarian cancer cell lines, lung cancer cell lines, and they always see the same. They have several different models in which they see the same tendency. Interestingly, they compare it with hematological malignancies. They see a different picture. They have non-6 cells, but they are kind of a liquid tumor. And what is interesting is that they don't see this effect of resistance when interferon gamma is knocked out in these cells. If they have, for example, CAR-T cells against CD19 that is expressed on these cells. And so the question is, what is the difference? Why is this working? What is this affecting solid tumor response but not liquid tumor response?

And they start—

they do a bunch of things, but I think the most interesting is that they compare the rich pathways of cells exposed— so interferon gamma knockout and wild-type cells that are exposed to CAR T-cell killing, And they see that there's a particular enrichment on some cell adhesion pathways in the glioblastoma cell lines upon interferon gamma sensing, but this is not the case in leukemia cells. And they see that the CAR-T cells that are recognizing interferon gamma receptor 1 knockout cells form actually shorter synapses, so they stayed shorter time in contact with the tumor cell. Only in the case of glioblastoma cells. And they see that in particular ICAM is much less expressed in cells that lack interferon gamma signaling. ICAM, the ligand is LFA-1, which is expressed on T cells. And this is important for kind of stabilizing the immune synapses between the CAR T cell and the target tumor cell. They show that ICAM is upregulated when cells are exposed to interferon gamma, and they see that blocking of ICAM or knocking out ICAM can mitigate cytotoxicity from the CAR T cells to the tumor cells. So it's really showing that in the case of solid tumor cells, they don't see this in NOMC6 cells, only in the glioblastoma cells because that's when they're testing this. So this is a differential mechanism between the one and the other tumor type. And that kind of explains why only one of the tumor types has a different sensitivity when interferon gamma receptor is knocked out. What is interesting is that they do show that their data does not completely show that ICAM is the only the only gene because it's not complete. So if they— the absence of ICAM does not give the cells complete kind of immunity against the CAR-T cells. And they don't have— if they have like wild type and ICAM knockout cells, they don't see that the ICAM knockout cells really outgrow the the knockout, the wild-type cells. So I thought at the end I was a little bit confused because they don't seem to have— it was looking really good and then at the end it's like it's actually important but not crucial. But I do think that it's a nice food for thought about what are the kind of mechanisms that characterize CAR T-cell treatment in solid tumors that might help us understand how to improve this therapy for this purpose, which So far, it hasn't worked that well for solid tumors. So maybe this will now send us in the right direction.

Yeah. So is there— how much is known about regular T-cell killing of solid tumors generally?

I think there's a bunch. So when it comes to using T cells recognizing antigen and this expressed, presented on MHC. We know a lot and we do have therapeutic, certain, we have already had some data about the capacity of T cells to identify antigens being presented on MHC. And, but when it comes to CAR T cells, which is this weird thing, which you have the CAR, there's a chimeric antigen receptor that The signaling is this kind of funky thing. So it's not, it's not as clear. I think we know a lot more about regular TCR signaling than we know about CAR signaling.

Yeah, that's what I was wondering is where that missing link is, because my understanding is we know, you said like T cells can kill tumors, but getting it therapeutically to work isn't the same necessarily, especially if you're using modified T cells, which we are.

Yeah. And I also wonder whether the synapses, you know, these LFA-1, for example, locates itself around the synaptic area on regular synapses with TCRs. I wonder whether the CAR also allows for a— because you know the synapse, the T cell synapse is a very regulated and very organized structure. They have several rings of different adhesion molecules that have specific spots, and you have a lot of people studying how each individual molecule affects the final synapses, but I'm not sure if this is necessarily the case for CAR T cells, which are expressing something that is non-physiological at expression levels that are non-physiological. So they don't have all of the domains to bind properly to all of the other accessory molecules. So it's maybe, you know, that makes a huge difference.

Yeah, minimum functioning unit that matters, maybe more complex. Yeah. All right, well, today isn't about CAR T cells, but we're going to be speaking with Dr. Michael Rosenblum from the University of California, San Francisco in just a moment. But before we get to do that, are you looking for in-depth information on cell separation? Download the Cell Separation eBook from Stem Cell Technologies. It's a practical guide on everything you need to know about cell isolation techniques and includes a collection of protocols. Separate yourself from the pack.

Visit stemcell.com/cell-separation to download and explore the guide. Hello everyone, we are talking today to Dr. Michael Rosenblum. He is professor of dermatology at the University of San Francisco and is joining us today to talk about his research, I guess hopefully, on skin-infiltrating lymphocytes and other skin immune system So thank you so much for joining us today.

Thank you for having me. I'm really excited to be here.

Thank you for coming on board. All right. So Brenda's favorite cell is a T cell. Yes. Which we've admitted too many times in this podcast. I know. I'm giving too much information, I guess. You study Tregs, which are one of my favorite types of T cells. So where to start here? So skin T cells, skin inflammation. I think we, we've all experienced skin inflammation in our life. Um, when I think about Tregs though, at least in my universe, it's always been gut, gut, gut, gut, gut, and then secondarily the immune system, immuno-oncology. Where does the Treg field stand in the dermis, epidermis, skin, whatever you want to call it, in our understanding in relation to at least what I think of as where more of the work initially been done, especially with oral tolerance and all of that?

Yeah. You know, where does the skin sort of T-reg world, immunology world stand relative to the gut? I think we're catching up. So we're not there yet. You're right. There's been a lot more work done in the gut and in tumors and in skin, you know, we're catching up. But the nice thing about skin, I would say, is that it's easy to do translational work because it's so accessible. The human skin is accessible. And so, I think one of the advantages skin brings is the ability to not only study skin in rodent models as we do in immunology, but also to quickly be able to translate any findings to human beings because of the fact that we can get skin readily and sometimes in large amounts.

So we're catching up, Jason. Yeah. There's an awesome skin core at Penn I worked with when I was there that does some really good work and they had a similar comment. You can get skin a lot easier, which is nice. You can get biopsies. Yeah. Healthy people without as much of a pain as other tissue.

Absolutely. And actually there's large pieces of skin that are resected in different, in different procedures. So we get, we get not only small pieces, but we can get large pieces and sometimes even from diseased tissue we get large pieces. So it's opened up a lot of doors for us.

I would say, if I may ask, maybe going a little bit deeper into into the, the Tregs in the skin. Tregs come, you know, they're everywhere, like all lymphocytes. Um, but it's really interesting to understand, like, what makes a skin Treg a skin Treg, and how do they— because you have them, they have this, this population of Tregs that have this very characteristic, very particular characteristics. What makes a Treg a skin Treg? A member of the skin or an inhabitant of the skin, how does that compare to other T cells in the skin? And how does this compare with Tregs in other tissues as well, as Jason mentioned, the gut or even fat tissue or other of these types of niches?

Yeah. So I got to give a shout out to Diane Mathis, who did some pioneering work showing that Tregs not only live in secondary lymphoid organs, they like to live in barrier tissues as well. And two of the tissues they really like to live in is in the gut and the skin. And it's interesting, if you sample peripheral blood and you ask, you know, look at the chemokine receptor profiles of Tregs in peripheral blood, most of them are either on their way to the skin or the gut, with a small proportion also going to the lung. And so, Brenna, to answer to your question, so what makes a skin Treg a skin Treg? Well, It's really interesting. Once these cells migrate there and take up residence, they start to take on features of the tissue as dictated by their environment and sort of where they're living. And so, when you think of the gut, you think of a highly vascularized organ that's involved in a lot of exchange of nutrients. And it's a nutrient-rich environment, actually, because that's one of the focuses of the gut is to absorb nutrients. The skin, on the other hand, is completely different, right? It's actually made to keep things out. And it's not as well vascularized. And, you know, if you look at histology of the skin, it's kind of a desert. You know, it's a lot of fibrous tissue, you know, to keep the structural integrity of the organ. And so, the Tregs that are in skin have adapted to sort of— what makes them a little bit special is they've adapted to live in this environment of low, low nutrients, low oxygen, sort of a barrier tissue where you're trying to keep things out, not, not exchange nutrients. And so I know you said you've done a little bit of work in immunometabolism. So the metabolism of Tregs in skin is quite different than any other organ in that they've, they've adapted to exist in sort of this low nutrient, low oxygen environment. One of the other things I'll quickly say about Tregs in skin is that they like to be around hair follicles. So, a hair follicle is a mini organelle that's really— or a mini organ, I would say, that's— it's basically not present in any other organ. It's only in skin. And so, the Tregs in the skin have been specially adapted to live around hair follicles and play, I think, a significant role in hair follicle biology, which really makes them, I think, unique relative to Tregs living in other organs such as the gut and the lung.

So along that line, so I think hair follicles or follicles in general, I go to acne, right? Are Tregs really principally important in the response to acne? Do people, like, for instance, with one variant of a T cell reg response, have cystic acne or no acne or really bad inflammatory responses or what have you?

Yeah, so, so it's interesting. So acne hasn't been as well studied from a regulatory T-cell perspective as other diseases. And I think the reason for that is it primarily occurs on the face of adolescents. So it's hard to get tissue to really do a deep dive. Right. And for very, very good reason, I would say. And so I think, you know, speaking about the role, potential role of regulatory T-cells in acne, It's a little bit of a black box right now that we're trying to figure this out. I think going to people that have more acne on their back and trunk is one of the ways we can start trying to address this question. And we're thinking about, you know, and talking about doing that right now. But I'd say one of the places where Tregs have really been implicated in hair follicle biology is in autoimmune-mediated hair loss called alopecia areata. And actually, alopecia areata is one of the most common autoimmune diseases in people, more common, we think, than rheumatoid arthritis. And it is, if you look at the GWAS studies for alopecia areata, there's clearly a Treg signature. There's some really nice work done by Angela Cristiano. And so, we've also done some work in that field and shown that, you know, Tregs are definitely abnormal in the skin of patients with alopecia areata. And when you knock out Tregs just in the skin, which we've recently been able to do, you see this striking sort of hair follicle infiltrate type phenotype that results in hair loss. So I think that is the area where I think Tregs have been best shown to play a potentially pathogenic role It's in this field of autoimmune-related hair loss. I think they'll probably play a role in acne, but I think we just know a lot less at this point. How about pemphigoid? Pemphigoid is an interesting one. Actually, I'm also a practicing dermatologist, and that's the disease that I specialize in seeing. I see patients with pemphigus and pemphigoid. It's interesting that the pathology in pemphigoid and pemphigus is actually in the secondary lymphoid organs. It's a humoral disease of, you know, an antibody-mediated disease. And so, we think the Tregs probably play a role there, but the— I mean, obviously, I see the world through a Tregs lens. So, if you're going to ask me, I think they play a role in everything. But the Tregs actually in pemphigus and pemphigoid, which again are autoimmune-mediated blistering diseases by autoantibodies, mediated by autoantibodies, these autoantibodies are created in the secondary lymphoid organs such as the lymph nodes. And we think that there might be a defective immune regulation there, possibly abnormal Tregs that allow these autoantibodies to be generated.

There is clearly a lot of promise. Well, and I'm also— I also tend to see Tregs everywhere. And I guess that I'm also guilty of that. But I do feel like there's still so much untapped potential on harnessing Tregs for this kind of of diseases. And well, it's always exciting to see where the research will take us and also the clinical, the development in clinical trials and such. And I think we want to talk to you maybe about your incursions in clinical products. But first, I would just like to just insist a little bit on the science and some of your recent, recent publications. And particularly, I think it's a paper from last year in which you also looked into the effects of early life inflammation and how that actually affects the fibroblasts later on in life. So I just was hoping you can introduce our readers, our listeners very quickly to this story because I thought it was very interesting.

Absolutely. I'd love to. This is one of my favorite stories, actually. I got to give a shout out to a really, really awesome MD-PhD student in the lab that drove this project. His name is Ian Boothby. We started this story by trying to understand what the role of regulatory T cells are in skin early in life, because a former postdoc in my lab, Tiffany Scharschmidt, who's now a junior faculty member at UCSF, when she was a postdoc, she made this observation that Tregs accumulate in skin early in life in mouse skin, and to the point where like almost all the CD4s are Tregs in mouse skin at about day 13 of life. And what's interesting about that is this is really a period of time where the skin is still developing. If anyone's handled mice that are that young, they don't really have hair yet. They're sort of pink, and there's a lot of skin development and hair follicle development that's still happening postnatally. And there's just a lot of Tregs in the skin at that time point. And so, we hypothesized that possibly these Tregs were playing a role in early skin development. Making sure that, you know, there wasn't tons of inflammation while hair follicles were still being generated and the hair was poking through the skin and the microbiome was coming down through the hair follicle. The Tregs were playing a role sort of making sure that that all happened properly. And so, what Ian did for his project is he hypothesized that basically that these cells are playing a role in that developmental process. And he was able to delete them just during or deplete them from the skin just during that window of time where they're really accumulating early in life.. And he made this really interesting observation that what happened was these fibroblasts started to grow out. These bands of fibroblasts started to grow out deeper in the skin through sort of the subcutaneous fat, almost looking like a fibrosis kind of picture that actually resolved. And when you did the same thing in adults, you didn't see it in adult mice. So it was very specific to this neonatal time period. And we took a deep dive actually to look at these fibroblasts. And it turns out that these fibroblasts are immune-interacting fibroblasts. They're fibroblasts that have receptors for cytokines such as IL-13 and IL-4. And these fibroblasts played a role actually in maintaining specific immune cells in the skin. Specifically, they maintained, they helped maintain these Th2 cells. So these IL-4 and IL-5, IL-13-producing Th2 cells in the skin. And so what we basically found is that if you depleted Tregs early in life, you didn't allow them to accumulate in the skin, you established this fibroblast Th2 niche that stayed in the skin throughout the life of these animals. And basically, if you didn't deplete Tregs, you never saw this niche. Th2 cells aren't there. These fibroblasts are very low in proportion, but they grow out and create this niche if you deplete these Tregs. And that having this niche in the skin actually led to increased disease susceptibility later in life. So, when we wounded the skin later in life, they had increased Th2 response and enhanced wound healing. So, it was really sort of an interesting finding. And we basically coined this new sort of fibroblast subset, TIF cells, standing for Th2-interacting fascial fibroblasts because we think they come from the fascia underneath the skin., and they interact with Th2 cells to maintain them in the skin long-term. We've also translated this finding to humans. As I mentioned earlier, it's easier to do in skin. So we got some skin from patients with specific diseases and showed that these TIFs actually exist in human skin and potentially play a role in several Th2-mediated diseases that we see in humans. And so that's also this very interesting translational aspect of this work. And so, We're super excited about that. It started with Tregs, but then it went into stromal cell biology and fibroblasts. And it's an exciting new area for us that we're actively exploring and was uncovered by our ability to sort of manipulate Tregs early in life.

What I also found fascinating is the fact that you also showed in other work that you did that when it comes to resolving skin damage and fibrosis and preventing fibrosis, the Tregs in the skin are actually expressing GATA3, right? They are. Yes. So which is a transcription factor associated with Th2 responses. So I like because this is another example of this idea that Tregs are kind of mimicking the cells that they are antagonizing in a way. So, it's very interesting. I don't know if you—

whether when you realized that, it also surprised you. Yeah, it's super surprising. So, it's really interesting. And this is great work done by Dan Campbell and others, Sascha Denski and others showing that, yeah, Tregs really like to mimic the cells that they most potently suppress, right? So, it's sort of like if you're going to suppress a certain cell, you got to know what that cell is like, right? And so, you adopt a little bit of that. Of the features of that cell to be able to best suppress it. And GATA3 is one of those ones where you look at Tregs in skin, and you see that they express a lot of GATA3. And that's most likely for two reasons, I think. One is that they are uniquely poised to suppress Th2 immune responses in skin because it seems to be a place where Th2 immune responses, if they run rampant, can create bad things like eczema and atopic dermatitis. Things like that, and maybe even fibrosis. But also, the Th2 program is really a reparative program, right? Like, there's many people that think that the whole purpose of Th2 cells is to actually repair damaged tissue. And I think that's one of the main functions of Tregs in skin and in barrier tissues is to help repair, you know, not only suppress inflammation, but facilitate repair of the organ after it becomes damaged. And Jason, you mentioned everyone's had skin inflammation. Think about how many times a week we bump our skin. We run into things and we scrape it. And so, the skin has really a unique ability to repair itself, and it has to repair itself or the organism doesn't survive, right? And so, these Th2 cells play a major role in repairing damaged skin, and these Tregs play a major role in that process. Either helping repair independently of, of Th2s or actually suppressing that Th2 response so that repair doesn't become too fulminant and maybe turn into fibrosis.

So it sounds like similar to the gut, actually, there's a massive influx early in life of Tregs, very early with very critical responses so that you establish kind of your, your host microbiome interactions, which I know you've talked about back in papers in 2015. You, you know, your stem cell niche has a regulation by it, your wound healing repair. So where do you think it goes wrong in people? I mean, there's probably many ways it goes wrong, but is there, is there, is there any common threads you've been able to pull out where like Tregs gone bad that then seem to be a recurrent thing? I also ask this in context of, you know, the, the segue to, I know you have a company Yeah, which is named after a dinosaur.

Yes, it is. And we can get to that for sure.

But, but when you start thinking about it, it's like, okay, cool, T-Rex are important. Yes.

What do you do about it? Yeah. So, so speaking, getting to your first question, you know, do we know ways that they've gone awry in humans? Um, I mean, the clearest example of that, if you take it to the extreme, right, to make, to make the point, is the patients with eye the IPEX syndrome. In patients with the IPEX syndrome, the etiology of that disease is dysfunctional Tregs. They actually have Tregs, but they have mutations in FOXP3, which is the lineage-defining transcription factor for Tregs. And those mutations in FOXP3 render these cells dysfunctional. And those patients develop, you know, Th2 immune responses in the skin. Like, they develop atopic dermatitis, they develop a lot of other things, but when focusing on the skin, it's really this this atopic derm type phenotype. And so it makes me think that maybe patients that don't have IPEX but develop atopic dermatitis might have transiently had dysfunction in their Tregs early in life, maybe relative to microbiome interactions, maybe relative to other genetic factors. And then that's allowed them to have this decrease in Tregs or a limited time where the Tregs were needed to be high and they weren't high or they were dysfunctional in some way. And that set the stage for them to develop these sort of Th2-type responses throughout life and later in life. And I think the opposite of that might be the hygiene hypothesis where we know the more that the kids play in the dirt and get dirty, the more that they are protected from these diseases. And maybe that's augmenting their Tregs and really allowing this suppressive ability to actually max out such that they don't develop these immune cell niches, these potentially pathogenic immune cell niches. So I think that's one area that seems to lend itself to your question about, you know, are there examples where Tregs, you know, play a role in disease pathology in humans, and how does that work? Your other part of your question about the company. So I think that one of the really attractive things right now therapeutically is augmenting Tregs and the ability to manipulate these cells to increase regulation, to suppress autoimmunity inflammation, or to inhibit or deplete these cells to actually generate better antitumor responses. And I think either way, if you can make them better or knock them out, you have a play, you know, in terms of our potential therapy in autoimmune disease or cancer. And I started a company called Delinea originally that co-founded that with another friend of mine, Jeff Greve, where we developed a mutine that selectively basically augmented Tregs. And that's what went into clinical trials to try to treat autoimmune disease. And then multiple other companies have those mutines now, and some of them are showing really, really nice efficacy. In certain diseases with the preliminary data coming out. And then, as you mentioned, the dinosaur, I co-founded another company called T-Rex. And that was kind of a confusing name because it sounds a lot like T-reg, but it's a great company and it's focused on augmenting T-regs to treat autoimmune disease and also has some focus on inhibiting these cells in cancer, but mostly focused on autoimmune disease. And I think one of the interesting things that I would say is that I don't think the claim of anybody that's augmenting Tregs or inhibiting Tregs is that in all human diseases, it's a dysfunction of Tregs that results in human disease. But I think most of us will say that we know making them better will suppress the disease. So, it doesn't mean that the disease is a result of dysfunctional Tregs, but we know that if we can make them better or augment them, then we will be able to suppress the inflammation. And that's, I think, is the key sort of therapeutic hypothesis. In contrary, I don't believe that cancer arises because of Tregs, but I think if we can knock Tregs out in tumors, that we can develop better anti-tumor immune responses. And so that's sort of the the thinking around, you know, the therapeutic manipulation of Tregs in autoimmunity and cancer.

In your case. So we also had several episodes— several episodes ago, we spoke with Jeff Bluestone, who's also another person that thinks Tregs have therapeutic potential, but he's doing a different approach than yours. I know you cannot apparently disclose a lot about the products that you have, But maybe what kind of examples, or maybe things that are published that you can give our listeners as examples of how can you improve the function of Tregs for the purpose of therapy?

Yeah, I think, um, so I think there's two main approaches right now. And Jeff is a colleague and friend who's done some unbelievable work. And so the approach that Jeff is taking and others are to adoptively transfer Tregs into people and try to suppress inflammation, autoimmune responses. Our approach is to say the Tregs are already there. We're going to pharmacologically manipulate them using drugs, either make them better or delete them in the context of cancer. And so I would say that I can't divulge the specific programs that we're working on to do that, but I think the best Examples of that are published in that, and, you know, Shimon Sakaguchi originally identified Tregs based on the expression of the high-affinity IL-2 receptor. And IL-2 biology is intimately linked to Treg biology. And the first attempts at therapeutically manipulating Tregs have focused on the IL-2 pathway. And I think the whole idea that that Tregs express this high affinity IL-2 receptor marked by expression of the alpha chain CD25, there's been specific modalities that actually target just CD25, that receptor. And that's resulted in some really interesting, I think, clinical— some really interesting clinical efficacy in certain diseases. And I think that's That's one of the main things I'd point to, Brent, in terms of like what's out there in the published literature about the idea of being able to augment Tregs in the context of autoimmunity is go after that high affinity IL-2 receptor that's not only expressed by Tregs, but really highly and preferentially expressed by Tregs. And what's interesting though is we've sort of found out that, you know, once a Treg gets to a tissue, it might be a lot less dependent on IL-2. And so that approach may actually only be augmenting certain Tregs. And not necessarily the ones in the tissue that you really, you want to get. And so, that's yet to be determined, I think, clinically, but those studies are going on right now. And suffice to say that there are other ways to augment Tregs, specifically tissue Tregs. And I'm a big proponent of that because I feel like that's where the action is, right? You know, if you want to treat skin disease, you want to augment the Tregs that are actually in the skin. Not somewhere else in the body. And so, I point to the IL-2 approach and the IL-2 mutine-based approach as the best published examples of ways that, you know, you can manipulate Tregs for therapeutic benefits, specifically in inflammatory disease.

You're referring to kind of low-dose IL-2 as an— I said maybe paradoxically for those who think of IL-2 as a proleukin cytokine, but actually it is a Autoimmune treatment, the treatment for autoimmunity is low-dose IL-2.

Yes, that's one of the big ones. So that was the first one. And that's a lot of work being done by David Glatzman over in the UK. But I would say this, it's interesting. The IL-2 biology is very interesting because many people that took immunology way back in the day, you know, they studied IL-2 and it's like IL-2 is a growth factor for T cells. And when you knock IL-2 out, you should have no T cells. You know, they shouldn't be able to grow. And you should have basically an immunodeficient mouse. And then, the IL-2 knockout mouse was made and they developed fulminant autoimmunity. They have tons of T cells attacking all of their organs. And the reason is because IL-2 is actually required for Treg development. And if you don't have IL-2, you don't have Tregs, and that allows all the other immune cells to sort of go crazy. And so, that That aspect of Treg and IL-2 biology has been exploited now for therapeutic benefit. And you're right, low-dose IL-2 is the first, I think, parlay into that. And that was some really nice work done. I think the best work done at the Dana-Farber with Jerry Ritz, where he basically treated patients with chronic graft-versus-host disease with low-dose IL-2 and had really, really nice responses, specifically in the skin. And actually, they were able to show a decrease in fibrosis because some of the chronic graft-versus-host disease patients develop fibrotic disease in the skin. And low-dose IL-2 and the augmentation of Tregs was able to reverse some of that. And that's some of the earliest work. And now those patients are 6, 7 years out.

So really quick, you mentioned mutine.

I don't know if our audience knows what that is. Okay. Yeah, yeah. I'll explain. A mutine is a mutated protein that has specific characteristics that the that the wild type or native protein doesn't have. And so in this case, IL-2 is mutated so it does only bind to a specific receptor because the IL-2 receptor complex can be in the low affinity form or the high affinity form. And you can mutate IL-2 to only bind to the high affinity form of the receptor and not bind to the low affinity form. And so mutine is basically a mutated protein that has selective selective properties that you want it to have based on what you want it to do.

All right. So then my follow-up is, so you've gotten to found a few companies. I know often after you found and you can go on the websites, you're not in business meetings all day, but what has your experience been for people in kind of interest in the biotech world in terms of that process of, oh, I want to found a company and then you start talking to people and you put some pen to paper and then you have one. Then you need money.

I'm in the biotech world, but I'm at a different stage of the biotech side. Yeah. I would say that it's, you know, the things that I would say is it's super exciting. It's super exciting. And the reason why it's super exciting is because you're taking these scientific discoveries and you're trying to translate them to treat people that are suffering. And that's really, really exciting. I'd also say that it's a lot of work. So it's not just, you know, let's start a company. There's so much work that goes into it. And, you know, I consult for the companies that I've founded and I enjoy that, but it's a lot of work. And so it's one of these things where it's great reward, super exciting, but also, very time-consuming. And what's nice, actually, in my situation, what I've been able to do is I've been able to have research collaborations with the companies that I've started. So I'm able to not just, like, start them and kind of walk away, but start them and work with them through my lab at UCSF to help further the biology as the companies are furthering the drug development. And that's actually been super exciting and fun. And I think the people in my lab really enjoy it as well because it's sort of, again, that real translational aspect. I think a lot of us go into science because we want to help people. And this is a way where you really can see it happening.

I guess that this is a good segue to a final question. You, as you said, you are really busy doing all this entrepreneurship and funding, founding companies and doing research and impacting people's lives. So we like to ask a fun question at the end of our interview, and I guess it is appropriate to ask what is a hobby you always wanted to pursue but were never able to?

I guess because of lack of time. Yeah, lack of time. This one's really easy for me, and I think it has immunology implications. And so that's why I think this is a good one. So, so, um, I, I've, I'm a musician and I, I've, I've been playing guitar for about 20 years, but because of the pandemic, um, I was able to spend a little bit more time doing that. It just opened up my time a little bit. And so I really dove down, uh, and started playing a lot more guitar and started writing songs and, um, and started to actually had a colleague of mine who's also a scientist at UCSF, a developmental biologist, Arnu Richard Schneider, and another friend of mine who's the chief of neurology at Kaiser here in San Francisco, is also a musician. And so the three of us got together and we formed a band during the pandemic. And now we're recording an album, actually. And this is the best part. The name of the band is called Lightchain, which is based off the antibody structure light chain because this whole thing came together during COVID and everyone was talking about antibodies directed at COVID. So, we thought that Lightchain was a good name for the band. And yeah, we're right now in the studio recording an album. So, I really feel like I would have been a musician and the pandemic, the silver lining is it allowed me to to dive a little bit deeper into music. And as a consequence of that, I'm, I'm, I'm with the rest of my time basically pursuing that endeavor.

I, I have to say though, you cheated because it was supposed to be a hobby you could— you haven't been able to pursue, but you have. Congratulations.

I mean, that's, that's very exciting. There's a very famous cover band, you know, very famous, right, uh, in the GI community called GI Distress.

Okay.

Plays only at conferences. Okay. Do rehearsals like one time a year on Zoom ahead of time. So they've played more live than they've rehearsed. But I think there's, there's a proud tradition of, uh, of scientists and musician overlapping.

Well, I mean, in the immunometabolism field, there are, there, there is a band. I, I see you can join Luke O'Neill's band.

Yeah. The next meeting. Luke O'Neill has one. And then also Jim Allison.

I, I know.

Oh yeah, of course. Yeah. Nobel laureate Jim Allison plays in a band as well. So I don't know what it is. I mean, I mean, there are a lot of people are musicians, so I don't know if it's any strong association with science or not, you know, but, but I think science lends itself to be very creative. And I think there's other ways for creative outlets. And I think music is definitely another creative outlet. Jason, I see a guitar on your wall right there.

That's a Schecter Hellraiser.

You know the type of music I play.

Yeah, exactly. Well, thank you very much for coming on.

We hear you have already fans in the lab ready to go that we need to reach out at. Oh, yes, yes. I got to do a shout out to Madison Moss, who is a technician in my lab who came up to me yesterday and said, "Hey, Mike, I see you're going to do the Immunology Podcast." And I said, "Yeah, how did you know?" He's like, "Well, it's listed on the website, the people that they're going to have." And like, "It's the only podcast that I listen to." So big shout out to Madison, who is—

definitely listening to this.

Thank you, Madison.

We do it for people like you. You're the reason why we're doing this.

Thank you very much for those kind words. Well, thank you again for coming on and discussing not only Brenda's favorite cell type, but her favorite subtype.

Exactly. Very precisely, the cells I love. Thank you so much for joining.

It was great conversing with you today. Thank you so much for having me, guys.

Really appreciate it. That brings us to the end of our show. Don't forget to subscribe to our newsletter at www.immunologypodcast.com to get the show notes, including an episode summary and links to all of the interview and roundup papers. You can also reach out to us on Twitter @AminoPodcast or by email at info@immunologypodcast.com for feedback or to suggest guests. See you next time. Mm-hmm.