 Good afternoon. I'm Carol Christ, Chancellor of UC Berkeley. It's my privilege and my pleasure to welcome all of you to this year's 109th annual Martin Meyerson Faculty Research Lectures. It is so wonderful to see you here in person and to be able to gather together for this event. For more than a century, Berkeley's Academic Senate has singled out distinguished members of our faculty whose research has changed the trajectory of their disciplines and of our understanding. These lectures shine a light on an essential part of our mission, the creation of new knowledge. The curiosity and creativity that fuel the quest to learn and to understand are at the heart of Berkeley's core commitment to making the world a better place through what we discover, through what we teach, and through the public service that we provide. This year's pair of lectures represent the continuation of a treasured tradition that is recurred annually with one exception. In 1919, in the wake of World War I and the influenza pandemic, the faculty research lectures were suspended. Unfortunately, there was no Zoom back then and our predecessors were unable to do as we did last year as virtual events were in vogue and these lectures went on. I should note that today's lecture will be recorded and available to view on YouTube or from the faculty research lectures website for those who are unable to join us today. Being selected to deliver a faculty research lecture is rightfully seen as one of the highest honors that Berkeley can bestow on its own, on a member of our faculty, to stand out among peers who exemplify academic excellence as no small thing. For students, members of the campus community, and the public we serve, this is all a unique and wonderful forum that offers access for all to scholarly research of the highest caliber. I'd like to ask that you join me in welcoming the past faculty research lecturers who are with us today. Professors, please stand as I read your name, but let's hold our applause until everyone has been recognized. The two individuals chosen by our Academic Senate to give the 2022 faculty research lecturers are David H. Rowley, who will speak to us today, and Timothy Hampton, who will deliver his lecture next Monday. A few words of introduction for Professor Rowley. David Rowley is distinguished professor of molecular and cell biology and the Esther and Wendy Scheckman chair in cancer biology. He also serves as faculty director of Berkeley's Immunotherapeutics and Vaccine Research Initiative. He's been recognized with many awards for his scientific contributions, including election as a fellow of the National Academy of Sciences and being named a distinguished fellow of the American Association of Immunologists. Our university has established itself as a center of breakthrough research in cancer immunotherapy, starting with the pioneering work of former faculty member and Nobel laureate Jim Allison. Today, David Rowley is not only carrying on this tradition of innovation and discovery, he's taking it in exciting new directions. Rowley's research addresses how the immune system recognizes and responds to cancer cells and virus-infected cells. While his early work focused on T lymphocytes, current research focuses on another cellular player in our immune system, the natural killer cell. Related to T cells, natural killer cells employ completely different strategies to attack cancerous and infected cells. And Relay has discovered essential ingredients of their capacity for recognizing and destroying many types of cancer cells. Relay's recent work aims to understand why in some cancer patients, natural killer cells fail to activate or become inactive. Part of his ongoing quest to decipher the mechanism of cell activation in order to devise therapeutic approaches that will mobilize natural killer cells to eliminate cancers. Today, he will present a lecture entitled Not All Killers Are Bad, How Natural Killer Cells Protect You from Cancer. Please join me in welcoming to the podium Professor David Rowley. Thank you. Thank you Chancellor Christ. And thank you for being our Chancellor by the way. We love you. Yeah, I'd like to talk today about natural killer cells and how they protect you from cancer and to explore ways that we may mobilize them for immunotherapy of cancer. To start with, I must disclose relationships with the biotechs that I have here. And I want to start by really acknowledging all the wonderful colleagues we have and that have made such an impact on the research in my laboratory over these I don't know, 30 years that I've been at Berkeley. So here are the colleagues in the division of Immunology and Molecular Medicine. It's really been an incredible experience to be around these people. They've had such an impact on our work and I'm really grateful for that. In addition to the current members I want to acknowledge former member Nelab Shastri who passed away recently and of course Jim Allison who Carol mentioned at such a formative role in our division and also in cancer therapy. And I also need to acknowledge the larger molecular and cell biology department I brought this slide from Rebecca Heald. So it's really a compilation of the much larger department and the impact of all the different kinds of research that goes on here at Berkeley and it's been incredibly informative for the work we do and I'm grateful for that as well. And finally of course I have to thank my laboratory and this is a historical talk. I'm going to cover some decades really of work that we've done in a sort of narrative form and I want to acknowledge all the people in lab who've done this work over these many years I think probably missing some people in these various group photos but this covers a lot of the time. So thank you guys. I don't know about everyone's work but the work we do is informed by the work that all these people did. And this is a recent group photo in a retreat that we had and I have to shout out to Lily Zong who's the manager of our lab who's been with me for 22 years now I think and really makes lab run well and she's I have to say puts up with me so that's a good thing. And in that light I have to acknowledge my family as well who are here. Michael and Gabriel and what they put up with me too and that's great and actually my sisters are here as well nice and recent and they put up with me for a really long time so thanks for that. So the main take away I'd like to make today for those of you who are not in the field is that curiosity driven studies of cellular and molecular mechanisms drive the invention of new medicines and I'll take you through our work that we hope is leading in that direction and mention as well some other work along the way. So I'm going to do a kind of deep historical dive here into the immunology and going back now to actually the 1800 when Jenner invented the smallpox vaccine and of course Pasteur invented several other vaccines in the late 1800s and what's remarkable about them is that they really had no concept of an immune system dedicated to protecting us against pathogens or cancer. They really didn't know how vaccination worked but did work and so they exploited it and it really wasn't until around 1899 or 1900 that antibodies were discovered by von Bering and over the next years immunity was attributed largely to antibodies these proteins that can recognize things or to phagocytes which are cells of course that eat other cells and it wasn't until the late 1960s that the era of cellular immunity was born really with the discovery of T-cells and these are cells that kill other cells in our body and the concept is that killing infected cells prevents a virus or other pathogen from replicating within that cell and therefore can limit the infection and killing nascent tumor cells obviously has obvious value as well. So this background I want to introduce T-cell recognition because it's really the starting point for my presentation so T-cells recognize what we call antigens and antigens are what the immune system recognizes so I think spike protein of COVID and COVID's been great for educating people about immunology a little bit but what's key in this recognition and different from antibodies is that it involves these proteins called MHC proteins that are displayed on the surface of every cell in our body so that's the blue thing here and they are antigen presenting molecules and what they do is they survey the otherwise hidden internal contents of cells for antigens and then they present them on the outside membrane of the cell for recognition by the immune system so in the example of a viral protein that's a long polypeptide chain it's degraded into peptide fragments which bind to this protein then they're displayed on the outside and then T-cells have a so-called antigen receptor which enables them to recognize that complex and they have different T-cells have different specificity for different antigens and when it interacts it's a... pathogens can run but they can't hide and inside cells is my point there so these antigen receptors are triggers and when they recognize and bind to the complex here it triggers them to dump cytotoxic molecules onto the surface of the target cell to kill it so that's really how both T-cells actually and antigen cells kill target cells although as we'll see the receptors used are different so now this was our understanding as around in the 1980s but it rapidly became clear there's many other molecules interacting between T-cells and other cells not just the T-cell antigen receptor but other molecules and many of them but there was some that were particularly interesting and that were being studied in the early 1990s by many of us and one that Jim Allison seized on was this receptor protein called CTLA4 and it's on T-cells and Jim proposed that it engages molecules on other cells and in so doing that it inhibits the T-cells so this was a different idea it was actually preventing killing of a target cell and he made this specific jump then to suggest that maybe this inhibits T-cell responses to cancer and that perhaps you could block that interaction with an antibody against CTLA4 and therefore unleash the T-cell to kill the cancer cell so this was the concept you block that interaction the inhibition is gone and now the T-cell can kill the target cell and remarkably that worked in the first animal experiments he did and that led to the development of this approach in industry and I have to give a shout out to my colleague friend Alan Corman is in the audience who was working at Metarex and then at Bristol Myers Squibb and who oversaw the development of a human antibody that did this and this of course turned out to be very important therapeutic the first checkpoint therapy these are called checkpoint receptors and this shows clinical trial data 10 years out for patients with metastatic melanoma with chemo there was very little help for these people but a fraction of people about 20% would survive essentially permanently with this therapy and indeed if it was later combined with additional similar approaches to block other checkpoint receptors the outcomes have been improved really substantially greater than 50% long-term survival so this really caused a revolution in cancer therapy I think I would argue it's the single biggest cancer therapy breakthrough in I don't know 50 or more years not more and of course Jim was awarded the Nobel Prize for this work but it's also caused a revolution in the field of cancer therapy and in the biotech industry as well so the I should point out that this therapy is also effective against numerous other types of cancer but not all cancers so there's still a huge unmet need for treating cancer patients beyond this so we were studying and I want to go back in time a little bit earlier so we were studying T cell development when this story I'm going to tell you was initiated and we wanted to define how killer T cells develop from immature cells and then become activated during an immune response so this I would say was curiosity driven basic science question and the question we asked was what role does MHC play in T cell development and to do that we wanted to test it in mice useful animal for immunology research with mutations so that they were in MHC components so that they did not have any MHC on their cells so we were fortunate that we collaborated with Rudolf Janisch a colleague of MIT where I was at the time and Janisch was one of the first to successfully generate a so-called knockout mouse where you could take a normal lab mouse and through a feed of genetic engineering and a lot of time and effort you could generate a mutation in a component of MHC for example and then generate a pup that lacked that finally a mouse strain that lacked MHC proteins on all their cells so this was a very important technique and it's been replaced by Jennifer Doudness CRISPR technology more recently which is much easier to do so it took it by storm quickly so we studied these mice in collaboration with the Janisch lab and we showed that MHC deficient mice lack killer T cells altogether so not only is MHC important for presenting antigens but it's important for the development of T cells from immature stem cells and that was an interesting and important finding and we wanted to understand it better and so we did so and of course this led to a chance observation that diverted us and this was worked by a graduate student in the lab, Mark Bix and a postdoctoral fellow, Nancy Leal and they decided to investigate the following question do the T cells themselves need to have functioning MHC genes or do they need to interact with MHC displayed on other cell types and so this was important for various reasons and to ask this question we decided to perform bone marrow transplants from MHC deficient mice to normal mice so this was a way you could have bone marrow from one source and the rest of the animal from another source and so the idea was to prepare bone marrow cells from MHC deficient mice and that has all the stem cells that form all the bullet cells and then you would take a normal mouse that has MHC on all its cells and you treat the mouse to destroy most of its own blood cells and then you give it the transplant to replace the blood cell system with the donor type stem cells and then they would of course develop into mature cells with time and you could then see what happened well we did this experiment and much to our much to our surprise the mice receiving the MHC deficient marrow cells became very ill 7 to 10 days later and we had to euthanize completely unexpected we really didn't expect that to happen so the question was did the what happened here and did the MHC deficient bone marrow cells fail to engraft and the symptoms and the timing of this syndrome were consistent with bone marrow failure where donor bone marrow cells fail to engraft and you ultimately get anemia as a consequence so we thought perhaps the MHC deficient marrow cells were being rejected by the normal mice sort of transplant rejection type response but bone marrow cells the bone marrow cells we were using lacked the MHC molecules and normally you need MHC molecules for graft rejection that's what T cells actually target in graft rejection and also the recipient mice should lack T cells due to the pre-treatments we perform so they don't have any T cells in theory at least so it seemed unlikely that T cells were responsible but there were hints that another type of cell was involved the so-called natural killer cells and I'll tell you more about that in a minute before I introduce natural killer cells in a little bit more detail so what are they? they were discovered in the mid-1970s as cells lymphocytes like T cells present in mice or in humans and they killed tumor cells that was how they were initially defined they were called natural because they were present without previous exposure to the cancer cells as late as 1990 with the understanding of how they recognized cancer cells we had nothing about the receptors they were basically as we say naked cells no known antigen receptors we had no idea how they would recognize any target cell they killed numerous types of tumor cells which you would think would be exciting but actually immunologists at the time didn't like that because they were fascinated at the discriminatory powers of T cells or antibodies and so they considered these cells boring if you will and as a result there was really not that much interest in them and others were actually even skeptical that they existed they thought there was some sort of artifact or some kind of weird T cell or something like that so they were the Rodney Dangerfield of cells and I was raised in that environment where we disrespected natural killer cells and I was not particularly interested in that but we thought maybe they have a role though in this bone marrow rejection phenomenon and to test it we asked whether if you deplete NK cells from the recipients do you reverse the rejection and indeed that was the case so the way we could do this is you can deplete NK cells by injecting specific antibodies and a day later all the NK cells are gone so here's the experiment we can quantify the amount of graft acceptance here and when we transfer normal bone marrow to normal recipients you know very nice graft acceptance but again the MHC deficient marrow was actually completely rejected but if we depleted the NK cells immediately before the transplant the grafts were accepted perfectly well so that made the strong case that NK cells were indeed rejecting these MHC deficient bone marrow cells and suddenly they seemed extremely interesting to us it was claimed that they were nonspecific and did not recognize MHC but clearly MHC did influence the rejection of the bone marrow cells by NK cells so MHC was recognized in a way but the mode of recognition was very different because immune cells were thought to recognize the presence of foreign things right that's how we generally think about it but NK cells somehow detected the absence of a self thing so really they flipped around in terms of what was going on here so how can that even work it was unexpected at the time so the hypothesis arose and by the way we didn't do all this work in isolation there were many people that were working in the field but I don't have time to call out everyone in the field but there was a hypothesis that NK cells have a new class of receptors that bind MHC and the idea was that these receptors instead of activating the NK cell inhibited them and this by the way was before Jim Allison's work so the notion of checkpoint inhibitory receptors was not out there if anything Jim may have been influenced by this kind of work but this concept was completely new that there was inhibitory recognition and that the MHC binding would actually inhibit NK cells and prevent them from killing target cells so this is one way this could work and this was called a missing self recognition where self is self MHC so other groups actually came up with the first good candidates to be these receptors right there was a search for receptors like these and the live 49 and here I won't burden you with the names of these things but they were identified subsequently and they were good candidates but it was worked by a postdoc in the lab Werner Held who's now a professor at the University of Lausanne who that really demonstrated that these function in vivo to inhibit NK function and he did it by introducing one of the receptor genes into a mouse chromosome in a manner that resulted in all NK cells expressing that one receptor and that receptor bound to a specific MHC variant and he found that NK cell killing of target cells with that MHC variant was prevented so this was a genetic demonstration of the receptor inhibited recognition this worked both in cell cultures and in intact mice so this supported this idea of inhibitory recognition so what's the biological purpose of it well killer T cells recognize viral antigens and tumor antigens presented by MHC proteins and this exerts a selective pressure for loss of MHC by mutation for example and you can imagine that happening in a cancer cell mutations in MHC and they no longer express that they become invisible to the T cells and viruses it turns out frequently interfere with MHC expression to try and hide from the T cell component of the immune system in doing so the cells become sensitive to being killed by NK cells and so you can appreciate that NK cell killing of cells like MHC is an evolutionary countermeasure that we've evolved over the years okay so that was really important steps that we and others made in the field to get this model of inhibitory recognition but is that the whole story and we thought it couldn't be we thought NK cells must have activating receptors too because in order to be triggered immune cells generally need some kind of activation receptor and so the hypothesis was there was some unknown activation receptor on NK cells that engaged some kind of molecule on a target cell and just for nomenclature purpose a thing that a receptor binds to is called a ligand so I'll use that term subsequently a ligand is what the receptor binds to so we thought there must be activating receptors and we thought that they must recognize ligands on target cells and we wanted to identify both of those but NK cells don't have T cell antigen receptors and they don't have antibodies so is there something else involved that was the question so here's where great graduate students make a big difference and I've had many great graduate students some of them are here and this one took advantage of what was happening in terms of gene sequencing in the late 1990s before the full genome sequences were done random gene sequences in the late 90s began to be deposited weekly in large numbers into online databases and Russell Lantz who many of you know is our colleague here at UC Berkeley an esteemed colleague in our department at the time was a graduate student in the lab and he searched the database weekly for genes with sequences related to the inhibitory receptor family he was working on he did this on his own volition not on my suggestion and one day up popped this sequence NKG2D and NKG2D was fascinating because it had features suggesting it might be an activating receptor and we really immediately took interest and began to pursue it so we asked ultimately does NKG2D bind to something on tumor cells is it a tumor recognizing receptor and this is how we did that this is the notion of a membrane a plasma membrane of an NK cell and here's the NKG2D receptor anchored in it well it turns out by engineering you can readily generate a version that's truncated and is soluble it's not anchored in the membrane anymore and then by doing a little bit more engineering you could create a what we call a multimer but it was basically just several of them and also it's attached to a fluorescein reagent so it makes it light up and it's fluorescent so this could be a reagent as we call it to stain cells and if it stained cancer cells that would suggest that NKG2D binds to cancer cells so this was such an experiment and this is the machine we run to test this called a flow spectrometer and on this axis is the intensity of the staining and this is just the number of cells in each channel and the main point is that the normal cells had negligible staining with this reagent but the tumor cell line shown here stained very intensely all the cells did essentially so this suggested indeed that this receptor does bind to tumor cells and in fact it bound to 11 of 14 tumor cell lines that we tested and most primary tumors were stained as well so this is very frequently on tumor cells whatever it's recognizing and this immediately made clear that NKG2D sees cancer cells and then of course it becomes very interesting to ask what does it see exactly on the cancer cells and we wanted them to identify that so to quote the Taj Mahal the musician, many fish bite if you've got good bait well we had this staining reagent that's the bait and we could try and use it to pull out the fish which would be the gene that encodes the ligand the thing that the receptor binds to so basically the way that works is that the messenger RNA from the tumor cell is complex there's 10,000 different kinds and messenger RNA encodes proteins and we reasoned that maybe there was a protein that was being recognized and but only one or a few of those 10,000 different kinds would bind to NKG2D so it's a needle in a haystack and you have to find a way to find the one or few that are in there that actually do that and so the way we do this without going through details is what's called expression CDNA cloning and we did that and we succeeded in pulling out some gene so this is a method to use the bait as the NKG2D to pull out the proteins that bind to NKG2D so we pulled out a couple of different proteins one called ray 1 and H60 and again don't worry about the names I'll collectively call them NKG2D ligands what was remarkable was that there's a whole bunch of them ultimately once they were all defined each of us humans has eight different versions and they're distant relatives of the MHC proteins that I've already told you about and they all bind to NKG2D and what we then showed was that one of the genes was introduced into rare tumor cells that lack all of them the tumor cells are now killed by the NK cells so that was a kind of principle and indeed if we put those tumor cells into mice the mice rejected those tumor cells so another line of evidence that this is a NK is a cancer recognizing system came from a graduate student Amanda Jamison who's now a professor at Brown University and what she did was to generate a monoclonal antibody that binds to NKG2D the idea was if NKG2D is an activating receptor involved in cancer cell recognition then an antibody that binds it may block the tumor cell recognition and tumor cell killing so she made a monoclonal antibody she took six tries in two years of spending a lot of time in the tissue culture hood she finally succeeded and she named it Mission Impossible 6 or MI6 and it proved to be a quite useful antibody and this is one example where what we're doing here is asking whether NK cells kill tumor cells so we're adding in a tissue culture well we're adding more and more NK cells from left to right and then we're testing the amount of killing of the tumor cells in the tissue you can see in blue that goes up now you repeat the experiment but in the presence of the antibody and you can see it completely blocked the killing so it doesn't always block this successfully because it turns out that NKG2D is one of several tumor recognizing NK activating receptors but one of the more interesting one okay so that was nice finding currently in this system we could say NKG2D ligands are kill me signals and then the question becomes what turns them on and they are self proteins they're encoded by our own genes so it's not foreign recognition the genes are off in normal cells but on in tumor cells and so what cellular signals turn them on in tumor cells and why became a question now we knew that cancer cells are highly dysregulated in numerous respects and this leads to cellular stress and the idea emerged in the field that maybe stress pathways induce NKG2D ligands and we decided to really begin to explore that so to get a little more detail cells have what are called pathways or mechanisms to counter stress excessive heat will activates the so called heat shock stress pathway unfolded proteins or amino acid starvation activates the integrated stress pathway and damage to the genome to DNA activates the DNA damage response stress pathway and we wondered whether those pathways were involved here so a postdoctoral fellow Stefan Gasser joined the lab and he decided to investigate this and he began to specifically explore the role of the DNA damage response pathway so let me just summarize that pathway briefly what happens here is that there can be errors in DNA replication as cells divide but also there can be damage to DNA from irradiation or DNA damaging chemicals those can cause breaks in the DNA in the genome and these are these lesions these damage is detected by some sensor proteins that are called ATR and ATM so they detect the damage and then they activate a cascade of underlying events phosphorylation events and this leads to two things first of all you stop the cells from dividing so that you can do some work on them and you induce the expression of DNA repair functions which attempts to repair the damage and repaired cells can then go on their merry way but the cells that are too seriously damaged may undergo senescence or program cell death so that is the general features of the work that emerged around the time we were doing this these studies showed that this pathway is often activated in cancer and the reason is that in cancer DNA replication in cancer is unscheduled in effect and errors occur frequently in DNA replication and cause damage and that can activate this pathway constitutively in many cancers it is on all the time in many cancer cells so that became quite interesting that this could be a way to detect cancer and what Stefan showed that indeed this pathway induces expression of NKG2D ligands on cells and therefore killing by NK cells and this was a really interesting finding because it connected this pathway which had been seen really as kind of an intrinsic pathway to mobilization of an immune response for the first time so here's a way you could link cancer to expression of these kill me proteins and Stefan showed this in two different ways one was by starting with normal cells and then damaging the DNA in them and showing they gain NKG2D ligands and that depends on these ATR and ATM proteins or you take cancer cells which express NKG2D ligands all the time and then you inhibit these sensor proteins or you temporarily knock them out or something like that and you sow loss of NKG2D ligands so that together made the case that indeed this pathway is important for expression and so you have this picture then of a normal tissue with a lot of cells and then there's some tumor formation one of these cells goes awry and begins to accumulate and that generates DNA damage stress that causes expression of these ligands and enables NK cells to now try and kill those tumor cells and then work by two subsequent members of the lab graduate student Benjamin Gowen and graduate student Haeyun Yong investigated other kinds of stress and showed that the unfolded protein stress response would induce one of the ligands or hyper proliferation would induce another of the ligands so this seems to be a system to detect different kinds of stress and to enable NK cell killing as a consequence of that okay so then there's a picture emerge then of NK cells with both activating and inhibitory receptors and in stress cells you increase the expression of the stimulatory ligands and that can activate the NK cell to kill but you still have these inhibitory receptors which can either partially or even completely prevent that but then many tumor cells lose MHC1 it turns out and so then that inhibitory signal goes away and you have a killing due to lack of inhibition in this case so it turns out basically that loss of MHC can be sufficient for being killed cells that strongly express these ligands that can be sufficient for being killed and when both things happen the cells are really killed very efficiently they're the best targets for NK cell killing okay so the next question we asked does NKGGD play a role in immune surveillance so the notion that the immune system often stamps out newly arising cancer was first proposed by I believe by Paul Ehrlich in the early 1900s and this idea was wax and waned over the years but by the 1970s a study appeared suggesting the answer was no and it kind of lost favor and for 30 years no one thought much of that idea but then around the year 2000 Robert Schreiber and his colleagues basically resurrected the theory and specifically implicated T-cells in immune surveillance of cancer in some cases so that was an important finding but building on our findings we hypothesized that NKG2D and NK cells play a role in immune surveillance of cancer in some instances and we wanted to test that so Nadia Guerra a postdoctoral fellow joined the lab and she's now a faculty member at the ICRF in London and she decided to investigate this question by disrupting the gene that encodes NKG2D in mice so here again we use the same method I showed earlier where you can disrupt a specific gene and in this case she disrupted the NKG2D gene so these mice she could make a mutant mouse strain that specifically lacks just the NKG2D gene now to study cancer incidents in these mice we would have had to look at probably thousands of mice to get to look at fully spontaneous cancer so the way to get around that the way we do the experiments is to equip the mice with oncotrans genes and these are basically oncogenes cancer causing genes that increase the incidence of specific types of cancer and then to ask the question if we compare the normal mice that have normal NKG2D or they're NKG2D deficient but they have the oncotrans gene in both cases would this lead to more cancer or more severe cancer so you age the mice and you monitor cancer incidence and severity so Nadia did this study and first she investigated a model of prostate cancer and what we found is after 10 months with this model there's a low frequency of massive early arising adenocarcinomas in the mice with normal NKG2D but in the mice that lack NKG2D it was several fold higher incidence so really making the case that NKG2D is suppressing cancer and then she also did it in another system this was a model of B lymphoma and in this case with time the mice with normal NKG2D get all get lymphoma but they get it faster in the NKG2D deficient mice so both of these studies support the idea that NKG2D recognition helps to eliminate some cancers spontaneously but obviously not all of them so if NK cells can kill cancer why do they so often fail that becomes the question and what can we do about it so we think that one reason for failure is likely to be poor initiation of the NK cell response and that's because immune cells generally when in the absence of an infection or a tumor or etc. are in a so-called naive state where they have little functional activity they're kind of minding their own business they really have to be activated to become active killer cells and that wasn't actually thought to be the case for NK cells but we developed evidence that it clearly is the case that NK cells require this sort of event as well and the signals that do this activation are called innate immune signals and they emanate from either the recognition of pathogen molecules or other kinds of events that signal something is wrong and that can result in the production of goodies if you will you know cytokines and other signals that lead to the activation of the NK cell cytokines I should say are basically hormone like proteins that regulate immune responses so we think we began to study this and again it's great that we have such great colleagues here while Russell Dan is now a faculty member and Dan Portnoy both members of our department have done pioneering work in elucidating what's called the Sting Pathway and this is one of these innate immune pathways where the presence of pathogens through a series of events results in the production of what's called a second messenger a small molecule inside cells it's called a cyclic dinucleotide and I'm just going to call it a CDN from now on and those small molecules bind to another protein in the cell called Sting stimulator of interferon genes and that leads to the production of all these goodies and those goodies activate immune responses so that really is kind of how innate immunity works more broadly but the role of this specific cyclic dinucleotides is specific to this particular pathway so that was an important pathway for recognizing pathogens but evidence some of it from our lab showed that in fact this happens in cancer as well and it happens downstream of the DNA damage that I've already told you about that happens in cancer that you can generate signals that activate CDN production and activate this pathway so Asaf Marcus a postdoctoral fellow joined the lab and he showed that that pathway this Sting pathway is essential for NK cell responses to tumors for the spontaneous NK cell response to tumors so that suggested that this really was an operative pathway for activating NK cells in response to cancer but his evidence also suggested that this was a relatively weak and often insufficient pathway to strongly activate the immune response the good news was that there are other ways to activate the pathway and work from Tom Gajewski's lab and also from the group at Aduro Biotech including Sarah McWhorter but also Tom Dubenski and others showed that if you inject cyclic dinucleotides into tumors in sufficient quantities you actually super activate that pathway and you get very powerful activation of the immune response so this is a potential therapeutic approach for cancer then alright but before I show you some data another reason that NK cells fail we believe is the process of desensitization so here is basically years of work we've done that shows that persistent unresolved stimulation leads to NK cell dysfunction so the way you think about that in the context of cancer is you get an active NK cell it goes into a tumor if it can't completely eliminate the tumor it's persistently stimulated and they give up you know they they give up NK cells give up and they're dysfunctional they can't really kill tumor cells anymore well what we were investigating were ways to wake up these NK cells and that by the way it's still a very active area of research in my lab I think there's much more to be done here but Michela Artilino who's a post doc in the lab now a professor at the University of Ottawa decided to investigate whether he could wake them up and he tried to do it in collaboration with Chris Garcia a colleague at Stanford with a so-called superkind so these I mentioned cytokines are like hormone like proteins well Garcia's lab engineered in especially active form of one of these cytokines called interleukin 2 and he calls a superkind and Michela's work showed that this were versus NK cell dysfunction so we could restore some NK cell activity by injecting this superkind into animals so that led to the idea then that we could maybe do both things we could strongly initiate the response with the CDNs we could prevent NK cell desensitization with the superkind it's called H9 MSA or better yet we could combine the two things and hopefully get synergistic effects so these were studies initially carried out by Chris Nicolai and Natalie Wolfe to show that this maybe could work and the idea then was to inject CDNs into tumors and then inject superkinds into the animals and see if we could see therapeutic effects against cancer models and so this turns out works great for NK cells as I'll show you but it works very well for T cells too and we like the idea that we're getting sort of a 1-2 punch from both NK cells and T cells in this approach so here's just this is my last data slide where it could be applied in a very difficult to treat colorectal cancer model MHC deficient tumor cell there's no MHC on it so T cells can't even recognize it and it's very hard to cure but so what Natalie did in this experiment was inject the tumor cells the tumor becomes established and then she applies therapy beginning at day zero and with no therapy the mice are all going to succumb we euthanize them before they do but the tumors are growing with the CDN or with the superkind however there's massive delay in tumor growth so that was good news and in fact we cured about 20% of them in both cases so that was also promising but the most remarkable results were when they were combined where we basically eliminated the tumors and the mice were cured curative 100% taking these mice out for many months and there's no evidence of cancer and this was mediated solely by NK cells T cells played no role in this response so we were quite impressed and happy that NK cells could do such a good job on their own so then Christina Blage investigated an even harder to treat kind of cancer this is a sarcoma cancer induced in mice by carcinogens and sarcomas are an unmet need the existing checkpoint therapies that I told you about are ineffective in most sarcomas and you know so there's not much if they're advancing there's not much to do about them and indeed in this sarcoma model that Christina investigated the tumors would grow and as I said the checkpoint therapy was completely ineffective in even delaying the tumor formation but what was nice was that the CDN in particular had a major effect in delaying the tumors super kind maybe had a small effect but we didn't get any cures here the tumors grew out and all the mice but when we combined them we began to see this again this synergistic effect and about 12% of them lived long term seemed to be cured so that was encouraging and then we decided to say when you we knew that checkpoint therapy is most effective when the immune response is actually ongoing so the idea is we're making an immune response go here so maybe now the checkpoint therapy will help if we add it to the mix here and so we did that and that turned out to be quite remarkable when we combined the three agents we saw 44% cure rate in this model so that was really very encouraging so this you know we're obviously and we're actually exploring many different ways to to generate NK dependent immunity I should point out that this rejection was mediated by both T cells and NK cells so we're getting that you know one-two punch they work together in this case neither alone is sufficient so we're really mobilizing both kinds of immune responses okay so I'll just summarize here and I'm not really summarize but just make some closing remarks about discovery science because I think I think that I think that's a way to generate effective therapies when I was a grad student there was one biotech I believe Genentech and not much coming out of that at the time and developing new treatments was a dream we all had but it was rarely a reality we just didn't know enough and today there are thousands of new therapeutics that we treat myriad disorders we basically take them for granted now and at each step I would say the main driver has been discovery science curiosity driven so that's really the way these discoveries are made we had to learn to have an eye towards applications to ask what does this finding suggest a potential therapeutic approach Jim's example was a good one he asked that question instantly in his work and I think all of us now realize that we can do this now and try to do that but there's many more of these approaches to come probably accelerating dramatically I would imagine and I would point out the most important ones will still be based on curiosity driven research and the reason is we still have so much to learn occasionally you can do anything like we've done enough basic science now it's time to just cure things well that's stupid because there's so much we don't understand there's so much more research to be done and that's where all these new findings will come from and Berkeley is good at this I have to say when you think that I would say the two major therapeutic breakthroughs in the last I don't know how many years came out of Berkeley Jim Allison's work on cancer Jennifer Dowden's work on CRISPR which is going to have many many applications it's really Berkeley's an impressive place and not to speak of the work of everyone else here at Berkeley okay so I'll stop here and just acknowledge again I think I've shouted out all these people and I'd be happy to take any questions thank you very much Dave the the interleukin to the Super Chi I assume that's also promoting proliferation of the NK cells so is there any thought to from any given patient to take out their NK cell culture them so that the total number of NK cells that you could put back into the person to go after the tumor might that might be a strategy or not well there's many efforts now to expanding K cells ex vivo to take them out of patient expand them in tissue culture make them better even engineer them to make them better and then put them back in huge number of efforts like that are underway so yes that approaches is widely being explored not going to be able to hear you Doug so we know the T cells have a lot of all bad side effects because a lot of times you kill the patient trying to treat him with a T cell I'm sorry I don't hear very well too many rock concerts a lot of times the checkpoint therapy kills the patient so in the NK cells what are the sort of bad side to this great therapy you described yeah so there is toxicity associated with checkpoint therapy I don't know about a lot of times I think because I understand clinicians have become better and better at recognizing toxicity and you know backing off when necessary but there may be actually the notion in the field is that NK cells are likely to be much less toxic than T cell therapy but I think that remains to be seen in actual practice but they could apply therapy because they make cytokines that cause you know problems so called cytokine storms but clinicians have just gotten very good at recognizing that and backing off when necessary we have time for one more question hello professor thank you very much for your time and for speaking today I have a lot of questions but I guess like one that I would actually curious to know is that have you ever been in like I know you've been experimenting on mice in your research and there has been ever some times that when you're studying mice that research has failed and at what point when you are studying a certain study that there's enough failures that you just decide to pull the plug and decide to do some other research I'm sorry could you repeat the last part of that sorry I'm not making sense here but yeah at what point if there's enough studies that turned out that experimenting on mice have failed that you decide to just pull the plug on such research at what point does it fail did you say enough time I'm sorry I didn't understand I apologize oh sorry it's my bad I just don't sound systems here are never very good and I don't hear very well sorry I said again how can you judge a failure ah well when nothing happens that would be one one indication I mean it's I mean it's certainly the case that when these drugs are translated into humans they sometimes don't work often don't work actually so that's obviously a key step but of course that's not something we do in a university laboratory that has to be done usually outside and usually in an industry actually so that's a key step that there's often issues with because human cancer is a little different but again one of the reasons one of the ways we try and deal with that models of cancer in mice that are perhaps more similar to human cancer by for example the carcinogen model I mentioned is it really involves initiation of the cancer in a natural environment and that's really more similar to human cancer than some of the other models that people use in mice