KENDALL A. SMITH: Today I'm going to tell you about a new theory of how the immune system works that I've developed over the last several years. And how it explains, in molecular terms for the first time, why autoimmunity develops.
A while ago, my grandchildren came to visit me and we went to Broadway here in New York and saw The Lion King. And The Lion King had this line in one of the opening scenes from that play where both the Lion King and his son Simba are standing on a bluff overlooking their domain. And the Lion King said, everything exists in a delicate balance and it's our job to understand it. And I said to myself as I listened to that, that this guy must be an immunologist.
Now the immune system, it has been known from the beginning of time in immunology, which only dates back 00 years. To be entirely regulated by outside forces, the antigens in the environment, and which I've called the primacy of antigens. Because the idea has been is that the immune response on the part of the host is entirely antigen regulated.
And that means that the immune system is absolutely quiescent until and if an external, environmental antigen is introduced in the form of a virus or a bacteria, or even a pollen. And when that happens, then there are cells within the immune system that recognize the antigen and react to it.
And then ultimately, the reaction leads to the ability of the immune system, the entire system, to clear the antigen from the body. And when there is no more driving force on the system, it returns to quiescence and becomes in steady state again.
Now if you look back in the history of immunology where that idea came from, it really is traced back to Macfarlane Burnet. It was first published in the Australian Journal of Science in 1957, where he for the first time enunciated what became the clonal selection theory of the immune system.
And it's only a three or four page paper. And if you haven't read this and if you're interested in the history of immunology, I recommend it. It was published in Nature Immunology in 2007 as a part of the 50-year anniversary of Burnet. And in this three or four page paper, there's no data, there's no figures. He just simply opens his mind as to how he thinks the immune system is regulated.
And he said several things that ultimately over the last 50 years has come true. First, he postulated that when an antigen contacts an antibody, that takes place on the surface of a lymphocyte. And the cell is then activated to undergo proliferation, to undergo a variety of descendants.
And until that time, what lymphocytes did in the body was entirely unknown. There were inklings over the decade prior to this, that perhaps they had something to do with immunity. They knew that cells that were plasma cells were the sources of antibodies, but they didn't know that lymphocytes became plasma cells.
So he was revolutionary in that he focused his theory on the lymphocyte. And he also, the fact that he said that these selected cells and proliferate to produce a variety of descendants, was very, very important. He went on to say that the increase in the number of lymphocytes and the clone selected will ensure that the response to a subsequent entry of the same antigen will be extensive and rapid. And that is a secondary immune response or an anamnestic immune response.
And so that gave a cellular basis to the phenomenon of memory. He went on to say then that any clones of cells which carry reactive sites corresponding to body determinants will be eliminated. Now he didn't say how that would happen and this subsequently became called the [INAUDIBLE]. In other words, you don't react to your own body.
Now what Burnet did not anticipate was the following. First, he did not anticipate there would be such things as interleukins, which are molecules that are molecular regulators or hormones that regulate the whole immune system. Secondly, he didn't really deal with what would happen if some of these auto-reactive cells escaped the deletion that he talked about. And thirdly, he didn't say, well, what would happen if antigens persists for a long time in the host and it's not clear.
And so you ask yourself, so what kinds of antigens might persist? For sure autoantigens, self molecules are never going to go away. And tumor antigens, because they arise in cells of the host are not going to go away. Alloantigens, or the antigens that are present on allografts, like a transplanted kidney, they don't go away. And lastly, persistent infections, for example, HIV and hepatitis C virus, and tuberculosis where the immune system fails to clear the infection. Those antigens persist.
Now the premise that we started with is that although the immune response is initiated by the antigen, its tempo, its magnitude, and its duration are regulated internally by leukocytotrophic hormones. And internally regulated is the way all the rest of the organ systems that we know of in the body are regulated. What we're seeing here by this premise is that the immune system is like every other system within the body. And there is such a thing as a leukocytotrophic hormone and this is also known as interleukins or cytokines, the same words that have been used to describe these molecules.
So if you're thinking about deriving a new therapy that would follow on from Burnet's clonal selection theory, because absolutely everything, as I said, in his initial paper, discourse, has absolutely come true. But any new theory has got to explain how the entire system works. And it has to be able to be reduced to the level of an individual cell to discriminate about how each cell reacts. And furthermore, it has to be reduced from the cellular level to the molecular level to understand how the critical molecules function that determine how the cell reacts, which then determines how the system reacts.
And finally, this age old question of the self, what immunologists call the self, non-self discrimination which prevents a reaction against self. Any new theory has got to explain how that happens and it has to explain it at the molecular level. Now as I was preparing this lecture, one of the things I was reading in a new biography by Albert Einstein, and I came across this statement of his. A theory is more impressive the greater the simplicity of its premises, the more different things it relates, and the more expanded its area of applicability.
So I thought about that for a while and I said to myself, you set the bar pretty high in terms of what a theory is supposed to be. But I formulated the quantal theory of immunity. And first published it in 2004 and then another review in '06, and another review in '08. And then I published a full book about this just last year in 2010.
And the quantal theory essentially states the following. Number one, the systemic immune response is dependent on the number of cellular clones that respond and the extent of their proliferative expansion. And if you think about it, this is simply a restatement of Burnet's clonal selection hypothesis or theory.
But it goes beyond that because it states that individual cells must receive a critical number of triggered receptors which the cells have an ability to count. And when the critical or crucial number of receptors have been triggered, the cells respond in an all or none fashion. And all or none, the word for it in cell biology is quantal.
And this quantal decision or response decides the fate of these cells. And they're basically only three different kinds of fates that the cells can undergo. First, they can become activated. They then will proliferate and differentiate into effector cells. And then ultimately they can persist as long-lived memory cells within the body.
The second thing is energy. And rather than becoming activated, the cells become paralyzed. And those cells, have recently in the past several years, been suggested. And there have been suggestions that these cells, once they become energy, they become suppressor cells or regulatory cells. And I will talk about that later on in this lecture.
The final fate is cell death, or apoptosis, which is programmed cell death. Now as I said, most people didn't really know what lymphocytes were for or what they did in the first half of the 20th century. And it wasn't until this paper came out in 1960 in cancer research from a young physician at the University of Pennsylvania, Peter Nowell, who accidentally discovered that a plant lectin that he was using, which is called phytohemagglutinin or abbreviated PHA. He was using it to unglutinate red cells because he was a cytogeneticist and he wanted to get his hands on leukocytes, white blood cells that were uncontaminated by red blood cells.
And he left his white blood cell rich plasma in the incubator over a weekend. And he came in on Monday morning and he found that no longer were they very small lymphocytes. Most of the cells had transformed into large lymphoblasts. They looked like the lymphoblasts of a leukemic patient. And some of the cells were undergoing mitosis. At first he thought he discovered the cause of leukemia, but he hadn't. And what he had discovered was is that phytohemagglutinin or PHA would cause the cells to proliferate for several days.
Now this is where I entered the picture and it wasn't until almost 20 years later in 1977. Because what we had found was, or other people had found, that if you used conditioned medium from lymphocytes, you could get cells to start to proliferate. And I wanted to find out whether I could get antigen-specific cells to proliferate. And I wanted the antigen to be a tumor antigen.
And the very first experiments worked so that we were able to generate the long-term culture of tumor-specific cytotoxic T cells that we published in Nature on Bastille Day in July of 1977. And this is just a picture of some of our original cultures. And you can see that there are large blastoid cells, some mitoses. There were chock-full of granules and ultimately it was found that these are the cytotoxic granules that contain the proteins that enable the cells to kill their target cells.
And this is against the dogma, because the dogma in immunology was that antigen was the thing that caused the cells to proliferate. Not in a lymphokine or cytokine or what we call the T cell growth factor was in the lymphocyte condition medium. So the question immediately became, what is causing these cells to proliferate? Is it the antigen or is it the T cell growth factor?
And that became the question that we focused on. And I felt in order to answer this question, we had to do strictly reductionist scientists. And we, science, we needed to have monoclonal T cells. We needed to make monoclonal antibodies to the growth factor. We needed to purify the growth factor to homogeneity so that we can be sure what we were dealing with.
We needed to identify the receptors for the growth factor, so we had develop an assay for the TCGF. And we needed to get monoclonal antibodies that would react to the receptors so that we can quantify the receptor at an individual cellular level using the flow cytometer. And this is just our first paper that we published on the clones, of monoclonal cytotoxic T cells in the Journal of Experimental Medicine in January of 1979.
And what we had done was is that we had mixed tumor lymphocyte cultures where the tumor cells were allogeneic or a different histocompatibility than the lymphocytes. And so ultimately derived cells that we could keep growing in culture, in our conditioned medium, that would lyse allogeneic tumors. But they would also lyse identical syngeneic tumors that were from the same tumor virus.
And this is a cytotoxicity assay where we have the percent of cytotoxicity with an increasing killer target cell ratio. And as you increase that ratio, you get more and more and more killing of the cells. And so in order to find out whether or not we had individual cells, we cloned these cells in individual wells of microtiter plates. And then tested the clones for cytotoxic reactivity, either to the allogeneic or to the syngeneic tumor. And this clone is only reacting with allogeneic cells, whereas this clone is reacting to only syngeneic tumor cells. This clone reacts to both and this clone reacts with neither one of them.
So we could prove that what we had then was the asexual progeny of a single cell. Now the next thing that we did was is we took these cloned cells and devised an assay for the growth factor, which we then used to purify the growth factor so that we could immunize mice and see if we could get monoclonal antibodies. And this paper was published in 1983, "The Production and Characterization of Monoclonal Antibodies to Human Interleukin We had renamed the growth factor, the T cell growth factor, interleukin-2 by this time.
And what we had done was, we'd taken our monoclonal antibodies purified in a very small immunoaffinity column. Purified liters and liters of conditioned medium into a single molecule that we could elude off the column. And then we've analyzed it here by HPLC. There was a single peak on a reverse phase, HPLC showing that we probably only had one protein. We had one band on an SDS page. And we had one, a single internal amino acid sequence that we were able to discern from our purified growth factor. So we could be able to state then that we had homogeneous purified T cell growth factor in interleukin-2.
The next thing we did was to radiolabel amino acids to the cultures of the cells that were making the growth factor. And here we used it, S35 methionine, labeled our growth factor. And then we purified it and showed that it was only one radiolabeled protein on the gel of around 14 or 15, or 15 and a half kilodaltons. And once we only had one radiolabeled species, we could do binding assays on receptor positive cells. And this shows the saturation binding curve on the three different cell populations.
Where we're talking about the bound, we have plotted here the bound molecules per cell against the concentration, the free concentration of the radiolabeled growth factor. And if you then take the bound over the free ratio and plot that on the y-axis against the number of molecules bound by the cells, you generate straight line plots. And where the data crosses the x-axis or is a direct read out of the number of cells, number of receptors per cell. And the slope of the line allows you to calculate the equilibrium association constant, or the infinity constant.
And what we found, as we went on and characterized these receptors, that these interleukin-2 receptors have all the characteristics of true hormone science. And that is, number one, a high affinity. And the equilibrium dissociation constant was only 10 Picomolar, or 10 to the minus 11th molar. There are a finite number of receptors, and that's why they were saturable. If you added more and more and more hormone, you reached a point where you plateaued and didn't get any more binding.
There was stereospecificity in that these receptors would only bond interleukin-2 and all the other factors and cytokines that we'd try to compete for the radiolabeled IL-2 bonding were negative. We found that the concentrations of the radiolabeled IL-2 that bound to the cell were identical to the concentrations that promoted the proliferation of the cells, so that we had a physiologic basis for the receptor.
And when we found all of these characteristics, I went to my colleague Allen Monk, who was an endocrinologist and really the person who taught me how to do hormone binding assays. And I said, Allen, we've got a hormone, right? And he looked at me and he said-- he was 20 years my senior at that stage. He said, no, no, no, Kendall. He says, not until or unless you can show that there's negative feedback regulation to the system, do you really have a hormone.
Well, one of the things that we wanted to do away was to use our IL-2 receptor binding assay to find a monoclonal antibody. And in order to do that, we collaborated with Tom Walden's group at the National Cancer Institute because they contacted us after they saw our receptor paper. And said they thought they had an antibody. And so I said, well, send up the antibody and I'll do a competition assay.
And you can see in this plot right here, which was one of the very first experiments that I ran, as we added increasing amounts of antibody, this is the percent inhibition of the radiolabel TCGF binding. We get more and more and more inhibition of the radiolabeled ligand binding.
So that was the first antibody that had ever been devised to react to a cytokine receptor. And one of the first things we did with that antibody was to use it in the flow cytometer. And Doreen Cantrell, who was a postdoctoral student from Nottingham in the UK, had come to the lab. And she did a series of experiments that we published in 1984 in Science that laid out the ground rules in terms of how the IL-2 receptor interaction actually causes these cells to proliferate.
And this is just a plot of a flow cytometry profile of increasing concentrations of fluorescent labeled antibody. And as you can see, in the cell number on the y-axis, you can see as we increase the concentration of antibody, you get more and more antibody bound. Until down to here, there's only very few cells that bind very high concentrations of antibody.
Now we knew at that time that if we did dose curve of using IL-2 receptor positive T cell clones and using non-labeled interleukin-2 that we would see a symmetrically sigmoid dose response curve. And that is, is that in very low concentrations of ligand, only a very few cells would get very low, incorporation of tritiated thymidine, which was our routine way of measuring proliferation at the time.
And as we increased the concentration of the ligand, we got more and more thymidine incorporation into the population of cells to the point where we finally got 100%. Now what we didn't know at the time was that if you look at the 50% point, for example, does that mean that at the 50% point of thymidine incorporation that all of the cells that have incorporated only 50% of the thymidine that they are able to? Or does it rather mean that 50% of the cells have left the G0, G1 phase and have gone into the S phase to synthesize DNA?
And so we took advantage of the flow cytometer and quantitative staining, the flow cytometer allowing us to look at individual cells and quantitative staining with propidium iodide. And here's a PI stain of cells at the very lowest concentration where the vast majority only have a single peak of DNA. And then here at the top here, we have two peaks of DNA of a G1 peak and G2 peak here with the inbetween are the S phase cells.
So that told us that at the individual cell level, say at the 50% point here, 50% of the cells have left G1 and started to synthesize DNA, whereas 50% of them have not. So that then led to the question, well, what's the difference? We thought that this is a cloned cell population. Why should there be any difference? They should all incorporate thymidine at the same time.
The most logical explanation of that was that here, where we have very high densities of receptors. Probably then, those are the cells that are responding to very low concentrations of ligand. And as we increase the ligand concentration, we're essentially marching back on this curve to the point where when we have low receptor densities, we need a higher ligand concentration. And then we get all of the cells to proliferate.
Well, in order to test that hypothesis, we took advantage of the fluorescence-activated cell sorter. And Doreen took the cell population-- here is the parent cell population. And this is a linear scale rather than a logarithmic scale, so it's skewed and some cells here are off scale.
She separated that into the low density receptor population and the high density receptor population. And synchronized the cells, put them back into saturated concentrations of interleukin-2, and then did frequent pulses over time thereafter. And what she found was that the low receptor density population took longer to proliferate, as measured by thymidine. Then the high density population, which was this one, and the unseparated population was inbetween.
And it was also something about the timing of this that told us that what we were looking at was really an arithmetic manner in which the cells were detecting the presence of the triggered ligand. And one of the things we did very soon thereafter was to look at the effect of interleukin-2 on this phase of the cell cycle, the G1 progression phase.
And in order to do this, Julie Stern, a post-doc in the lab, put lymphocytes in the culture. And they're small, round, resting lymphocytes. She triggered the T cell receptor here at minus 12 hours. And then added IL-2 at zero time, zero hours. And then looked at five hours, at 10 hours, and 20 hours.
And you can see at five hours, some of these cells have started to become large blastoid cells, which is clearly evident by 10 hours. And then at 20 hours we can actually even see mitosis. So if you don't add interleukin-2 at time zero, all of the cells in these time points stayed as round small lymphocytes.
And so we concluded that the T cell receptor was really causing the cells to become competent, in cell cycle terms, to the growth factor, which was giving them the progression factor. Around the same time, Doreen and I collaborated with Ellis Reinherz, who's a good friend down at the Dana-Farber in Boston, who had used our method to generate the first human T cell clones. And then used the human T cell clones to generate mouse monoclonal antibodies reacted to the T cell antigen receptor.
And Ellis's group in 1983 were the first to describe the T cell antigen receptor. So with his unique reagents, both cells as well as monoclonal antibodies to the T cell receptor, and our cells had monoclonal antibodies, we collaborated to show-- it's depicted in this figure, last figure from the paper.
Whether you use antigen presenting cells and antigen, or whether you use the clone-specific, T cell receptor-specific antibodies, you can activate the cells. And when you do activate the cells, what they start to do is produce the interleukin-2, which is the small spheres here in this graphic. And at the same time, this stimulation here causes the cells to now express, newly express, T cell growth factor, interleukin-2 receptors. And it's the IL-2 receptor interaction that actually causes the mitosis and the production of two different daughter cells.
So a pictorial representation of this is shown in this slide. The T cell receptor interacts with the T cell antigen receptor on small lymphocytes. And that antigen receptor interaction causes the cells to start to produce interleukin-2 in the green squares here, and also express IL-2 receptors. And the IL-2 receptor interaction then causes the blastic transformation as Julie Stern had seen. And at the same time, causes the replication or the duplication of the DNA. And ultimately then, to undergo mitosis and create two daughter cells.
So in cell cycle terms, the G0 to G1 competence is given to the cells by the T cell receptor. Whereas the G1 progression part of this system is donated or mediated by the IL-2 receptor interaction, to the point where it progresses past the restriction point where you no longer need the IL-2 receptor interaction anymore. You'll undergo DNA replication and then subsequently mitosis.
The same thing happens to other hematopoietic cells. And down here what I've shown is the hematopoietic stem cell ultimately becoming competent to the granulocyte colony stimulating factor. And it's this cytokine receptor interaction that does exactly the same thing in the myeloid series that the T cell growth factor IL-2 does in the lymphoid series.
And the fact that the receptors that are identifying their different ligands cause different things to happen in the cell. The competence caused in the T cell system by the T cell receptor activates a series of enzymes, of kinases, that activate now at least three families of transcription factors that then migrate to the nucleus.
And there are specific sets of genes that are turned on by the T cell receptor, the importance of which, for a cell cycle progression, is interleukin-2 and interleukin-2 receptors. Now by comparison, the structure of the cytokine receptors is different from the structure of the antigen receptors. The IL-2 receptors are composed of three different chains that are called alpha, beta, and gamma.
And when you add the ligand to cells that have these receptors, it forms a very tight quaternary molecular complex that brings together two kinases of the Janus family or the JAK family of kinases. JAK1 and JAK3, which phosphorylates the beta chain of the IL-2 receptor, which then serves as a docking site for the signal transducer and activator of transcription 5, STAT5.
And once STAT5 docks onto the beta chain, it becomes phosphorylated by the JAKs, undergoes a conformational change, translocates to the nucleus, and activates a different set of genes from what the T cell receptor had activated. The most important of which, for cell cycle progression, I think are the Cyclin D genes and also survival genes that keep the cells alive, Bcl-xL.
Exactly the same thing happens in the hematopoietic system with granulocyte and arithroid and platelet derived growth factors, except for the fact that they only have one chain rather than three chains in the receptor. It's complex to only one kind of JAK. It happens to be the JAK2 molecule, but it does the same thing, phosphorylates cytoplasmic domain of the receptor, which serves as a docking point for STAT5. And exactly the same thing happens when the same genes are turned on.
Now this is very important from the standpoint of the molecular aspects of what drives cell cycle progression, because it's really the progression factors that do this rather than the competence factors or the T cell receptor. So what we're saying is that T cell clonal expansion, the molecular explanation of that depends on the affinity of the the receptor interaction, which is very high. It depends on the concentration of the ligand. And therefore, we have to know about IL-2 production and persistence. It depends on the number of receptors per cell, the duration of their interaction.
And also because of the way the system seemed to have worked, the number of triggered receptors accumulate, which the cells mysteriously are able to count. And that's the molecular definition of Arthur Pardee's restriction point or our point. And that's the molecular explanation of the G1, S phase transition.
So if T cell clonal expansion depends on the concentration of interleukin-2, what determines the interleukin-2 concentration? That's one question. And another question is desire to regulate itself. So we did a series of experiments early on that told us that interleukin-2, once it was bound to the receptor, it's internalized and degraded by IL-2 receptor positive cells.
And this is from the earliest paper that we published on the growth factor assay in 1978. Once we had the assay, we wondered what would happen if we seeded a very low concentration of cells in culture. This is 6,000 cells per milliliter in the open bars with what we thought was a high concentration of the growth factor, one unit per mL.
And then every day then we would take a sample of the supernate and assay, its concentration, and also count the cells. And as you can see over six or seven days in the culture, the cell number gradually progressively increased to the point it went up 100 fold to 600,000 cells per milliliter. But during that time, the concentration of the growth factor progressively decreased. Now if you leave this culture at six days, one more day without adding new growth factor to the cells, they will all die overnight. And so we didn't know exactly what was going on, but this was the first inkling that the cells were consuming the growth factor.
Now over the course of a decade, from 1982 when we first collaborated with Tom Waldman's group, up until 1992, a series of papers were published by ourselves and others that defined the three different chains of IL-2 receptor. And this is a flow cytometry profile from our lab where we have radiolabeled antibodies to the alpha chain, the beta chain, and the gamma chain. And we've simply taken human T cells and activated them so that they become IL-2 receptor positive.
And what you see here, this is cell number against the fluorescence intensity. And this is a logarithmic scale. And now what you see is the log normal presentation of the alpha chain receptor. There are some cells with very low numbers of receptors, but some cells with a high density of alpha chain. And as you go here, most of them are distributed about the mean.
But in the same profile, you see of the beta and the gamma chain, what is very striking in this plot is that there's about almost 100-fold increase of alpha chains, as compared to the beta and the gamma chain. And the beta and the gamma chain looked to be identical in terms of their density on the cell surface. And one of the questions was, why is that? Why do we have this discrepancy of number? And does it make any difference if you have a high density versus a low density in terms of any of these things that we're dealing with.
Well, here's one of the earliest experiments where we took radiolabeled IL-2 and bound it onto the surface of the cells, washed off the excess, and then put the cells at 37 degrees. And over time took samples to see what happened to the radiolabeled IL-2 molecules. And if they were bound only to cells that had an alpha chain, they stayed very stable over time, over the next 30 minutes.
Whereas if we had the trimeric high affinity receptor, there was a progressive loss of surface bound IL-2, radiolabeled IL-2, and it was going inside the cell. The half time for this process was only 15 minutes. And so it was a very rapid internalization process. And that had been shown for the ETF receptor, the insulin receptor, and other receptors at that time. So that wasn't too surprising to us.
You know some things, I tell my students that anything's possible in science. But some things take a very long time. We tried to crystallize IL-2 bound to its receptors way back in the early 80s. But it wasn't until 2005 when Chris Wilson and then in 2006, he and Wilson and his students and postdocs were able to co-crystallize IL-2 bound to the three chains of the receptor.
And here you see IL-2 bound only on the alpha chain. And you can see one face in the IL-2 molecule. It binds on the alpha chain face. And here it is over here. But now we've got the beta chain in blue and the gamma chain in yellow. And when IL-2 is first bound onto the alpha chain, it very rapidly brings in the beta chain, which then brings in the gamma chain so that you make a quaternary complex. And that's the complex that signals the cell.
And once that happens, once you make that stable quaternary complex, the next thing that happens is that you start signaling into the internal molecules. And this is a flow cytometry plot, a Phospho-STAT staining against cell count, cell number over here over time. So in the gray here is at time zero. You add interleukin-2, you see at one minute, at five minutes, at 15 minutes, and at 60 minutes in the yellow here on the same normal distribution of Phospho-STAT molecules within the cells. But it's a very rapid process and it became a new assay for the ability to assay for interleukin-2.
But what we found is that the efficacy of IL-2 binding, signaling, and consumption are determined by the IL-2 receptor densities, which isn't too surprising either. Recently in collaboration with Greg [INAUDIBLE] across the street at Memorial Sloan Kettering, and [INAUDIBLE] who is one of his postdocs, they found that the density of the alpha chain becomes extremely important. Because as it varies, it varies the affinity of the trimeric receptor to bind IL-2 and to signal and to consume.
And here is just a plot from their paper that was published last year of cells that have only 100 alpha chains right here, versus cells that have 100,000 alpha chains here. And this is the response of the cells as determined by Phospho-STAT5. And you can see that there's a thousandfold difference in the IL-2 concentration curve when there's a thousandfold difference in the number of alpha chain receptors. And so when you go from a Kd of 10 to the minus 10 to one that's 10 to the minus 13, so you're down here in the femtomolar range, 100 femtomolar when you have a very high density of alpha chains. And that becomes extremely important for a lot of reasons, as we'll get into subsequently.
Well, what we found with another series of experiments was that IL-2 production is transient because there are negative feedback, classic negative feedbacks that Allen Monk wanted me to find way back. that IL-2 inhibits IL-2 gene expression. So now we have a system whereby we can say finally that this is a hormonal system.
The transient nature of IL-2 production that we found in our very, very first experiments is now known to be due to these three different negative feedback loops, one that works at the level of the TCR signaling. So once you start signaling the cell via the T cell antigen receptor, IL-2 is produced. IL-2 then causes the production of another molecule that's called CTLA-4, that's an IL-2 induced gene. And CTLA-4 feeds back and stops T cell receptors signaling.
The other thing that happens is that IL-2 induces inhibitory cytokines that the cells make. So in addition to the fact that it makes the cells and gives them a positive signal, it causes them to produce inhibitory cytokines. Interleukin-10 is one of the most important. That feeds back and stops these cells from responding to the interleukin-2.
And finally, there is a transcription factor, FOXP3, which is an IL-2 induced gene which feeds back and blocks the T cell receptor activation of the IL-2 gene. So you're clamping down in three different molecular mechanisms on the ability of the cells to make interleukin-2. And this is a series of experiments that we did where we tested the hypothesis that we knew and other people had reported that the T cell receptor induced FOXP3, which is a transcription factor. And it induced FOXP3 expression.
And our hypothesis was is that was part of an IL-2 dependent negative feedback loop that limits the immune responsiveness by restricting IL-2 gene expression. And our first paper came out in 2008 and reviewed immunological reviews. And then I treated that extensively in the book on the quantal theory and immunity, if you're interested.
And this is just one figure from our initial paper where we could see that FOXP3 expression, which is listed here on the y-axis. So as we increase the fluorescence intensity here, we've got increasing amounts of FOXP3 within cells. It doesn't suppress IL-2 production, which we're measuring here, intracellular IL-2. To Because we've got a fair number of FOXP3 negative cells that can express interleukin-2. But the FOXP3 positive cells, for the most part, do not. Only very few of these cells are capable of making IL-2.
So given the fact that FOXP3 shuts down IL-2 production, T regulatory cells that have become FOXP3 positive, they can suppress effector T cells by primarily [INAUDIBLE], due to interleukin-2 receptor mediated IL-2 consumption. So if FOXP3 positive cells cannot produce IL-2, but they can bind it, internalize it, and degrade it, and because IL-2 activates FOXP3 expression and FOXP3 restricts IL-2 expression, this is a classic IL-2 dependent negative feedback regulation.
So the question is what happens if IL-2 stopped being made and being consumed when it becomes depleted. And there was a very nice paper done by Pushpa Pandiyan and Michael Lenardo. It was published in Nature Immunology in 2007, where basically they showed that the regulatory T cells, the FOXP3 positive cells, actually kill effector T cells.
Here is a phase contrast microscopy of the effector T cells alone. Here is the effector T cells with control non-regulatory, non-FOXP3 positive cells. And they're all live lymphocytes. And here they are if you co-culture them with regulatory T cells. The blue cells are cells that have had their membrane damaged. I guess now they're taking up the vital dye of Trypan Blue.
So the premise is that the immune system regulates responsiveness to antigen by controlling IL-2 production via negative feedback loops. And that's an extremely important concept. This is a concept that the endocrinologists would find no problem with, but most of the immunologists have never taken any courses in endocrinology.
There's a whole series of papers that were published in the early '90s that discovered that if you have an IL-2 deficiency, you get an IL-2 deficiency autoimmune syndrome. This was first described after knocking out the IL-2 gene out of mice by Ivan Horak over in Germany. And he showed that IL-2 knockout mice as well as then subsequently, others showed that IL-2 receptor knockout mice. They're both immunocompromised and they're are autoimmune, which is sort of an oxymoron.
The immunodeficiency is easily seen because mice that are IL-2 knockouts or receptor knockouts cannot develop memory to infections. And so if you expose mice to influenza virus, they get over it, the initial infection, primarily from their innate immune system. But then if you rechallenge those mice with the same virus, they have no memory and they will succumb. And we've done that ourselves in our own laboratory, and others have shown it with other microbes.
The autoimmune aspect of this IL-2 deficiency syndrome is manifest by lymphoproliferative infiltration of non-lymphoid organs, many of the organs of the mouse, over time. And this develops over several months, ultimately become infiltrated with lymphocytes, activated T cells. And the mice die prematurely due to inflammatory bowel disease and autoimmune hemolytic anemia.
At a similar time that the IL-2 knockout mouse had been found to have this weird immunodeficiency and autoimmunity, there was known to be an X-linked immune dysregulatory polyendocrinopathy syndrome, or IPEX, abbreviated IPEX, that had been found in humans. And in 2001, people showed that if you mutated FOXP3 or knocked out FOXP3 in mice, that you would get this IPEX syndrome in both mice and in men. Then in 2003, people showed that the IL-2 knockout mice had a deficiency of FOXP3 positive cells from IL-2 knockouts, from IL-2 receptor alpha chain knockouts, as well as beta chain knockouts.
And so that seemed to provide the molecular explanation of why the IL-2 knockout mice got this weird autoimmune syndrome. However, 1/3 of the IPEX patients had no FOXP3 mutations. And so therefore, it got to be a little bit more complicated. In 2007, there was a human, an infant that was identified that had homozygous mutations in the IL-2 receptor alpha chain gene. And that young infant developed an IPEX-like syndrome. It was first published in the current opinion in Rheumatology in 2003 and the Journal of Allergy and Clincal Immunology in 2007.
And the homozygous IL-2 receptor alpha chain deficiency, there's an absolute absence of the IL-2 receptor alpha chain positive cells, paradoxical coexistence of immunodeficiency and autoimmunity. And here's what happened to this poor child. The first symptoms were found at six weeks of age, severe diarrhea with a loss of cytomegalic virus in the stool, the onset of insulin-dependent diabetes, the onset of an acute respiratory distress syndrome with viral pneumonia due to CMV, and chronic CMV viremia.
Then at two years, the child developed severe eczema and had diffused lymphadenopathy that was filled with cells that were EBV virus positive and hepatosplenomegaly. At three years, they developed Hashimoto's thyroiditis and became hypothyroid. Started making antibodies to the red blood cells, autoimmune hemolytic anemia.
And then at five years, antibodies were found to their neutrophils, so that they had antibody-positive neutropenia. Persistent sign, bacterial infections is a consequence with sinusitis and otitis media. At eight years, there was asthma and recurrent pneumonias, persistent eczema, lymphadenopathy. Inflammatory bowel disease then came upon him with protein-losing enteropathy. Multiple different kinds of immunologic treatments we used, steroids, IV immunoglobulin, rituximab, cyclosporin, and so forth. And nothing really seemed to do much good.
You would have thought that this child would not have any FOXP3 positive cells, but this is CB4 against FOXP3. This is the control, normal person. This is the child. And you can see that the FOXP3 expression was entirely normal. So this is one of the IPEX-like syndromes. It's not due to FOXP3 deficiency, but rather to the IL-2 receptor alpha chain deficiency.
So to sum up, the quantal regulation that we first talked about in Immunological Review in 2008, the proliferation of immunocytes is antigen initiated, but it's IL-2 mediated. And it's quantal. The energy that occurs in some cells is due to IL-2 negative feedback regulation of IL-2 at three different quantal levels, TCR signaling, CTLA-4 shuts down TCR signaling, IL-2 signaling, IL-10 does that. And IL-2 transcription, FOXP3 does that.
Any suppression that's going on in the system is entirely passive, and it's due to the IL-2 consumption. And that's quantal. And ultimately it leads to apoptosis, due to IL-2 deprivation and then cloning deletion of cells. So that the pathogenesis of autoimmunity in the year 2011, I think can be explained by mutational defects and negative feedback of the IL-2 receptor interaction. CTLA-4 is important, FOXP3 is important, IL-10 is important.
And you couldn't have a genetic predisposition to then somatic mutations in any of the genes encoded in these molecules so that you could have germline mutations in one allele within a single mutation in one of the somatic genes encoding CTLA-4, FOXP3, et cetera. It's also clear that the antigen concentration is very important. And so that if you have situations where you have increased autoantigen production, either through inflammation and tissue destruction, either sterile or non-sterile inflammation, or if you have germline mutations.
In different tissues, in molecules in different tissues, you could account for specific tissues causing autoimmune kinds of diathesis where you get premature apoptosis and necrosis and release of more and more antigens. And if you're interested in reading about this, [INAUDIBLE] and I, a colleague in Cornell, published a paper in Immunology & Cell Biology in 2010.
So that's all. And I will end here. And thank you for your attention.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact email@example.com if you have any questions about this request.
Dr. Kendall Smith, the Rochelle Belfer Professor of Immunology and Medicine at Weill Cornell Medical College, reviews the progress that has been made since 1960 in our understanding of the molecules that operate to make the decisions of the immune system when contemplating whether to react or not to invasion by a foreign microbe or molecule. Because the immune system is tightly regulated by both positive and negative molecular influences, abnormalities in especially the negative regulators can lead to recognition of self as nonself and thus result in autoimmunity.