Robert Horvitz (2010) - Programmed Cell Death in Development and Disease

Hans, thank you for the introduction, and to the Countess and Count, I’d like to thank you very much for inviting me to join this wonderful celebration and for inviting all of us who are here. I must say that I’m very delighted to have had the opportunity and I look forward to the continuing times of interacting with the truly outstanding and interesting students who are gathered at this meeting. Now, in the context of advice and thinking a little about what Roger Tsien talked about earlier today, I think I very briefly will cite the advice I was given when I was a graduate student by my PhD advisor, Jim Watson, who is known for the Watson Crick double helix. And Jim basically said: In choosing a problem to think primarily about two things. First, the problem should be important, because it’s no harder to work on an important problem than on a non-important problem. And secondly, the problem must be tractable, because no matter how important the problem, if you can’t make headway it won’t help. Now, that’s not to say that the problem has to be easily tractable and you may have to innovate to make it tractable, but if you can’t make it tractable, nothing is likely to come of it. So what I’m going to do this morning is tell you a little about a problem called programmed cell death. And I’ll focus on findings from our research lab. Talking mostly about some historical findings, but also trying to at the end bring you up to date with a message that basically says “science doesn’t end”. Discoveries lead to progress, but always I think to incomplete stories. And inform us better about what questions to ask going forward. So let me make some introductory comments about programmed cell death. Programmed cell death refers to the cell death that occurs as a normal aspect of the development of animals. And also to other cell deaths that use the same mechanisms as used in those naturally occurring cell deaths. And there are many examples of programmed cell death in biology. For example, if you think about the metamorphosis of a tadpole into a frog. The tail of the tadpole is lost, that tail consists of cells, the fundamental units of life, and those cells die. If you think about the development for example of a human embryo and a baby, in utero, between our digits, between fingers and toes there are cellular regents that make for webs and those cells die so that our fingers and toes become separate. This regulation can be controlled among species, so that if you think for example about ducks, some have webbed feet, very useful for swimming, some do not, and the difference is a difference in the regulation of programmed cell death. In the development of our brains, programmed cell death is very important. As many as 85% of the nerve cells that are generated as the human brain forms, die. In our immune systems, as we sit here today, as many as 95% of certain of the blood cells involved in our immune responses we generate, as many as 95% die by programmed cell death. Programmed cell death is pervasive in biology. And yet I think it’s not so many years ago when biologists thought about cell death, and I probably should say, if biologists thought about cell death, the basic thinking was cell death is not interesting, rather it’s a phenomenon to be avoided because what does a biologist do? Study cells that are alive, if cells die you can’t do your work. That's a problem, but it’s not an interesting problem. What we know today is this is not the way to think about cell death in many cases. Cell death can be an active process on the part of cells that die and particular genes can act in cells to make those cells die. So there is a biology of cell death, every bit as much as there is a biology of other fundamental processes like cell division, cell migration and cell differentiation. Now, where there is a biology there also can be a pathology. Any normal biological process, if it goes wrong in us, can lead to disease and program cell death is no exception to this. There are many diseases now, and this is just a short and somewhat old list, diseases that are known to be associated with abnormalities in cell death. In some cases cells that should live, die, for example neural degenerative diseases, Alzheimer’s, Huntington’s, Parkinson’s, ALS, Amyotrophic Lateral Sclerosis, heart attacks, congestive heart failure, liver diseases, kidney diseases. We can go on and on with disorders in which cells die. And in at least some of these disorders, but not necessarily all, it has been shown that what is going on is an unleashing of the normal developmental program for cell death in the wrong cell types or at the wrong time. Conversely there are disorders in which cells that should die instead live. For example autoimmune disorders. In our immune system we generate cells that have the capacity to recognise cells in our own body. And normally these cells die by program cell death. If they do not autoimmune disease results. Cancer people often think about, as uncontrolled cell division, cells divide and divide and divide leading to too many cells. Well, in fact the number of cells in our tissues is defined by two opposing processes, the process of cell division, which adds cells, and the process of programmed cell death, which removes cells. And the number of cells can be too high, either because of too much cell division or too little cell death. And certain cancers are fundamentally cancers that are caused by too little cell death and it appears today that most if not all cancers involve too little cell death. Furthermore, some of the classic remedies, some of the treatments for cancer, radiation, chemotherapy they act by activating the endogenous process of programmed cell death. So what this says is that an understanding of the biology and the biologic mechanisms that are involved in programmed cell death is very important both to understanding biology and also to understanding and approaching treatments for a broad variety of diseases. So what I’m going to tell you about now is the discovery and characterisation of genes that function in this process of programmed cell death based upon work primarily from my laboratory over the years. And I want to say that we made these discoveries not focused on a human disease and even not focused on studies of a mammal, but rather using a very simple animal, a microscopic nematode or roundworm, known as caenorhabditis elegans. Now, this animal was introduced to modern biology by my friend, colleague and mentor Sydney Brenner, with whom I shared the Nobel Prize. And it was studied in detail by another friend, colleague and mentor, John Sulston. And one thing that John did was to analyse the development of this very simple animal. It turns out this animal contains as an adult only about 1,000 cells, this is to contrast with us, where in our brains alone we have more than 10 to the 11 cells. And what John did was to study the cell lineage. Starting as all organisms do from a single cell, C. elegans divide from 1 cell to 2 to 4 and so on. And this diagram depicts the developmental origin of every cell in the animal. The Y axis here is time and each vertical line terminates in a cell. If you count quickly going across, you will see that there are 959 cells generated in the adult animal and an additional 131 cells that are generated but are not found in the adult animal. These 131 die developmentally by programmed cell death. And this is one thing that John Sulston established. Now, what we did was to seek genes involved in this process, we did this by looking for genetic variance, mutants, abnormal and the patterns of programmed cell death, cells that should live, die or cells that should die live. And in this way we identified four steps in the process of programmed cell death. First, every cell in the animal must decide: I will live or I will die by programmed cell death. Second, for those cells that decide to die, they must literally execute that decision. Thirdly, dying cells, the corpses must be engulfed by neighbouring cells to remove those dying cells from the body of the animal. And fourthly, the macro molecular debris of the cell corps must be degraded. Another way to remember these steps is very simply indicated here. Identify the victim, kill, get rid of the body and destroy the evidence. So that’s the essence of programmed cell death. How did we get there? Well, again we began with genetic studies. This is the nose of an animal and in essence what we did, and by we I mean a graduate student, Hilary Ellis, and all of the story that I will tell you about today was carried out experimentally by young students, graduate students and postdoctoral researchers in the lab. Hilary Ellis is a graduate student. Basically looked for mutants in which those cells that die normally by programmed cell death do not. And what you can see here, depicted with these arrows, are cells that are in the process of undergoing programmed cell death. And here is a mutant in which those cell deaths cannot be seen. This mutant defined a gene which we called CED-3 for cell death abnormal gene number 3. And it turned out that the mutation that perturbed this gene eliminates the function of the gene. So without CED-3 function, cells that should die instead live, this very simply tells us what CED-3 does, CED-3 kills. Without CED-3 function programmed cell death does not occur. In further studies, this finding, let me back up for a moment, this finding lead to a very important conclusion because it said since programmed cell death requires the function of a specific gene, this means that programmed cell death is an active biologic process. Cells are not dying by default, something biologic is going on to cause them to die. We continued our studies, identified a second gene called CED-4 which behaves very similarly to CED-3, it too is required for killing. Then another graduate student, Junying Yuan, asked where do these killer genes function. Do they function essentially as suicide genes within the cells that are about to die or do they function elsewhere in the body, perhaps sending some signal throughout the body, telling certain cells to die. And what she found was the answer is the former, both of these genes act within cells that are going to die, saying that to at least this extent programmed cell death is a process of cellular suicide. Junying Yuan and another graduate student, Shai Shaham, then characterised the CED-3 gene molecularly and discovered that the CED-3 protein looks like a protein just discovered, actually we had to wait for its discovery, by two pharmaceutical companies interested in human inflammatory diseases. The CED-3 protein is similar to a protein known as ICE, interleukin-1 beta converting enzyme. This is a protease, we heard a little about proteases from Roger earlier this morning. And this finding told us that CED-3 likely acts as a protease cutting up other proteins to drive the process of programmed cell death and post doctoral researcher Ding Xue who had previously been trained by Marty Chalfie, from whom we’re going to hear from next, demonstrated that in fact this is the case. It turned out that CED-3 and ICE are the founding members of a family of proteases, today known as caspases and many studies from many laboratories have shown that caspases drive programmed cell death, not only in C. elegans but also probably in all animals, certainly including ourselves. Now, meanwhile we kept looking for genes. CED-4 we characterised molecularly. Four years later, Xiaodong Wang, working in Dallas, Texas, in the US, found the human counterpart of CED-4. It looks similar and like CED-4, it can drive the process of programmed cell death, which I should say often is referred to by a word that has four syllables, beings with A, and is pronounced in at least seven different ways. I’ll leave it to you to choose how you would like to pronounce this. The Greek speakers in the room have probably the right answer, which none of the rest of us ever use. Then Ron Ellis, a graduate student in the lab, found another cell death gene. And this one was different from CED-3 and CED-4, instead of being a killer, it is a protector, it protects cells against programmed cell death. Graduate student Michael Hengartner cloned this gene CED-9 and discovered it looks like a human cancer gene. A gene called Bcl-2, Bcl stands for B-cell lymphoma and this is a cancer gene that causes cancer by causing B-cells in our immune systems to not die. And that survival of these cells leads to their proliferation and the cancer is state and Bcl-2 like CED-9 protects against cell death. Now, people sometimes have asked me if you ever had an ah-ha moment, eureka, discovery, you know you’re on to something that’s important. And the answer here is that on February 12th 1992 I was attending a conference and graduate student Michael Hengartner was doing something worthwhile. He was working in the lab, he had identified the CED-9 sequence, he did a computer search at this relatively early data base and he sent me this fax at the meeting which says, ‘guess what ends up at the top of the search, the answer human Bcl-2’. And this was the moment that we knew we not only were discovering a fundamental biological pathway in C.elegans that had some counterparts but also that this pathway we were working on was going to be very important in the context of human disease and at this point in particular in the context of cancer. Now, the similarity between CED-9 and Bcl-2, both had similar sequence information, both protected against cell death, led us to ask the question how close are these genes really. And we were able to show that the human Bcl-2 gene, if expressed in C. elegans, would work and furthermore if expressed in a C. elegans mutant defective in CED-9 function, Bcl-2 would substitute for the worm gene. The fact that the human gene could substitute for the worm gene said not only are these genes piecemeal similar, but also there must be similar pathways. And we then worked out the basic pathway and this core pathway looks as indicated here. CED-3 caspase kills, CED-4 kills by promoting the activity of CED-3 and CED-9 protects by preventing CED-4 from promoting the activity of CED-3. We kept looking. Next we found in work from post doc Barbara Conradt, a third killer gene called egl-1. And what Barbara discovered was that egl-1 acts in the pathway at a different point than the other 2 killer genes, before rather than after CED-9. So egl-1 kills by preventing CED-9 from preventing CED-4 from activating CED-3. And this then proved to be the core molecular genetic pathway for programmed cell death and subsequent experiments by a variety of labs showed that these proteins each interact physically sequentially. Egl-1 with CED-9, CED-9 with CED-4, CED-4 to CED-3. And so this is both a biochemical and a genetic pathway for this core component of programmed cell death. Now, with this information one can begin to think about applications to medicine. And the biotechnologies in pharmaceutical industries took note, in fact the day we published our paper about CED-3 is a caspase. I received telephone calls from five pharmaceutical companies, including one from an old friend who said: Since then, a variety of companies have pursued the human counterparts of these players in attempts to discover and develop therapies. For example caspases are killers, so that if you could inhibit a caspase, you could prevent diseases that have too much programmed cell death. The clinical efforts that have gone furthest, are actually in the context of liver diseases in programs that are active today. Conversely, if you could inhibit a CED-9 Bcl-2-like protective protein, you could activate cell death in cells that were otherwise being protected, inhibition of Bcl-2 family members is an active area of interest for cancer and there are a number of anti Bcl-2 programs that are currently in what to my eye look very promising clinical trials for certain specific cancers. Ok, now let’s go on beyond this core pathway. Very succinctly, this is just the beginning, the next step, that of engulfment of phagocytosis of the cell corps, we also have worked out in some detail, we have studied what happens upstream of this core pathway and it turns out that that first step of egl-1 driving the pathway is regulated by cell type in a very specific way. At the level of gene expression with specific transcription factors controlling this process. Different cells are controlled by different transcription factors. We have worked out many cases in C. elegans. In each of these cases we find human counterparts and many of these cases the human counterparts are shown to also act in cell death and apoptosis and human disease. Most of the ones that we’ve characterised are players in cancer, here’s one set of cells, here’s a second set of cells, but that’s not always the case and without going through the details, here is a picture of how this first killer gene in the killing pathway is regulated by different transcription factors in different cell types. We go on to another story, to ask if the generalisation, specific control of this gene egl-1 and cancer is generalisable and the answer is not always. There are always, as we have heard this morning already, surprises and I won’t go through the details. But in fact here is a case where we studied a particular kind of nerve cell death, a neuron in the head of the male that he uses to chemically sense the presence of the opposite sex. And in the other sex, which is not female but hermaphrodite, these cells are generated but die, so this is sexually dimorphic programmed cell death. And the primary regulation of this cell death is not the control of the egl-1 killer gene, but rather down stream the control of the gene CED-3. And this is work both done by Hillel Schwartz, a graduate student in my lab, and Barbara Conradt, who had been a post doc in my lab, working independently. And interestingly, when we ask about human counterparts now and human disease, there are counterparts, but the disorder at least from what's been seen in studies of mice, is not cancer, but rather a neurologic disorder, deafness. If the counterpart in mice is inactivated, the mouse is deaf, the reason is that the hair cells in the ear die and what it looks like, but this is still a work in progress, not by us but by the people who are interested in mice and deafness. What it looks like is that these hair cells are dying because the process of programmed cell death is activated and that leads to the deafness. And this story comes again, the understanding of this story comes from the study of a sexually dimorphic programmed cell death in C. elegans. Now, to go from there, many more genes, much pathway, human counterparts, various diseases, big picture and then is this it. And the fundamental answer and what I said at the beginning is discovery leads to exception, leads to new discovery. As we look very carefully, we find that even in the absence of the CED-3 caspase there is some programmed cell death. Furthermore, the entire pathway that I’ve told you about is not needed for these few programmed cell deaths. There are other caspases encoded by the C. elegans genome and we have eliminated all of them and there is still programmed cell death. So no caspase, no core pathway, still a little bit of programmed cell death. What do these cell deaths look like, there are various criteria for defining the landmark morphological set of changes associated with programmed cell death apoptosis, these are apoptotic deaths by every criterion we can test. And yet this core fundamental pathway is not needed, there is some other way to get there. We have some hints about how this pathway is activated, but the mechanism we do not know and that means we’re not out of business yet, there’s still more things to discover. So with that I want to make one over reaching statement. The studies that we did were fundamentally basic research. We did not target any disease, we worked on an organism that at the point I started was not only obscure but people thought irrelevant. The studies we did were fundamentally genetic in nature at the beginning, genetic studies are often abstract and formalisms. I didn’t know if what we found would be relevant to any organism other than C. elegans and yet our findings have established mechanisms that appear to be universal amongst animals and that are providing the basis for a variety of explorations into new treatments for a very broad variety of human diseases. And I think there is a very fundamental message here and one I hope that every one in the audience, students and the rest of us, will always keep in mind, be it for defining our own research programs or in following the advice we heard at dinner last night, and talking to funding agencies and the public. Basic research is the driver, I wrote here of biomedical knowledge, but for this audience I would say of scientific knowledge. Basic research, curiosity based research is the future of all that we can and will know and it is crucially important for intellectual reasons and for pragmatic socioeconomic reasons that basic research be supported appropriately from those sources that are best suited to do this and that in my view is national governments. So with that little bit what I want to do is to end by showing you a celebratory reunion from 2002, thank you Hans for helping in making that party possible and also I list here those people currently in and previously in the lab, whose work I alluded to day, our studies for programmed cell death have been but one of a variety of adventures we’ve had and certainly one of the ones that has been most rewarding and most existing. And I’ll stop with that and thank you very much.

Robert Horvitz (2010)

Programmed Cell Death in Development and Disease

Robert Horvitz (2010)

Programmed Cell Death in Development and Disease

Abstract

Programmed cell death (often referred to as apoptosis) is a normal feature of animal development and tissue homeostasis. The misregulation of cell death has been implicated in a diversity of human disorders, including cancer, autoimmune diseases, heart attacks, stroke and neurodegenerative diseases. Our laboratory has analyzed the mechanisms responsible for programmed cell death by studying the nematode Caenorhabditis elegans. During the development of C. elegans, 131 of the 1,090 cells generated undergo programmed cell death. We have characterized developmentally, genetically and molecularly the roles of many genes that function in C. elegans programmed cell deaths. We have analyzed genes that control the death process, genes that act in the phagocytosis of dying cells by their neighbors, and genes that function in the digestion of the DNA of cell corpses. We have studied in some detail genes that specify which cells will or will not express this cell-death program. Most but not all of these genes involved in cell-type-specific programmed cell death encode transcription factors that specify whether or not the first gene in the core killing pathway, egl-1 (which encodes a BH3-only member of the BCL-2 protein superfamily), is transcribed. Many of the C. elegans genes involved in programmed cell death show structural and functional similarities to genes that act in mammalian apoptosis, indicating that the major mechanisms of programmed cell death are conserved among organisms as distinct as nematodes and humans. A number of the human counterparts have been implicated in human disorders, including deafness and cancer. We recently have been analyzing a few cell deaths that occur in the absence of the activity any of the components of the core killing pathway. These deaths occur even in mutants defective in all four C. elegans caspase genes (caspases drive apoptosis in worms and other organisms) and involve cells that normally are killed rapidly by the core killing pathway but in the absence of this pathway still die but do so more slowly. Thus, there is a second as yet uncharacterized pathway for programmed cell death in C. elegans, and for at least some of the cells that die by programmed cell death during C. elegans development two distinct mechanisms of cell killing act to ensure that the cells that should die do so.

Cite


Specify width: px

Share

COPYRIGHT

Cite


Specify width: px

Share

COPYRIGHT


Related Content