0:00 · Opening: immunity, time and persistent threats
0:00 Amit Agrawal
Hello everyone. Before we bring on our guest, I want to talk about immunity and its relationship with time. We usually think of our immune system as a defense system in the body. It wakes up when we get sick. It fights the war and then it goes back to sleep. But if you look deeper, the immune system is actually a historian and a record keeper. Its job is to remember a threat it saw once maybe 50 years ago and protect you from it today. For a long time, we were puzzled by this. We used to believe a simple idea that for the body to remember a virus, traces of that virus had to stay inside us. We didn't think the immune system was smart enough to just remember on its own. We didn't think it could hold a memory for a long time without a reminder. But we were wrong. Biology is much smarter than we thought.
0:53 Amit Agrawal
Then there is the second big question and that brings us back to the same theme time and immunity. What happens if there is a long long war? We know how the body fights a short battle like the flu. You get sick, you fight, you win mostly not always. But what happens when the enemy refuses to leave? What happens in a long war like HIV or cancer? For years doctors looked at patients with these long diseases. They saw that the immune cells had stopped fighting. The cells were there but they were silent. So we called them failure. We thought these cells were impaired. We thought they had died or run out of energy. But again we missed the beauty of it. These cells were not broken. They were just exhausted. Was this is a biological mistake or was this is a survival strategy? Did the body shut these cells down not because it was weak but to prevent the war from destroying the body itself?
1:58 Amit Agrawal
Think about that. If this is not a biological mistake, if the body actually choose to stop the immune cells to fight the war, it changes everything. Then if we release the brake of immunity, will the army wake up? Will it start fighting again? and can it clear the cancer or the virus? So the choice comes should we boost the number of immune cells or to fight the war or should we release the brake on immune cells that are already there. Today we are speaking with the professor who will answer this question. To start the dialogue I am handing it over to Julia. Please welcome Professor Rafi Ahmed.
4:20 · Immune memory after infection or vaccination
4:20 Rafi Ahmed
Thank you. Yeah, thank you. Thank you Julia for the kind introduction and thank you Amit for that wonderful summary. I think you captured the main concepts you know very nicely in your introduction. So again thanks for this opportunity to discuss two areas that we've been very interested in which one is immunological memory after acute infections or vaccination and second is what happens when the antigen persists in the context of chronic infections autoimmunity or cancer. So I'll first start with a few comments about immunological memory. so what is immunological memory? immunological memory really is as Amit pointed out is remembering the pathogens that we encounter. And there's historical data going back to you know thousands of years back that people who were infected once rarely got very severe disease the second time. So this concept was known for many many years. and then as we started understanding our immune system we knew about T cell adaptive immunity innate immunity T and B cells then it became very clear that the major cells that are responsible for protective immunity or immunological memory are really our adaptive system cells B cells and T cells and it was believed the strong belief was at least until the 19 1990s I would say the strong belief was that the way we maintain immunological memory is either by reexposure to the pathogens which certainly does help in boosting
6:15 Rafi Ahmed
responses but more importantly it was believed that in the absence of antigen that memory would disappear. This was a strong belief in the very strong belief in the 1960s and 70s and even in the in the 80s and at one level it made sense because if you get reinfected it was a a boost to the immune system but it didn't fit many of the observations were that were around at that time and I'll start first with the with the antibody issue so it was noted by serological analysis that people who were infected with you know acute viral infections or got live attenuated vaccines as children still had antibody levels detectable decades later. and even under conditions where there was not much reexposure, antibody levels were still detectable. They're of course lower than they were early on but even 10 years, 20 years, 30 years later, you could still have some circulating antibody. so we addressed this issue when I was a faculty member at UCLA in the 1980s using we used the mouse model of LCMV infection and then extended it to other mouse models. And the question we asked was you know how is this antibody maintained when the virus is cleared and we are not reinfecting the mice with the same antigen and what we discovered was that there are plasma cells and plasma cells are the cells that make that secrete
8:12 Rafi Ahmed
antibody that there are plasma cells that are long lived that is can live for the life of the mouse two two and a half years that is you immunize mice or infect mice with an acute infection, you know, at when they're about 2 months old and then you still will find antibody out up to 2 years. And more importantly, we identified these antibody secreting cells that were located in the bone marrow. So there were two major points that came from these early studies from our lab and also a group a group in in in Germany. Manz showed the same thing is that you can have these plasma cells which are generated early on during in the spleen and the lymph node. they will then migrate to the bone marrow and there a subset of them not all of them but some of these plasma cells reside in niches that can provide survival for these cells and just to remind everyone plasma cell is constitutively secreting antibody because you cannot stimulate a plasma cell with antigen because the B-cell receptor i.e the antibody is secreted also these cells don't divide. So basically what you have is a factory producing antibody that's the major function of plasma cells they don't divide more than 50% of their protein actually is antibody think of a factory that's just producing this wonderful molecule that we need for protective immunity and can live for
9:51 Rafi Ahmed
extended periods. Then these studies were extended into humans and then it was shown that people who got smallpox vaccine many many years back still have some plasma cells in the bone marrow. So again this bone marrow niche being critical for longevity generating these long live plasma cells now is well established. So this pre-existing antibody that is coming from plasma cells that were formed many many years ago and are very long lived this is our first line of defense because this antibody is pre-existing in particular if antibody is pre-existing at the mucosal sites okay then it becomes highly protective. So when we come to protective immunity against pathogens not just intracellular but all pathogens including bacterial infections pre-existing antibody at mucosal sites and also pre-existing antibody in the circulation is your first line of defense. in some unusual cases this antibody level is so high that minimal infection happens but in most cases and I'll get to this point later some infection some there will be some viruses that go through or bacteria that go through and infect the cells.
11:14 Rafi Ahmed
Now once you get the infection then there are two additional mechanisms of protection. One are T cells which are specific for the pathogen and T cells can also be very long-lived. again this these are observation that came from our lab and other groups showing that memory T cells essentially can live for decades. after in humans you can find them even 20 years 30 years after the vaccination or an acute live virus infection you can find them you know in the circulation and also some of these are present at at mucosal sites and often these are referred to as tissue resident memory cells that is they are at the sites of entry. So this second line of defense comes from both the circulating memory cells but also but more importantly from these tissue resident memory cells which are at mucosal sites and then the third mechanism of preventing severe disease and more spread of the virus is now recall response that happens very rapidly. So the memory B cells again a very long live population. So antibody secreting cells, plasma cells can be long lived providing that first layer of antibody that you detect. Memory T cells are very long lived. and so are memory B cells. So upon infection then the memory T cells will see the antigen.
12:48 Rafi Ahmed
they will expand elaborate function very quickly and the memory B cells now will provide a very strong recall response. and usually this is higher affinity antibody because it has undergone somatic hypermutation. So the memory B cells will now expand differentiate into new antibody secreting cells. so you get this burst of new antibody coming up. So basically to repeat what I said the first line of defense is pre-existing antibody. The tissue resident memory cells are there if any cells do get infected after several pathogens and viral infections but then you get this big burst of antibody coming up from the memory B cells that which are differentiating now to give you more plasma cells and then also the T cells will increase their numbers. So what I think
13:45 Julia Majewska
yes so when you're saying preexisting antibodies it means that at the beginning they there had to be exposure to the given pathogen right so the
13:57 Rafi Ahmed
yes abs and absolutely in a naive individual if you're naive and have not been exposed to a pathogen there's absolutely no protection at all and that's why whenever a new pathogen comes comes up. This is what happened during the pandemic. we were all naive to SARS-CoV-2 and that's why the infection rates were so much higher. Disease severity was so much higher. When I say pre-existing antibody, this is pre-existing antibody coming from the long live plasma cells that were generated upon the initial exposure. Great.
14:36 Julia Majewska / Rafi Ahmed
Thank you. Yeah.
14:38 Julia Majewska
and another question. So what we what we understand is that there are long lived plasma cells that are in bone marrow and they are constantly secreting protective antibodies that defend us from exposing pathogen.
14:57 Rafi Ahmed
Yes. And I think however yeah let me say something then you can continue. I think to me that is the most critical line of defense that
15:06 · Immune memory and questions about aging
15:06 Rafi Ahmed
we have because these especially if these antibodies are at the mucosal site this this first line of defense is really very very important. Yeah. Go ahead.
15:19 Julia Majewska
How does it change in aging? Because in aging bone marrow changes and it's getting more fatty. there's my there's production of a given cell types that is enhanced and simply there's depletion of lymphocytes B cells there's there's it's it's getting very crowded
15:42 Julia Majewska / Rafi Ahmed
this is what I want to say yeah
15:44 Julia Majewska
so if there's if it's crowded and there's no space one could think like how plasma cells could go and stay there to have their niche and to be these factories that are producing constantly that are giving us the defense long-lasting defense. So does it work in this way that when we are young and we are exposed to pathogens we get plasma cells that home to bone marrow niche and they are giving us long-lasting protection. However, if we are getting exposed to pathogens in later stages of our life, it won't be as much efficient or could you comment on this?
16:25 Rafi Ahmed
Yeah, those are those are excellent questions. I'll break it into two parts because I think you asked two important questions which I think are very distinct. So one is let's say you were given a vaccine as a child. Okay, you got the measles vaccine or any of the live attenuated very successful vaccines you get or you were vaccinated or infected as an adult for so for example you got the yellow fever vaccine which is very good in terms of long-lived immunity a live attenuated vaccine or you got one of our more successful vaccines like HPV vaccine or the hepatitis B vaccine so all of these vaccines will induce you know a reasonably good number of these long live plasma cells and then there'll be a subset that will migrate to the bone marrow and be there. So there's data showing that people who were infected at as children or as young adults or even adults okay not aged but adults 30 years 40 they have very long lived antibody responses now I want to emphasize that it's not antibody coming up and staying flat no antibody titers decline significantly over the first few years and they have to because otherwise we would all have hypergammaglobulinemia and you don't want that. So antibody levels have to be maintained by homeostatic mechanism. so antibody titers do decline significantly over the first few years but then they reach this plateau.
18:11 Rafi Ahmed
So most of the plasma cells we generate are short-lived because you need them immediately. You don't want to become a bag of plasma cells. So you have to eliminate most of the antibody secreting cells or plasma blast that were generated early on but then there's a subset of these long live plasma cells that then can live really for the life of the individual up to 40 50 years. Okay. So in that instance the response was generated in a younger individual was not an older person. Okay. so the response happened while there wasn't any immune senescence or aging. But now you ask an important question in that context which is you know what happens in terms of space are you is there still space in the bone marrow and that is an interesting question that I think still people are debating is there an upper limit is there overcrowding especially as an individual ages there will be more and more plasma cells and since as you pointed out the composition of the bone bone marrow also changes during aging.
19:25 Rafi Ahmed
So these are important questions for f for future studies. This this is an area that has not been explored in great detail. There were some suggestions about competition early on. but I think those were not very definitive studies. there was it was thought that if you generate new plasma cells you displace the earlier ones. that those data may be correct but again they're not definitive and so I think that's probably unlikely that's you make one you displace another one I don't think that happens but what you raised is as an individual ages and the bone marrow environment is changing then what happens I think that's a great question for future studies you know if that has some some adverse effect okay but then the Second question you asked is really different which is now generating responses in the elderly. That's different because there is we're not just talking about crowding of the bone marrow but we're also talking about the overall immune system in a 90 year old or a 85 year old person. Okay, as opposed to somebody who's much younger. I think you know more about aging than I do. But when I talk to aging experts studying the immune system, they usually put the cut off for immune aging above 70. Okay. And maybe even older than that. most of them don't feel that major immune aging happens in somebody who's 50 years or 60 years. but certainly we're now living
21:06 Rafi Ahmed
longer. So I think that issue of u generating new responses denovo responses in terms of both from naive cells and also recall in the in really the aged is as you know an area of intense investigation is the and one of the to me one of the important questions to address and I'm not aware of very definitive studies on that is I should make it qualify that our own lab does not work on the aging question. So this is my information that I gathered you know talking to my friends who study immune aging. so the one question is you know is it's very well established that a aged individual respond less efficiently than a younger adult. but is the main problem here due to the T cells and B cells or is it really the innate system that is are your DC's not as efficient are your innate cytokine production which is very critical for generating efficient responses is that less efficient so I think this distinction between poor responses in the elderly how much of that is due to defects in T cells and B cells as opposed to how much of that defect is in the innate response. I think is another area that requires you know major investigation.
22:40 Julia Majewska
Right. Definitely is very exciting. We for now what you just said we concluded that there are two arms b cell mediated defense and T cell mediated defense against pathogens they play they complete each other for defense against
23:05 Rafi Ahmed
they work beautifully in synergy so there's great synergy and these arguments about are T cells more important or B cells more important I think are really kind of juvenile. They both do distinct things and they both work in concert. Both are equally important.
23:23 Amit Agrawal
I see. So Rafi, if let's say there is no T-memory cells, so B cells will not be able to compensate the deficiency of T memory cells.
23:33 Rafi Ahmed
the there are data on that people who have T cell deficiencies are much more susceptible to viral infections but there is human data because of the genetic some of the genetic deficiencies in T cells also you know you know with the HIV which unfortunately was depleting the T cell compartment again there were issues there with infections and also there's a lot of data even more data in people with B cell deficiencies which are more common than the T cell deficiencies and those individuals have again problems with infections in general but more so with bacteria infection. So I think there some distinction could be made that the T cells would be I'm talking really more in particular about the CD8 T cells but also CD4s because you need them for helping the CD8s and you need them for helping the B cells that with intracell once it comes to intracellular pathogens I think you do need T cells to really efficiently combat that infection and that's what I'm saying that this It's very rare that you totally prevent infection.
24:56 Rafi Ahmed
And I'm going to get to that point. This is a good transition to get to my next
25:00 · Protection from disease versus infection
25:00 Rafi Ahmed
point. Great. So the so the next question really is okay, we have I've talked about the longevity of T and B cell immune memory and how it can protect us you know for decades. but the question then is are we getting complete protection from infection and I've already alluded to this fact or are we really getting protection from disease or more severe disease and this is not a trivial question because it's also has some public health in addition to the scientific issues it also has some public health issues in terms of informing the public you When we develop a vaccine and we want to convince the people to take that vaccine how do we market that vaccine scientifically? Do we say this vaccine is preventing infection or do we say that this vaccine will be resulting in much severe much less disease that is you will not get very sick and part of this is academic but part of this also is the messaging. So if you look at the at the available data in humans and this has been generated for many many years. many systemic infections like measles yellow fever take any systemic infection you get very good protection from from from disease very good protection. And the assumption is that these vaccines and are very good that you have very long live protective immunity against these systemic viral infections. I give you examples of measles, yellow fever, small pox there
26:53 Rafi Ahmed
are many many examples. Polio polio yes many many examples polio is a great example and many others. But then when you come to more what are u mucosal infections and in particular the respiratory infections we all get the common cold over and over again. We all get influenza many times. pretty much all of us even though we've been vaccinated and so on have probably had SARS-CoV-2 more than once. Very few people have only had it once. Fortunately the symptoms were not that severe. So it's so there is this thing this human data showing that there is long live protective immunity based on any disease with the systemic infections but there isn't much long-term that is you do get reinfection with many of these respiratory infections and also with some of the intestinal infections you see that with rotavirus you can get it more than once of course so I think that here one my feeling my potential explanation for this is is really based on how do you get the disease you know what causes disease what causes symptoms so basically our test for vaccines and infections is really the clinical readout nobody's going inside our body and saying did we have some viral infection okay basically is all based on clinical symptoms. So let me now illustrate an example. and I could have chosen many others but I chose influenza but in influenza I could have chosen any respiratory vir
28:47 Rafi Ahmed
2 to put in there and as a systemic infection I put there measles. Okay to just to illustrate that but again that would apply also to smallpox and many other viruses. because many viruses that cause systemic infections are also taken into this respiratory tract. So let's look at what happens after influenza infection and after measles infection. So after influenza infection if you get some infection of the you have pre-existing antibody but it's not completely preventing every virus from infecting the upper respiratory tract. Okay. some will some viruses will go through that pre-existing antibody level especially if it's on the lower side and it usually is very rarely is every viral particle coming in the that all viral particles coming in fully neutralized. Okay. So then if the infection is just in the upper respiratory tract you likely will have some very mild symptoms but you will have some symptoms. So that's the key that some symptoms will happen because the disease is caused by the initial set of cells which are infected. Now if immunity was really low in the in those individuals or they were elderly you know had some some of their immunity was kind of compromised due to a variety of things then the virus may spread to the lower respiratory tract and if it goes into the lung then you're very sick you could even die.
30:22 Rafi Ahmed
but so here we think of this as not being protected. Okay. But I what but what I want to point out is that the real difference between influenza and measles is not that one immunity is there and the other one is not there. It's how the disease is caused. So with influenza really everything happening up here. Now measles also gets in through the aerosol. it come it goes through the respiratory tract and but there's no respiratory disease. The virus really isn't infecting large number of cells in the respiratory tract or in the lung to cause any of those diseases. So how does disease happen there? And this is true for pretty much all systemic infections. The virus will first infect target cells that it can infect and then those target cells will produce a huge amount of virus. You get viremia and then you seed the secondary sites which cause the disease. So if you read classic papers on viral pathogenesis work done by Frank Fenner and Macfarlane Burnet in particularly Frank Fenner they talked about primary and secondary viremia. So basically the initial infection with any of these systemic viruses you don't pick up viremia initially it has to replicate in several s whatever cell it infects then there's a burst of virus coming out then it in a few days you will then have virus in the in the blood that's how then the infections spread. So this gives an incredible window and the window usually
32:09 Rafi Ahmed
is 4 days, 5 days, sometimes maybe a week but there's this window of at least 4 days to a week where the immune system can then kick in and prevent the spread of the infection. You gave the example of polio and that's a great example. I should have I in this case the virus is infecting the gut okay is infecting and any symptoms any the killing of cells within the intestine within the gut by polio virus which happens very often okay doesn't cause any severe symptoms the children are not that sick okay so that kind of pretty much goes unnoticed okay it's not recorded as as something bad with polio. Okay. But if there is high viremia from that then the virus will go to the central nervous system infect the motor neurons and then you have the terrible poliomyelitis and the data are that in of the if you have 500 kids infected only about one or two. So the typical ratio is about one in 300 to 500 kids will actually get poliomyelitis.
33:24 Rafi Ahmed
So again there is that wonderful window of protection. So then the question was with measles and the reason I put here measles is because there was a very nice study that was done in 1997 by Michael McChesney who's the first author of this paper and this was a study in non-human primates in monkeys. So there was actually a outbreak of measles in the primate center at UC Davis and this and the virus that was isolated from these from these non-human primates at that at the UC Davis primate center was very pathogenic for monkeys. Measles in general all wild type measles will infect monkeys anyway. this monk strain was very pathogenic in in monkeys and caused symptoms that were very similar to what one sees in humans. So this was a beautiful animal model where a monkey adapted measles strain really was causing disease just like measles does in naive individuals. so the experiment in this paper was they vaccinated the monkeys with the licensed measles vaccine. The measles vaccine that's protection that's protecting millions and hundreds of millions of people of children who get this vaccine from getting measles. Again very rare vaccinated individual to get you know clinical measles. So what they found was that so they had two groups of monkeys. One group of monkeys was naive and was infected with this very virulent measles strain and the other one did not get the vaccine. So
35:23 Rafi Ahmed
vaccinated with the licensed measles vaccine unvaccinated and they got what we see in humans that is beautiful protection. All of the monkeys that got the vaccine showed no pox, didn't show any of the spots. They did not get fever. They did not die. So, so that part was reproduced exactly that we see in human. Beautiful protection by the measles's vaccine. But here they could do more. What was done in this was they then analyzed the measles virus load in different tissues and they found that in monkeys that were vaccinated and showed no clinical symptoms there was still replication of the measles virus. So, so the vaccine did not result in what the term that is used often sterilizing immunity that is there was some infection but certainly no no spread not having any disease symptoms. So I think we should again in terms of how we sell our vaccines, how we market them, we should be a little bit careful and not say that this vaccine prevents infection.
36:36 Rafi Ahmed
what we should be saying is that this vaccine is going to prevent disease and either you will have no disease or you'll have very mild disease. I think that this is a important distinction both in terms of how vaccines work but also in terms of how we message and how what we tell the public about how vaccines protect us from from severe disease and death.
37:03 Julia Majewska
But the question comes But the question comes so if we are if even once we are vaccinated and we have the protection vaccine gives us protection but still infection can happen but no disease onset the I understand that the capacity of neutralizing antibodies cannot sterilize it cannot it doesn't give us complete 100% % sterilizing immunity which would just kill the neutralize the virus as it enters as we are exposed to so it can still infect cells but the disease onset is not happening so what about the approaches which would enhance T-C cell mediated immunity against infected cells so that you know
38:00 Rafi Ahmed
Right. Yeah. So I mean I that's a good question. So it that's why things get controlled very quickly because of the memory T cells that we have. Okay. pre-existing memory T cells which are there and can very quickly elaborate effector function. The reason the virus doesn't spread and cause disease is because the T cells are coming in very quickly eliminating those virally infected cells. So that elimination of the few cells that get infected really is predominantly by the T cell compartment. So that's why I said there's beautiful synergy in how things work.
38:40 Julia Majewska
So in case of viral infections respiratory for example viral infections influenza where we see the onset of the disease. So this is because there is not enough mucosal immunity which would protection locally. Yeah. I think we have we have to
39:02 Rafi Ahmed
different yeah that high levels you know and also it depends something I didn't mention but it's very obvious. It also depends on the exposure. You know, it's like what was the initial MOI of infection? You know, did you get infected with 50 particles or did you get infected with 500,000 particles? If you get infected with 50 particles, it's very likely that all 50 of those could be neutralized. Okay. But if you get infected with 10 times more, it's more likely that some of them will not get neutralized immediately or fagocy taken up but will actually infect the cell. So I think that there are some lot of nuances there. It's not extreme in one direction goes to the other. But the concept that no matter what the exposure is and when you're how far you're out that you'll always get no infection, I think that is not correct. Now the getting infected has another very important implication in terms of controlling an infection or controlling a disease because disease is spread by transmission.
40:15 Rafi Ahmed
So a so that positive thing still happens because transmission is more when you have more vir. Now in the case of someone who's vaccinated and is getting very mild symptoms or no symptoms but some infection those people are very unlikely to be transmitting much virus either.
40:41 Julia Majewska
So, very important for public health.
40:43 Rafi Ahmed
Exactly. Very important public health. So, so when I'm saying that sterilizing immunity is is more of a myth, reduce transmission after vaccination is not a myth. That indeed does does happen. So, vaccination will reduce transmission because the amount of virus that's there or the organ that there will be much less. So, vaccines do prevent or decrease transmission.
41:10 Amit Agrawal
But does it work 100% in all the humans who got vaccinated for?
41:15 Rafi Ahmed
No, nothing is ever well some vaccines are pretty close to 100%. You know but and again that depends a lot also on the on the pathogen and something I forgot to mention is that of course with many of these respiratory viruses you're also getting variance. So when the variants come up then of course that pre-existing antibody protection dramatically decreases and we see that all the time when you have a pandemic flu strain that emerges major changes in the hemagglutinin so protective immunity now is less and we saw this dramatically with the omicron in the SARS-CoV-2 pandemic you know the original Wuhan strain mutated into beta and so on but then we saw this big upurge with delta and That was interesting. Delta was not that immunologically different from Wuhan. So people so Wuhan induced antibody did neutralize delta quite well but delta was just much more pathogenic. So you get you got a lot of disease. Then Omicron emerged that was a dramatic mutation in the spike protein very most of the antibodies that were induced by Wuhan strain and also by delta were very ineffective in neutralizing Omicron and that's why you got this huge surge of infections in individuals who already were immunized vaccinated or infected previously but I still think that the disease severity was lower because we had this pre-existing antibody and the T cells if the so again the when you are looking at this protection from this
42:55 Rafi Ahmed
pre-existing antibody if a virus is changing dramatically then it has an impact on the ability to protect
43:04 Julia Majewska
but what about strategies to target conserved part of the virus of
43:09 Rafi Ahmed
and that of course everyone is trying to do that raising to the conserved part with influenza there's been major efforts the influenza hemagglutinin you know the stem has the head and the stem region the stem region is more conserved so there have been a lot of efforts and targeting vaccines to generate more stem specific antibodies but the head is the most critical part to neutralize because the head is what is binding to the receptor Okay. so I think we do need also regions within the head which are a little bit more conserved. So I think just a stem antibody is not the final answer. I think having more of the stem antibody will be helpful but I think we also need antibody to the more conserved regions of the head and of course with HIV this is the biggest the broadly neutralizing antibodies this is the biggest effort with you know and lot of very elegant and interesting work being done to kind of get more of those being had
47:27 · Stem-like cells and their cellular niches
47:27 Rafi Ahmed
This is a busy slide but I will walk you through it. Okay. So the major discovery that was made in 2016 by us and by others working in this field showed that there is a very unique cell population that really is critical for maintaining this ongoing immune response. And these cells I will in this slide are the red cells and I'm calling them stem-like CD8 T cells. these cells are also referred to as precursors of exhausted CD8 T cells. So you'll see the term Tpex used in many in many papers and publications. So basically what I'm calling stem-like CD8 T cells are in the Tpex cells. But what these cells do is they are essential cells for maintaining an ongoing immune response. So even though the virus is not eliminated, the cancer is not fully controlled and cleared. there is some functionality within these three subsets. It's not a dead system.
48:43 Rafi Ahmed
it is still trying to do its best to control the infection failing in most cases but still trying to do its best and everything any response that present under these chronic infections is dependent on these stem-like cells. So they are really the resource cell. This is where all responses are being generated from. And when did the stem-like cell arise? will arise very early on when the infection first started and came from a naive cell. but now in an ongoing chronic infection or in cancer there are three major subsets of cells you can find. So the red cell are the stem-like cells and there are many unique properties of these cells. The perhaps the most interesting property of these cells is that they is their location. So these cells are found predominantly if not exclusively within the lymphoid system. So in mouse models that we have studied chronic infection mouse models you can find these cells in the spleen. You find them in the lymph nodes. You occasionally find some in the lung and some other non-lymphoid tissues.
50:01 Rafi Ahmed
But the major but in most non-lymphoid tissues you don't find them. They are present in the spleen and in the lymph nodes. that's their main location and then within the spleen and lymph nodes they found in the T cell zones so and they're not circulating so basically they are resident cells within the T cell zones of lymphoid tissues when you go in human cancer you find them in in three different locations you find them in the draining lymph node You find them in tertiary lymphoid structures and within tertiary lymphoid structures they found closer to the T cell areas and not to the B-cell areas. So these will be closer in TLS have both B and T cells. These cells will be with closer to the T cell area and then they're also found in the tumor stroma. So in the tumor you have the real malignant cells the tumor cells but you also have tumor stroma. Okay. So these cells are in the tumor stroma. and you find them and this is very nice work that Nir Hacohen at the broad institute has done looking at lung cancer and he's identified what he calls the immunity hubs. So this is where the stem-like cells are in these immunity hubs and they're surrounded by myeloid cells. Again I we talked about niche for plasma cells in the bone marrow. There's also a very fascinating niche for these stem-like or the tpex cells within the tumors and during chronic infection. So
51:45 Julia Majewska
these stem-like T cells when they are found in the tumor tissue will it be only hot tumors which have infiltration of immune in immune cells?
52:00 Rafi Ahmed
Yes. I mean this term hot and cold tumor is not that absolute. Okay. So but whenever you can find you have TILs in most tumors you will find TILs. So whenever you find TILs, the TILs will contain these three subsets. There'll be some cells that will be stem-like transcriptionally and epigenetically. Some will be these the transitory effector cells. Some will be the most exhausted cells and the stem-like cells will be in the stroma area whereas the more differentiated cells will be where the tumor is. And we showed the same thing in the mouse model. that is the stem-like cells are in the T cell zones where there isn't too much virally infected cells but when you come to where the virally viral infection is it's the more differentiated effector cells so there's a very nice segregation in terms of the anatomic location the stem cells are kind of think of it they're sitting in a bunker okay they're kind of protected from excessive are over stimulation by the you know by where the infected cells are and they're surrounded by myeloid cells in kind of in a wonderful niche that's kind of keeping them quiet and keeping them resident there and preserving their pluripotency
53:26 Julia Majewska
standing the current understanding is that it is good to have stem like CD8 cells if we the more we have them within the tumor sta of the tumor it's the better the potential outcome.
53:42 Rafi Ahmed
Yeah, absolutely. There's data showing that I'll get to that in a second, but there's data showing that if if you have more of these pretty potent stem-like cells then you respond better to PD-1 blockade.
53:58 Julia Majewska
And secondly because because you mentioned that they u they are in this niche stru with myeloid next to myeloid cells or specialist their special occasions close to myeloid cells. So what do myeloid cells they protect them in which way?
54:20 Rafi Ahmed
Yeah there are yeah there are very interesting interaction. I let me come back to that. Okay. I'll come back to that in you know in my last slide. Okay. but let me finish this one. So what are the stem-like cells doing? Okay. so as I told you they are in these niches both in human tumors and in chronic infection mouse models where they were first defined. and I'd say a few comments about their transcription program and then tell you what they're doing continuously. Okay. So of course these cells express PD-1 they express TOX they are getting antigen signals PD-1 comes up if there's antigen signals are there so they are u expressing PD-1 TOX but then they express also TCF-1 this is a transcription factor that's essential for longevity and survival of T cells all memory cells have TCF-1 most, not all, naive cells have TCF-1 but in addition to having TCF-1 and PD-1 and TOX they also express BCL6 which is quite unique for this cell population.
55:30 Rafi Ahmed
So it has a bit of a TFH flavor and this was one of the most unexpected findings that we made is that they had BCL6 they express CXCR5. Okay. even though the expression they're not really going to B-cell area the germinal centers but the main reason I think the BCL6 is there is to really keep them is it's part of their program where these cells don't have any effector molecules because as you know BCL6 is kind of an anti-effector repressor of many effectors so these cells don't express granzyme B they don't express perforin so here you have a CD8 T cells that's not expressing the killer molecules but it does express co- stimulatory molecules and inhibitory receptors other than PD-1 are very low like TIM-3 all of these molecules are not there and they are very pluripotent so they do two things one is they can undergo a slow self renewal to maintain their numbers so this population these cells are very long live in our mouse studies we can find them even in a mouse that's about 2 years old and was initially infected when they were about 2 months old. So they can maintain a both self renewal and quiet in a very long lip cell. But what they're always doing continuously even without any therapy is they're generating and differentiating and to give you these transitory cells. So the first step in this pathway is downregulation of TCF-1. Okay. Upregulation of TBET. So T-BET comes up very high now. Okay.
57:12 Rafi Ahmed
These cells now start proliferating. Now they express effector molecules and they express exit molecule like S1PR S1PR5. They come out into the blood CX3CR1. So they really effector cells and these cells have functionality. The number often is very low and I'll get to that point in a second. But you always will be getting some transitory cells. These cells will be where the tumor is, these cells will be where the viral infection is. But then since the tumor is not completely clearing, the viral infection is not completely clearing because you don't have enough of these blue cells, they will then become terminally differentiated and exhausted. Also the neither the blue cell nor the green cell is very long lived. The only long-lived cell here is the red cell. It's generating the transitory effectors. They're doing something but not completely doing what's needed to eliminate the chronic infection or the tumor. So you end up getting many more of the green cells which have less less functionality can't proliferate. U so that's the transition that's happening.
58:24 Julia Majewska
So, so it means that red cells then like T cells are absolutely
58:32 Rafi Ahmed
they do they are essential. they're essential because and we have shown this in our in fact we showed this in our first paper in 2016 the Im et al. paper that if you don't generate the cell and you can do that by if you start with cells with a TCF-1 deficient you still generate the effector cells initially you don't generate the stem-like cells and then the whole system crashes within in 2 to 3 weeks there are no cells remaining okay because you didn't generate this cell population. So that's where you set.
59:07 Amit Agrawal
So Rafi, is it possible to transplant these cells in cancer patients?
59:13 Rafi Ahmed
Well, I think you'd have to get it from that same individual which will not be that easy. So I think that but I'll get to in my last slide how we might have strategies for expanding their numbers. Okay. So this is what happens when you do PD-1 blockade. the same slide but now I'm showing you with a different title showing you that the proliferative burst that you get and this has been shown in animal models of infections chronic infections and tumors but most importantly in humans. So if you get PD-1 blockade you get a burst of proliferative cells that come out into the blood. It has been shown by us in lung cancer other groups in melanoma. so this is well established that PD-1 blockade will result in a burst of effector cells that come out. All of that is coming from the stem cells. So when you remove the PD-1 brake, if this transition going from stem to transitory was happening at a rate of 1x, you will now have significantly more that can be 2x or 3x or 5x depending upon the context of the tumor and context of other criteria that are regulating this other mechanism regulating it. But you will always get this increase differentiation. you get many more transitory cells and it's having more of those blue cells that results in the reduction in the tumor and reduction of the viral load.
1:00:48 Julia Majewska
So we were talking about exhaustion of T cells. We said that T cells express exhaustion molecules PD-1 and here we see that stem like T cells which are quiescent low in number also do they do they respond to PD-1
because they express very high levels of PD-1. Yeah, they are also very tough. PD-1 when it comes to CD8 T cells PD-1 is the central regulator of of controlling the T cell response, mediating exhaustion and regulating function of all three of these cells.
1:01:32 Julia Majewska
So is is it right to say that in this context PD-1 PD-1 on stem cells CD stem like CD8 T cells it doesn't mark exhaust the T cells but it's controlling their function
that's very well put yeah it's still they still retain their pluripotency because of their transcriptional program and their epigenetic program u and the niche that they're in okay they retain and then the transcription factors that they have it's a very key cell whose function actually some similarities to hematopoietic stem cells even in some of the genes that they express. In fact, in that first paper, we showed that some the transcription program of these chronic stem-like cells has some genes that it shares even with hematopoietic stem cells which as you know are also quiescent and also are in HSC. Yeah. So it's again to go back to PD-1.
1:02:39 · Why the PD-1 pathway matters
PD-1 you asked a question why do we have PD-1?
1:02:44 Julia Majewska
Yes. Why do we have PD-1 on stem?
Why do we have forget about stem cells, why do we have PD-1? Because and then you ask the question, how many people are there? How many people have been reported to have genetic deficiencies in PD-1? Hardly any. You will hardly find any human being who has deficiencies in PD-1. a very few number of people were identified and this is Casanova's work the paper that he published a few years back and these unfortunate individuals just lived for a few years and then they died due to infections and immunopathology and if you these are experiments that we did and we also reported this in our original manuscript that we published in 2006 Dan Barber's paper where we described PD-1 as the major regulator of T cell exhaustion. Okay. if you have PD-1 deficient mice they are fine as they mature the immune system develops they are quite healthy. If it's on a autoimmune background then you get increased autoimmunity but normal B6 mice bal mice live happily ever after even without PD living in our clean colonies. Okay, they have no no problem. Okay, but if you infect these mice with the chronic LCMV strain, the one that disseminates and infects many tissues which is what we use for chronic infection, the LCMV Clone 13 strain. Can you guess what happens if you infect a PD-1 deficient mouse with the Clone 13 strain?
1:04:41 Julia Majewska
I guess nothing good there.
All the mice are dead within a week and we reported that in the original paper. In that one we had used PD-L1 knockout mice but the same thing you see with PD-1 knockout
1:04:56 Julia Majewska
because because you don't have modulator of immune responses.
No, no, you do you. No, it because PD-1 only comes up after the antigen stimulation. So if you don't have PD-1, the immunopathology is very excessive. And we showed that if you infect the PD-1 knockout mice with the chronic strain, they all die within 8 to 10 days. But if you deplete CD8 T cells from those mice, they live and become chronically infected. So it's the immunopathology. So the PD-1 evolved to prevent excessive immunopathology. It's something that's in the literature and known but people haven't really appreciated it. But then you know it is doing wonderful things in regulating responses keeping all of these cells still alive especially the stem-like cells are kind of maintained under these conditions. And again antigen signaling is needed for maintaining this stem-like phenotype or the stem-like properties and then we can exploit this by doing PD-1 blockade where you get much more effector cells and get control and as you know there's always some immunopathology associated with these with the with the beneficial things that happen. So
1:06:14 Julia Majewska
Rafi in the context of aging we recently showed with Amit that in the context of aging you have accumulation or increased load of damage increased number of senescent cells and this is associated with the exhaustion of immune responses PD-1 CD8 T cells and and we we thought that this is I mean this is the way because you have chronic damage and that is also associated with chronic lowgrade inflammation. Yeah. Yeah. and there is [unclear] exhaustion that probably is needed to some extent to prevent constant stimulation and destructive effect of over stimulation
points and I think we had discussed earlier in our earlier call that this issue of T cell senescence and T cell aging It's really quite fascinating and beautiful work has been done on that. But I think would be also important to do a more detailed comparison of T cell exhaustion that has happening in younger people you know this has nothing to do with aging. You have chronic infection. You will get this phenot you know you'll get cells of this phenotype and transcriptional program. And then how do these cells compare with truly aged and senescent cells? I think that's a very important area of future investigation and this repeats what I told you basically you have on PD-1 blockade you have many more of the blue cells which will have killing function and show some clinical benefit and so now that we have this and it gets to the immunotherapy issues that you were that you asked earlier Julia so now that we have this understanding of these three distinct subsets and I think pretty much any everyone working in the field of chronic infection or cancer and also autoimmunity agree that there are these three distinct subsets. Of course there are some differences depending upon the model and the disease but these three cells have been found by almost anyone studying chronic infections or cancer. So the so if we now design so
1:08:52 · Stem-like CD8 T cells and immunotherapy
this provides a potential platform for designing more rational immunotherapy. So what can we do to something that Amit brought up okay earlier what can we do to increase the number of stem-like cells can we and we have preliminary data that if you inject chronically infected mice with IL-7 and this was a collaboration with a company called NeoImmuneTech. they have a long acting IL-7 molecule called NT-I7. we see a selective expansion of the stem-like cells. This is not surprising in some ways because of these three subsets the only cell that express CD127 which is what IL-7 is binding are the stem-like cells. Okay. the other two cells really don't have CD127 but it provides a way of of course you also expand other cells because any cell that has CD127 will expand there could be approaches there could be targeted delivery if one wanted to develop that is have IL-7 going only to cells which express PD-1 in which case it would be very specific for the stem-like population within these three subsets but what was more interesting was we found that this IL-7 treatment of course it's very high dose of IL-7 you know way above physiological levels that not only did it increase the numbers of these cells and their self renewal but it also mobilized them. So we saw a from they being totally quiet and resident in the T cell zones some of these cells
came out there were some additional changes in the gene expression program that brought them out into the blood but still retained the TCF-1 all the other properties so I think there could be so some approaches could be tried for increasing the stem-like cells and if they're safe I think it's something worth considering because the more stem-like cells you have I think the better you will be in terms of keeping this response ongoing and more importantly responding to immunotherapy. in any regimen that we have used for expanding cells the expansion comes almost exclusively from the stem-like population the so I think this is potential one target in terms of immunotherapy and people haven't used that so far the second is can you improve the quality of the effector cells that is can you get can the blue cells be even more functional and here actually we have very nice data and this is published and that is if you combine PD-1 blockade with IL-2 then the this differentiation trajectory becomes even more efficient that is the red cells are still the cells that respond the stem-like cells but now they generate qualitatively much better effector cells cells which are longer lived effector cells that have better effector function in fact look more like effector cells you get during an acute infection or a good vaccine and these data were published a few years back and we also collaborated with Roche where we had PD-1 bispecific, PD-1–IL-2
you know targeting these cells so I think that this is a that's IL-2 I think is the key cytokine for this and of course there's a lot of interest in IL-2 many many many clinical trials have been done most have failed or have shown too much toxicity and we also show in our paper that this IL-2 needs to engage CD25. So many of the clinical trials were done with with IL-2 that was mutated not to bind the alpha chain the CD25 and not to bind the trimeric receptor. We show in this paper that in under those conditions the IL-2 effect is very suboptimal. So I think there is now more interest in this field after our papers came out the people are now making what they call alpha bias. So this is an exciting field. It's an old field in terms of IL-2. The classic studies of Steve Rosenberg were very successful even with IL-2 monotherapy but there was a lot of toxicity and so people kind of put it away and focused on decreasing toxicity which of course is important but I think there's a new phase now for IL-2 that you need to still conserve binding to the trimeric receptor. I think that appreciation has come now and there could be other cytokines, other ways of improving it. and probably the toughest question is the last one. What can you do with these truly terminally differentiated cells which have minimal to no proliferative capacity? We and
others have thrown everything we could find cytokines all kinds of things on these cells and they really have lost all proliferative capacity. that one may require some epigenetic changes because I think these cells really are at the dead end in terms of any proliferative capacity. They're also short-lived. and then I want to end on this last point here which is I think what you brought up Julia about the myeloid cells. So I feel that a critical area that has not received as much attention and should in the future a major attention is T cell myeloid cell interactions and this would be particularly important in terms of the functionality of the stem-like cells and their ability to differentiate because the myeloid cell compartment is critical for maintaining this niche. but also in general myeloid cells in the context of cancer, myeloid cells in the context of chronic infection are not very functional especially the dendritic cells in cancer dendritic cells in chronic infection are not very functional. So I think targeting the myeloid compartment so that they become more functional and also how they're interacting with the how they're interacting with stem like to me this is the most exciting new frontier in terms of cancer immunotherapy is really the right the focusing on myeloid cell T interaction.
That was my last point last slide.
1:16:03 Julia Majewska
What do you think like in cancer immunotherapies or for diagnostics is it used in clinics to look at stem cell like T cells?
1:16:17 · Clinical measurement and the tumor environment
Yeah, that will not be easy because it can be done and that's how the people showed that there's better responsiveness to because a tumor biopsy was taken you know pre treatment. so I think that's happening could be done more but I think more important is to do these immunotherapies to improve the function because knowing that is is very good to know that but I think then you need to do something
1:16:54 Julia Majewska
I mean is it because some big part of them are in lymph nodes and then you don't like maybe it's less successful.
No. Yeah. I mean the other thing which I didn't touch upon at all is so in the context of viral infections it's easy we know that these are specific for the viral antigen right I mean that's so that part is very known and easy to study but in the context of tumors basically these would be to the neoantigens the mutated regions and there I think is the more little bit more challenging there because as you know some tumors what you call cold right probably don't have enough mutations or the mutations they have did not really induce good T cell responses so that's a more difficult challenge you know and what I talked about today are really situations where you already have enough T cells and then this concept of therapeutic vaccination comes in which is quite active now and that I think is another way of generating new stem-like cells. So if you have the right antigen in your vaccine and it's now presented by better APC again I go back to APC okay you will be you should be able to generate new stem-like cells which were not there before i.e e convert what potentially is a cold tumor into a hot tumor by generating new stem-like cells using the appropriate epitopes in the [unclear].
1:18:42 Julia Majewska
the role of the system or of the environment. If we have lots of damage, chronic chronic settings, even though we try to push the red cells then like T cells into affect our T cells to fight to defend us from whatever it is. We we have all the settings that are not beneficial. meaning you know like damage and constant stress and challenge. So how to how to cope with it?
Yeah. I mean that's a very active area as you know I mean the tumor micro environment is very immunosuppressive. Yeah. And we see the same thing in the chronic LCMV model. Very very immunosuppressive environment. You have a model which is a lifelong chronic infection model. You have a lot of TGF beta is playing a major role in inhibiting. We have published on that in our mouse model and of course a lot of studies in humans. So TGF beta is a major player in that immunosuppressive environment work done by several people IL-10 playing a role and then you know the role of fibroblasts the more fibroblasts and then this is very well documented clinically that if the tumor is very fibrous then outcomes are much worse. outcomes are much worse not only in terms of their survival and prognosis but also outcomes not that favorable with the PD-1 blocking. So I think that the TME and again within that TME comes in the issue of how effective are the dendritic cells. I think all of that is within u this thing of the immunosuppressive activities what the tumor is making you know now recent work by [name unclear] and others have shown that the tumor metabolism is also playing a role so it's not just the Tregs but also the tumor metabolism but the tumor is secreting and what it's doing is is is immunosuppressive so the TME of course is the center of all of these
1:20:57 Julia Majewska
So many things have to come together but step by step we are
I think I think the progress has been phenomenal you know and yeah I think this is a you know one of the most exciting areas of biomedical sciences and lot of outstanding people working on these these issues.
