 If we could get everybody to take their seats, we'll get started. For those of you standing in the back, there are various seats, including in the front row, that are marked reserved, but they're not reserved anymore. So feel free to come take them. And there are also, I see, a handful of free seats on the sides as you come down. So again, feel free to find your seats during these introductory remarks. Let me start off by welcoming all of you to the 2009 Trent Lecture. Obviously, it is now 2010, and not surprisingly, there is a story behind that. But first, let me tell you a few things about the Jeffrey M. Trent Lecture in Cancer Research. Jeff Trent was the founding scientific director of the National Human Genome Research Institute. He came here in 1993 to essentially start a program from scratch. He was recruited here by Francis Collins, who was the previous head of the National Human Genome Research Institute, and part of Francis's recruitment here was being given the ability to start an intramural program focused on genomics research. And Jeff arrived here, and from 1993 to 2002 did a spectacular job of leading our intramural program and starting it from scratch, building it up, and making it really a world-class program in genetics and genomics research. He departed in 2002, and at that time I became the scientific director. And one of the first things I did was to establish a lectureship in his name. And this is the Jeffrey M. Trent Lecture in Cancer Research. And on the program, if you see, this is now the seventh such lecture. And you can see the impressive group of individuals that have given the previous six, and indeed the seventh as you will already know and will quickly find out. It's no exception to that trend. But there actually are a couple stories I wanted to tell before I turn this over to the individual who's going to introduce this year's lecture. It actually turns out there's three of us you're going to see on this stage, myself, Eric Green. Next you're going to hear from Kathy Hudson, and then you're going to hear from Carol Greider. A year ago, if you think back, or roughly a year ago, each of our three lives were a little bit different. At that time I was the scientific director of the institute, now I'm the director of the institute. Kathy Hudson was a faculty member at Johns Hopkins, involved in a whole series of activities and research programs. And she now finds herself as the chief of staff here at the NIH. And our speaker, as you will hear about, was just a world-class researcher and a prominent member of the research community, which is why we invited her at that time, essentially a year ago or so, to come give the seventh Jeff Trent lecture. As you will hear and undoubtedly you know, her life has also changed a bit in events that actually have influenced this lecture substantially. It would seem, though, that the three of us, Kathy Hudson, myself, and Carol Greider and the entire NHGRI, it should be a simple activity to put on a lecture like this, but I will tell you in all of my years of organizing symposium and lecture, this has been the most challenging one to have happen. And in fact, we weren't even successful at having it happen in 2009. It's taken us all the way to 2010. And why is that? Well, if you carefully look at your program, you will notice that the date on this lecture is Tuesday, September 29th. And that was the original date to have Carol come here and give this lecture. And it seemed like such an innocent date when we selected it. It certainly was good on my calendar and it was good on Carol's calendar and that seemed good enough. About a day or two before September 29th, we got a call that basically informed us that there was going to be a very, very, very, very very high level visitor to the NIH and that Massour Auditorium was essentially not available to us anymore and it was in our best interest to postpone this talk. And we did that and originally that it was supposed to be a different visitor, not the very, very, very top one but actually the next one down. And it was supposed to be on September 29th. It turned out it wasn't September 29th. That visitor came, but it was actually the next day. But what I did find out was that we couldn't use this auditorium on September 29th because when the President of the United States visits you, about the day before they just locked the room down. They have dogs in here sniffing and they're making sure that everything's fine with this room. So we got kicked out of this room by the White House, which is understandable. And Carol was actually very gracious about this. She completely understood. She said, not a big deal. We'll find another date later in the year. So we innocently, once again, picked a nice date. It happened to be, I think it was December 8th or something like that. It looked good on my calendar. It looked good on Carol's calendar. And off we went and here we were all set to do this. I didn't feel too bad because at least it was still 2009. But then Carol had to go off and win a Nobel Prize and that just completely threw everything because she had to go all the way to Sweden that week. And needless to say, I understood it when she called me this time basically to say, you know what, we're going to have to postpone this again. But the good news is it brings us to today. And in the effort of being green, we decided not to reprint these programs. So I do apologize for the fact that they don't have today's date on them. But we left it as the September 2009 date. But at least it seemed like a green thing to do. And more importantly, it sort of is a memento to the story of two cancellations finally being able to have the seminar today. So with that said, and that explanation of both how we got here and the rich history of this lecture series, I'm now going to turn the podium over to Kathy Hudson, who it turns out has a long connection with the speaker and will introduce her today. Thank you. It's my pleasure to be able to introduce Carol Greiter to you today. Carol and I have been friends for 25 years. We first met as graduate students at the University of California, Berkeley. When I started graduate school there, she was already there, busily toiling away in Liz Blackburn's lab and well on her way to her first cell publication. And this should be inspiring to all the graduate students among you. This is a picture of Carol at the lab bench in 1985. And we worked in a very small department, a couple members of that department are here with us today, where the students, the postdocs and the faculty were very closely tight-knit group. So you didn't have to, when we had a party, which we did regularly, sometimes two on a weekend, or if we were having a student seminar, we didn't really have much email back then. We just posted signs around in the building. And this is a student seminar signed from a very important presentation that Carol gave asking the important question, how crude do Carol's tetrahemina extracts get during telomere elongation? And there's a couple of features here. Figure A shows Carol's supercoil density and her nickname when we were graduate students was supercoil. And in figure B there you can see figure representing what was later to become named telomerase. What you can't read in this slide, but I'd like to read an excerpt for you, is an abstract that her friends, and I think it was Claire, who you'll see photographed later, who wrote this abstract for her upcoming cell paper. And it was titled, Partially Purified Proteinaceous Factors from Tetrahemina Do Nifty Things in Test Tubes. Chromosomes in eukaryotic cells have telomeres. Therefore, telomeres must exist. And given that they exist, they must come from somewhere, especially considering that cells divide and make more chromosomes needing more telomeres. So, do the cells themselves make these so-called telomeres, or do they buy them from yeast, known to be in the business of telomere construction? An allusion to Jack Shostack, who shared the Nobel Prize with Carol. Or does God make telomeres? Undoubtedly God does make telomeres, but he isn't telling how or necessarily even calling them telomeres, and you will have to read God's paper, Supreme Being, forthcoming manuscript, just to find out what telomeres are called. Anyway, if you do the right things, Tetrahemina will make telomeres for you in test tubes. The gels look great and all the controls turned out right. Indeed, they did. So, here's another picture of Carol back in the day with an ugly but no doubt important gel. And being Carol's friend for a very long time, I've been trying to understand the key ingredients for her incredible scientific success. And this is a photograph taken in my backyard in the mid-80s, and you can see Carol pictured there. And you also see pictured there one of the key ingredients for scientific success, both friends and beer. And in many pictures that I have of Carol, there's also this interesting book, a dictionary. Maybe Carol will explain that to you. So, sadly, Carol started before me, and she left well before me at Berkeley. And in 1988, Jim Watson had a particularly good idea. He had had good ideas previously, including those associated with the structure of DNA, but he had a really great idea in 1988, which was to bring Carol to Cold Spring Harbor as a fellow, sort of a super post-doc position. So, we had to say goodbye to her but not for long, and we stayed very close in touch. One of the first times that we got together after I left to graduate school in 1989 was to go on a bike trip to Alaska. And this picture shows the route that we took, and I don't know how to get a pointer to work so I won't, but we biked from Anchorage all the way around that dotted line to Valdez. And on that bike trip, I have no pictures of Carol on a bike at all. I have pictures of Claire who wrote the abstract and her now husband Roland and my now husband Joe, but I have no pictures of Carol on a bike at all. And the reason is because when we saddled up bicycles in the morning, Carol was off like a shot. She rode a bicycle like she does her science, fast and furious, and so we only managed to catch up with her in the evenings, again with that key ingredient, beer. We finished the bike trip, took a ferry back across from Valdez, took a train up to Denali, and tackled glacial rivers and bare and fested woods and had a really great time together. In the intervening years, Carol and I have shared holidays, we've shared vacations, we've shared raising families together, we've been colleagues together at Hopkins. And so Carol did go from Cold Spring Harbor to Hopkins where again she rose quickly through the scientific ranks, becoming the Nathan's professor and the chair of the Department of Molecular Biology and Genetics. So during this time that we've been friends, of course I've accumulated many hundreds of photographs and Eric wouldn't let me show all of them today, but I did want to show you a couple pictures from Stockholm where I had the distinct pleasure of being a guest of Carol's as she received the Nobel Prize. So this is the auditorium where the Nobel Prizes are given out and what you can see in the top layer there is a full orchestra that performed during the course of the ceremony. Down below are the Nobel Committee and sitting in the front on the right is the Swedish royal family and in the front on the left are the Nobel laureates from this year. If you zoom in you can see a bust of Alfred Nobel and you can see that there's green and white decorating above the stage and along the base of the stage and all over and it turns out that those are flowers that were cut from Italian villa where Alfred Nobel vacationed and that whole room was redolent with the smell of these flowers. And here's Carol receiving her Nobel having successfully walked, walked, walked and bowed, bowed, bowed. So and this is where the dinner was following the Nobel ceremony and the center table was for the important people and then on the sides were these tables for less important people and in fact those tables on the sides were gradients and the less important you were the further out you were at these tables. Carol had dinner with the prince, the princess of Sweden that evening. So I'm sure you're dying to know what I wore and what Gigi wore and what Jeff wore. And so on the Nobel website are instructions about how you should dress and as you can see here there's very specific instructions about what one should wear to the banquet and to the ceremony and one of the options is to wear your national costume and so we did actually have an opportunity while we were in Stockholm to wear our national costume but we did manage to get ourselves cleaned up and as a girlfriend of mine here at NIH told me when I said I wouldn't be at a meeting the following week because I was going to Stockholm she said I have the best girlfriend story ever. My girlfriend won the Nobel Prize. Carol? Thank you. Oops. I'm speechless. I must say I have never ever had an introduction like that ever before. Between the poster for the talk when I was a graduate student which basically gave away all of the science and then the history of the rest of what we've been doing recently I guess I can just stop and ask are there any questions? It's really a pleasure to be here. To come back and see so many friends I have a lot of people I come here to the NIH a lot so I really appreciate being able to come here to this much postponed talk. So what I'm going to do today is to give just a little bit of background on Tina Mir's and Toa Amris and then I'd like to tell you sort of what we've been doing over the last five or six years and at the very end tell you a little bit of a new story that I haven't had a chance to talk about yet. It's always good to be able to come somewhere and actually talk some science. So I'll tell you a little bit about the Toa Amris history and discovery just to follow up on Kathy's introduction from the Nobel. I'll tell you a very little bit about the role of short telomeres in cancer. I'll spend the rest of the time telling you about short telomeres in human genetics and age-related degenerative diseases. Notice the human genetics here for the NHGRI just to remind you. So telomeres are the very ends of chromosomes and you can see here in these blue tips and they provide two really essential functions. One is that they have to protect chromosome ends so they serve as caps to protect the ends from nucleases, from end-to-end joining, from various recombination processes. And the second essential function that they have to do is to maintain telomere length. And the telomeres are made of simple DNA repeats that are bound by specific proteins. So this is just an illustration of the very simple telomere repeats. The very end of the chromosome is single-stranded and telomere repeats in many organisms are these simple, tandem-repeated sequences and what's shown here is the human vertebrates have the sequence tta ggg repeated over and over again. You'll also see the repeat tt gggg which is the tetrahymone telomere repeat. So they're very similar conserved repeated sequences. Most of the telomere is double-stranded although there's a small region that is single-stranded at the very end. In order to provide telomere function you need not only the telomere repeats but also the proteins that bind along the length of the DNA. So there are proteins that bind to the double-stranded region and proteins that bind to the single-stranded region of the telomere in order to provide this protective function. So it turns out that the way that DNA replication occurs the very end of the chromosome isn't completely replicated. And so this is what was known in the 1970s. In fact, Jim Watson wrote in a theoretical paper about this problem of replicating the very end of the chromosome. And so without any other mechanism the telomeres would shorten every time cells divide. And so this was posed as a particular problem back in the 1970s and early 80s. And so we were just very curious to find out how is it that chromosomes can be maintained if there's this problem that every time the cells divide they should be getting shorter and shorter. And so just to set the stage about how people were approaching this this is a paper that was written by Jack Shostek and Liz Blackburn who were the co-recipients of the Nobel Prize this year. And what Jack and Liz did was to find a way to look at the function of telomeres. We knew what the telomere looked like in the single-celled organisms called tetrahymina where the telomere sequence had been clearly delineated and that was the TTGGG repeats. And so what they did was to take a circular plasmid from yeast that had a marker and to linearize that plasmid and they put onto the very end telomere sequences from tetrahymina. And when they put this linear plasmid with the tetrahymina telomeres and they put it into yeast they found out two remarkable things. First of all the tetrahymina telomeres functioned as telomeres in yeast that is they protected the chromosome end and this chromosome now divided successfully and was maintained as a linear chromosome. The second thing that they found out in subsequent follow-up paper with Janice Champay is that after this linear chromosome had been maintained in yeast for a number of divisions that there was yeast telomere sequences which were added on to the very termini of these tetrahymina repeats. And so that really got their attention and in order to understand how it is that telomeres are maintained they made a very bold hypothesis and that is that they wrote in this nature paper we propose that terminal transfer like activities are responsible for extending the 3 prime end of the GT-rich strand of yeast telomeres. Such activities could be added as single nucleotides, etc. But this was a bold hypothesis because it proposed that there was a completely unknown enzyme that would be elongating telomeres and there were other hypotheses that involved recombination and other mechanisms whereby telomeres could be elongated. So when I then joined the Blackburn Laboratory I wanted to be able to test is there really such a thing? And so we again started off with tetrahymina which is known to be in the business of building telomeres and what we did was to use a single-stranded telomere sequence oligonucleotide primer shown here as this very small blue arrow and to put those into extracts of tetrahymina along with radioactive DGTP and DTTP and what we found is that the activity in those cell extracts would add this sequence T2G4, T2G4, T2G4 repetitively onto the end of this primer and it turns out that this telomerase as we then subsequently named it is what is responsible for elongating telomeres. And so the very first gel where telomerase was identified is a little bit smeary here but this is a six-base repeating pattern that was extending up to the top of this sequencing gel. So having identified an enzyme activity which would add these telomeric repeats to a telomeric substrate, the next question of course became well, where does the information come from? And so we had various debates about it but what we proposed is that perhaps this telomerase enzyme may contain a template for this sequence within it and so sure enough by following up with various biochemical experiments we were able to show that there is an RNA component which has the complement of this DNA sequence within the enzyme and so the telomerase is actually both a protein and an RNA component that allows the synthesis of these telomeric repeats. So that's illustrated here. You have the telomeric sequence here and the telomerase enzyme which is made up of both a protein component as well as an essential RNA component within that RNA component is this complement to the telomeric repeat. So the template region can be used to fill out and then translocate and make many, many copies of this T2G4 sequence. So we're very fascinated by the telomerase enzyme and how you can have an enzyme carry out this function of having the RNA component as part of the enzyme itself but we're also became interested in what actually this enzyme may be doing in cells. And so as you heard, after leaving the Blackburn Laboratory I went to Cold Spring Harbor Laboratory where I started to get interested in what was happening in human cells at the ends of chromosomes and very fortunately at that time there was the very beginnings of actually the genome meetings that would occur every year at Cold Spring Harbor and I was fortunate that I was at the meeting where they first identified the human telomeric sequence repeat and at that time it wasn't known what the human telomeric sequences were in any organism but that was presented at a meeting in 1989. And so we were interested then in doing an experiment to ask what happens in human cells, in normal human cells to the length of telomeres and once we got a hold of the human telomeric sequence we could do a southern blot and what we found was that in normal human cells as you passage them in culture they normally undergo cellular senescence and this is a southern blot showing the sort of smear of the telomeric repeats because the heterogeneous nature of the length of telomeres and as those cells are passage for increasing numbers of doublings the telomeres are actually shortening. So it turns out that in many primary human cells telomerase isn't expressed and so just that same problem I was telling you before leads to some telomere shortening. So we got very interested in this and this set off a number of proposals about what might happen to cancer cells which have to divide many times and so we sort of started walking down the path of trying to understand the functional consequence of telomere shortening in human cells. So in order to ask critical questions about what would happen in cells that don't have telomerase we decided to use mouse genetics to be able to ask very specific questions and so to do this we generated a mouse in which one of the two copies of telomerase was eliminated. So what's illustrated here is the name for the gene this is the mouse telomerase RNA this is this essential RNA component of telomerase and what Maria Blasco did when she was in the lab is to generate a mouse in which one of the two copies of the telomerase RNA was deleted and when you cross two such mice together what you get of course is the wild type the heterozygote and the null animals and these null animals were perfectly normal when they're born in normal Mendelian ratios but we could show that there was no enzyme activity for telomerase in these null animals so we were then interested in what happens in future generations and when you take two such null animals and we call this the G1 for the first generation in the absence of telomerase and you breed those to each other you get the G2 the second generation null animals the G3, G4, G5, etc so these mice were breeding continuously we didn't see any phenotypes in the early generations I'll tell you about the phenotypes we did see in the later generations but first I want to show you what actually happens to the telomeres these animals are breeding progressively for a number of generations so in order to follow the telomere length we use this quantitative analysis of telomere length that was developed by Peter Landstorp what's shown here is a metaphase spread of mouse chromosomes where the chromosomes are stained in blue with DAPI and the telomeres are stained with a PNA probe this is a very short oligonucleotide probe and the number of probes that can hybridize to the telomere how long the telomere is so the signal intensity of each one of these dots is proportional to the length of the telomere so you can then follow the telomere length with increasing generations and what you get is when you look at the parents here these heterozygous mice there's this nice frequency distribution this being the telomere length and this being the number of ends that have that length and you can see that with increasing generations there was a progressive shortening of the telomeres in these mice so telomerase not only will make telomeres in vitro but it actually is what's essential for maintaining telomeres in the mouse the other thing that's really nice about this assay is that you're actually looking at the chromosomes and what we could see is that not only was telomere length shortening but the loss of the telomere sequence was leading to loss of telomere function so if you look here in the wild type metaphase spread you see these nice mouse metaphase chromosomes and all mouse chromosomes are telocentric that is the centromere is located all the way along one chromosome arm so you see these nice little boomerangs that look like the mouse chromosomes when you then look in the G2 the G4 and the G6 what you start to see is some chromosome ends where we don't detect any hybridization telomeres in the population they've fallen below the level of detectability of this fluorescence and C2 hybridization assay but then we saw some chromosomes that look like this and those are two telocentric mouse chromosomes that are fused end to end and it looks like a metacentric chromosome but in fact these are two chromosome fusion events so in addition to losing telomere sequence they've also lost telomere function I told you that the telomeres are essential for protecting chromosomes from such aberrant recombination events so there's clearly loss of telomere function so what actually happens in the mouse we saw no phenotypes in the early generation mice but in the later generation mice what happens is that we see increased apoptosis cell death or cellular senescence and at the level of the mouse what that translates into is the first thing we could see was a decrease in fertility and then later on infertility because the testes in the tubules the germ cells were undergoing apoptosis we also see apoptosis in the immune system in the B and T cells are undergoing apoptosis and there's a lot of apoptosis in the intestine so tissues of high turnover that are dividing many times are undergoing apoptosis so we and many other laboratories are then very interested in how it would be that a short telomere would cause a cell to die and so in a series of experiments I won't really have time to get into the model that the field really has developed is that short telomeres induce a DNA damage response so if you normally have double-stranded DNA and that DNA undergoes some sort of a break that broken region is recognized by specific proteins and they will signal through P53 and the cells can undergo either apoptosis or senescence depending on the cell type in the case of short telomeres if you have long telomeres these are protected these chromosome ends are protected by those telomere binding proteins I told you about but when the telomeres get to be too short they no longer function as telomeres and they're recognized again by a particular set of proteins that recognizes this as a short telomere that signals through P53 and the cells undergo apoptosis or senescence and we're very interested in understanding the components of this pathway that lead short telomeres to induce this apoptosis we're also very interested in the cellular consequences of the short telomeres so as I've been telling you the telomeres then is really essential for all kinds of cells that have to divide many times so one example of that is cancer cells if you have a particular tissue and one cell undergoes genetic mutation that will allow it to divide many more times than the cells around it you can see that those cells are going to divide relatively many many more times than the other cells would divide and we were able to show that these cancer cells need to have telomerase in order to be able to continue to divide the second situation is in stem cells and these are tissue specific stem cells where you have a particular stem cell that has to undergo self-renewal but also differentiation and that cell division gives rise to many many different tissue types to undergo tissue renewal and it turns out that the telomerase is really essential for this tissue renewal I'll tell you one story about cancer cells and then focus the rest of the talk here on what happens when you don't have enough telomerase for tissue renewal so once we had generated a telomerase knockout mouse we and Ron DePino with whom we were collaborating set out to ask what is the role of short telomeres in the ability of cancer cells to divide and so we crossed the telomerase to a number of mouse models and various kinds of cancer and this is just a table that summarizes some of those findings and the effect of short telomeres in these settings was to decrease the rate of growth of those tumor cells and in some cases when we could look at the mechanism it was clear that apitosis was being increased and that would then decrease the number of times that the cells could divide so we wanted to follow up on these models that suggested that short telomeres can limit the growth of cancer cells and we wanted to be able to look at specific mechanisms and the pathway by which the short telomeres may be triggering the cell death and so David Feldzer who was a graduate student in the lab decided to use another model of tumor genesis and that's a B cell lymphoma model in which the emu-mic sorry the mic oncogene was driven by the emu promoter in B cells and so what he did was to generate a transgenic emu-mic mouse and put it onto the genetic background with the heterozygous telomerase mutation and when you breed these together you can select for the transgenic G1 mice and then breed those and always selecting for the transgene breed the entire line of mice down like this and then we could ask about the rates of tumor formation in the different lines of mice and what David saw was really quite striking and that was that the short telomeres really limit the growth of the tumor cells such they dramatically protect the mice and increase their survival so what's shown here is the percentage of mice that were alive and if you look at the emu-mic mice on their own that are wild type for telomeres which is shown here in the black line you can see that the survival has significantly decreased and by 100 days half of the mice have died and they all have died of this B cell lymphoma when he looked at the mice which were the emu-mic transgene and null for telomerase but the first generation so these are the long telomere mice now what he found was there was no change there was a very similar rapid decline so the B cell lymphoma is growing very rapidly and will kill these mice however when you look in the emu-mic mice that had the very short telomeres what you find is that there was a really significant protection against the B cell lymphomas and actually you can go into these mice and you can look and there are very small micro lymphomas that start growing but then they stop growing so this again makes the point here that it must be not the absence of telomerase but rather the short telomeres that have to be causing this effect because genotypically these mice are null for telomerase and these mice are null for telomerase the only difference between this cohort and this one is that these have short telomeres because they've been bred for many generations so the conclusion is that it's the short telomeres which are inducing the apoptosis and then limiting the growth of this B cell lymphoma one of the nice things about this model it allows you to have access to the cells that are these pre-B cell lymphomas to ask a question about pathways and so what David wanted to test was to ask what happens if you have a short telomere and we know that that signals through P53 and then the cells undergo apoptosis what happens if you actually block apoptosis so he was able to in this model over express the BCL2 oncogene which blocks the apototic pathway and quite to our surprise what we found is that the mice were still protected the short telomere still protected the mice against the B cell lymphoma so the cells stopped growing and David was able to go on and show that even when you block the apoptosis that comes from P53 that cellular senescence will occur and this will also limit the tumor growth so quite strikingly the telomeres can operate through both of these pathways senescence and apoptosis in order to block the growth of these tumor cells the second place where telomeres are really essential in cells that have to divide many times is in this question of tissue renewal and this really came to the forefront when a paper was published in 2001 by the DOKAL group this was a paper published in Nature and it's entitled the RNA component of telomerase is mutated in autosomal dominant dyskeratosis congenital so this is a group in England who were doing family studies and mapping genes that were involved in this particular inherited human disorder and what they found was that in their mapping studies it was the RNA component which was tracking with the disease in their families so what is dyskeratosis congenital? so some of the clinical features of dyskeratosis congenital there are these dermatologic criteria by which the disease gets its name of dyskeratosis there's skin hyperpigmentation or leukoplakia and abnormal growth of the nails but the mortality of the disease was primarily attributed to aplastic anemia or bone marrow failure there's also pulmonary fibrosis as well as increased cancer risk which would increase the mortality in this disease and so our hypothesis given that the human telomerase RNA was implicated here the disease is due to decreased tissue renewal capacity due to the short telomeres so we then wanted to ask specific questions about this disease model and ask what is the role of the short telomeres in this dyskeratosis congenital? now there are several modes of inheritance of dyskeratosis congenital what I was referring to now is autosomal dominant which was identified by the docal group had mapped the gene that's encoding the X-linked form of inheritance and that gene they called dyskeren and it turns out very interestingly that the dyskeren also has a link to telomerase so the docal group had shown that the RNA component of telomerase the human telomerase RNA is mutated they had shown that dyskeren is mutated and in work that we had done biochemically on the telomerase enzyme that there is this specific stem loop in the telomerase RNA which is a box HACA binding domain to which this dyskeren binds so it turns out that the X-linked form of the disease you don't have the dyskeren binding you destabilize the telomerase RNA and you get this effect on telomeres the docal group showed that mutations in the RNA component can lead to dyskertosis congenital and I'll show you in a minute mutations in the protein component of telomerase which we call TERT for telomerase reverse transcriptase also lead to dyskertosis congenital so we got interested in this when a patient came into a clinic at Johns Hopkins with aplastic anemia and it was apparent that this was a family with dyskertosis congenital and the previous examples of autosomal dominant were mapped to the telomerase RNA and the time was a fellow in the lab and now has her own lab at Hopkins was able to follow this family what we call Hopkins family one and to study the autosomal dominant inheritance of this disease within this family and I want to make a couple of points from this particular pedigree first of all the docal group as well as Mario Manios showed that there is a genetic anticipation in this disease that with each generation there is an earlier onset and a worsening of phenotypes within this family and this should remind you of the telomerase knockout mouse where we see progressively worse phenotypes in the later generations the other thing is that the affected people are heterozygous for mutations in either the RNA component or in this particular family it was the protein component which was mutated and so as a geneticist this interested us a lot because if there are heterozygous and it's inherited as an autosomal dominant trait then that suggests that either there's a dominant negative effect or haploinsufficiency so it could be that if you have two alleles of telomerase that the mutant allele is somehow interacting with the wild type allele and taking that out that's a dominant negative effect or if you have a wild type and a mutant allele and it's not sufficient to have a wild type function that's haploinsufficiency and we realized we were in a situation where we could genetically test this very rigorously using our telomerase knockout mouse so we set out to ask the question about whether or not haploinsufficiency was the mechanism by which this disease function was carried out and I'll show you that in fact it is due to haploinsufficiency so the mouse model that I've been telling you about up to now is the telomerase knockout mouse that's on the C57 black 6 genetic background and this is the southern blot a pulse field gel because mouse telomeres are very, very long you can see this long heterogeneous smear which is why we normally quantitate them using this fish assay however there are a number of species of mice and these are all recently derived from the wild so more wild type mice and you can see that the telomere lengths are very similar to human telomere lengths and so what we had done was to cross the telomerase null allele onto this background here of this cast EI which has human like telomere lengths and we were doing that in order to do various other studies and we recognized that having these mice with the knockout allele on this other genetic background would allow us to test the question of haploinsufficiency so what we were able to do then is once we had had this now back crossed for seven generations onto the Castanius background is to now do a cross like we had done before but rather than following the null mice for progressive generations we kept the telomerase heterozygous so this stands for heterozygous generation 1 so these mice are maintained as heterozygous heterozygous generation 2 heterozygous generation 3 etc. and to ask the question what happens with the telomere length and what we found was that there was progressive telomere shortening in the heterozygous state so what's shown here is this quantitative fish assay wild type Castanius is shown here in the black bars this is the signal intensity and these are the number of ends that have that intensity and we're focusing on the lower end of the distribution here so everything on the upper end is put into this last bin which is why you see this particularly high bar here so we can focus on this end of the distribution now when we look at the heterozygous generation 1 we see that the telomeres are shorter than wild type but even more importantly there's progressive shortening in this telomere length distribution with increasing generations showing that at the level of the telomere length clearly haploinsufficiency can contribute to this disease now of course when we're breeding these heterozygous mice when we breed and look at the heterozygous you of course also get the wild type, excuse me, also get the knockout at each different generation and in studying the colony of mice we were able to find that there's actually a decreased survival depending on what generation the knockouts came from so this is a survival curve days versus the number of mice alive and in the wild type the colony is doing quite well if you look at knockouts that come from heterozygous generation 3 parents they have a slight decrease in survival knockouts from heterozygous generation 6 versus heterozygous generation 8 is shown here and here there's a real dramatic decrease in survival in these mice as they are progressively bred when the parents then have shorter and shorter telomeres so this indicates that the inheritance of short telomeres decreases the survival in these mice and it resembles the genetic anticipation that's seen in dyskeratosis congenital in this heterozygous state now when we're breeding these mice we breed the heterozygous and we know that the telomeres are shortening progressively and of course every time we do this breeding we get out the knockouts as well as the wild type and we were very curious to know what happens in these wild type mice when they've inherited short telomeres through many generations and what I'll show you is that these mice actually have very short telomeres even though they're wild type so again I'm showing you for reference the wild type normal wild type telomere lengths and the knockout generation 1 I've already shown you here in this green that the heterozygous generation 5 sorry in the blue the heterozygous generation 5 has short telomeres but surprisingly the wild type 5 litter mates also had short telomeres so here's a cross to heterozygous generation 4 we get a knockout heterozygous generation 5 and the wild type 5 star and we call this wild type 5 star because these mice are not actually wild type and I've been making the point that it's the short telomeres that are causing the disease in these families that it's not the absence of telomerase per se and so in this particular setting telomerase is completely restored to wild type and so we were very surprised when we then looked at the phenotype in these mice and we found that there was a significant decrease in the testes weight and an increase in aberrant tubules so these normal wild type mice compared to the wild type 4 star mice there was clearly a phenotype in these mice again indicating that it's the short telomeres that are causing the effect even when telomerase is restored to the wild type setting and so we called this a genetic disease in the absence of telomerase mutation or a form of occult genetic disease and this is particularly important because these are the alleles that are normally mutated in the human autosomal dominant inherited syndromes and so this suggests that in families with this disease that there may be people where you can sequence the genes that are contributing to telomere length all you want but you will find that they are wild type where it's the actual telomeres that are contributing so it's a form of a hidden or occult genetic disease and we think that this is important to inform in terms of the clinical setting where the telomere length may be really the thing that matters now whenever I show this about the wild type telomere lengths being shorter in these wild type stars people always ask yeah but what would happen in the normal setting that is can you ever get these to be back to normal wild type telomere lengths so we undertook this experiment which took another three years of mouse breeding and what we set out to do is to ask the question about whether the normal wild type length equilibrium could ever be reestablished so we took these heterozygous mice and crossed them to each other as I've shown you and we got out these wild type 5 star mice that have phenotypes not only in the testes but as well as the intestine and the immune system and now we cross these mice to each other to ask for how many generations would you have to cross mice in order to restore the normal genetically determined telomere length and what we found is that after four generations of breeding we did in fact restore the normal genetically defined telomere length equilibrium so here is the heterozygous generation 4 mouse in red compared to wild type mice in blue and the wild type 5 star 2 which is two generations of breeding wild type mice to wild type mice hasn't quite gotten back up to the normal genetically defined equilibrium but now when we get to wild type 5 star 4 now we have overlapping telomere length distributions so this indicates that it's a slow process that we know that telomerase is limiting in cells we know telomerase is limiting because having half the level of telomerase is sufficient to maintain telomeres in these inherited syndromes and again telomerase is limiting and that it takes four generations of breeding with wild type levels of telomerase to reestablish the normal telomere length equilibrium and then when we look at what happens with the phenotypes in these mice if you look at the aberrant testes tubules I showed you that the heterozygous mice have aberrant tubules the wild type 5 star mice do but then now when you breed these back and re-store the telomere length you find that the testes tubules come back to normal so the phenotypes go with the telomere length so we've been very interested in these questions of what are the role of telomerase in disease and how the telomerase RNA and the protein component contribute to this genetic anticipation that is seen in this inherited syndrome in the last few years there have been suggestions from a number of different groups that indicate that the protein component of telomerase may have additional roles besides the telomere elongation role that I've been talking to you about and this comes from over expression studies that suggest that this protein component has telomere length independent functions in one study it was found that the TERT component regulates chromatin and the DNA damage response in another it was proposed that over expression of TERT promotes cell proliferation and related gene transcription and here is another study indicating that the over expression of TERT modulates the wind signaling pathway now we thought it was very important to understand if there is a secondary role of telomerase because in these families that have mutations in telomerase we would want to know about those other roles that telomerase may play that are independent of telomer length and so although these were over expression studies we decided that we wanted to look now in the protein component knockout and ask what happens in the setting of knocking out the protein component would we see different phenotypes that would have to do with these telomer length independent roles of telomerase so one of the first things we did was a gene expression array and we are using the black six mice that have very very long telomeres where we see no phenotype at all in the first generation mice with the long telomeres and this is simply looking at the TERT knockout G1 mouse gene expression compared to wild type and what you are looking at is what is called a volcano plot where we are averaging a number of biological replicates and so these are the normal levels of gene expression and anything that would be different between the G1 and the wild type would be found in these quadrants here in terms of outliers so in both the case of knocking out the RNA component or knocking out the protein component we found no change in the gene expression profile in the knockout setting and we specifically looked at Wint signaling pathway genes and they are in here but they are not altered we also looked directly at the DNA damage response so we took mice that were either knocked out for the RNA component or knocked out for the protein component these are first generation mice and we either didn't irradiate them or irradiated them with five gray and looked at the induction of the DNA damage response this is phosphop53 and you can see a nice induction of the DNA damage response here and the same thing occurs here in these knockout mice and when we quantitated the number of the amount of p53 induction in either no irradiation or the five gray irradiation we found no significant changes in either the RNA component knockout or the protein component knockout so then we wanted to ask a little bit more of a subtle question which was if we don't see any changes in gene expression or changes in the DNA damage response maybe we would see something at the level of the phenotype of the mouse so what we've done now is to take the protein component knockout mouse this turt component knockout mouse and backcross it for seven generations onto this castanius background where we can see very nice phenotypes in this particular genetic background and we can look now at the heterozygous mice we can look at the the null mice as well so we can compare what we see in the RNA component knockout to what we see in the protein component knockout so here's the RNA component and the protein component and when we look at the telomere length again I'm showing you here the wild type telomere length versus the RNA component knockout in the first generation castanius the wild type versus the protein component there's very similar levels of telomere shortening in both the RNA and protein component knockouts when we ask the question about haploinsufficiency we can look at telomere length in the wild type and the protein component null here's the wild type distribution the null and when we look at two independent heterozygous now so these are heterozygous for the protein component the telomere length is intermediate indicating that there's clearly haploinsufficiency for the protein component as well as for the RNA component and we're still early days in really analyzing this colony that we've been developing for the past few years but we can already clearly see that if we look at survival these are the wild type mice shown here the survival curve in our colony and when you look at knockouts from first generation parents versus knockouts from second generation parents you can see that there's a decrease in the survival in these protein component knockouts so we can now look at some of the phenotypes in these mice one of the phenotypes that we've been able to look at both in the Castanius mice as well as in the black 6 mice is the skeletal development the wind signaling pathway is known to be involved in skeletal development and it's been suggested that the number of ribs that may be apparent in this kind of an x-ray may be altered if the wind signaling pathway is altered however in the black 6 case looking either at wild type or the null for the TERT or in the Castanius genetic background looking at wild type or a number of null animals we haven't yet seen any disruptions to the wind signaling pathway so the pathology that we do see in the TERT knockout mice in the testes we see empty seminiferous tubules just like we saw in the RNA component knockout we see the intestinal epithelium we see villus atrophy there's something called extra-medually hematopoiesis and this is when there is bone marrow failure and hematopoiesis starts happening in the liver, spleen and other organs and as I showed you there's decreased survival with increasing generations so our analysis so far of this colony is that the phenotype of the RNA component and the protein component knockout mice is similar indicating again that short telomeres are the things that are be causing these phenotypes and again this is important to know from the clinical setting in terms of what diseases you might be anticipating in families that have these inherited mutations so now just to finish up I just wanted to point out that many of the phenotypes from these syndromes of telomere shortening share features of age-related disease so I've told you about dyskeratosis congenita it turns out that there are a number of other diseases that are associated in these families that have short telomeres so there is inherited bone marrow failure or familial aplastic anemia, immunosinnescence, chemotherapy intolerance there's a significant rate of pulmonary fibrosis liver disease and increased cancer incidence so these are all things that are seen in these families that have these inherited syndromes of insufficient telomerase and telomere shortening as in many cases where you have a particular genetic disease that genetic disease will give you the strongest case in terms of the phenotype but you may find that those pathways that are disrupted are present in other individuals and so we think that by studying these families that have insufficient telomerase it may tell us about what happens in individuals that have short telomeres even when they have normal wild type telomerase so it implies that short telomeres play a role in certain human age related diseases without frank telomerase mutations and these are all situations where there are tissues of high turnover so I just want to leave you with this slide here which shows the very wide distribution of telomere lengths in the human population it's been known for some time that if you look at telomere lengths in humans versus age and this is in total leukocytes but so normal white blood cells that there's a progressive decline in telomere length in age and this we think has to do with the fact that there's limiting amount of telomerase in cells and so the number of cell divisions that occur over the lifetime of an individual in the hematopoietic system outstrips the ability of the telomerase which is clearly active in these cells but there's not enough telomerase to be able to keep elongating these telomeres but the point I want to make about this slide is the huge amount of heterogeneity in the human population in terms of telomere length that there is a very wide distribution and what we are focusing on is the particular subset of individuals which may have a variety of different genetic inputs as well as environmental inputs because we're talking about tissues where there's a lot of cell divisions so any situation where you may have an insult for instance to the immune system and you have to divide more times that's going to cause more telomere shortening so it's a combined effect of both the genetic initial telomere length as well as the environmental history that may put individuals with shorter telomeres at a higher risk for certain of these age-associated degenerative diseases so the combined effect of initial telomere length and environmental history could contribute to these age-related diseases so just to summarize what I've told you I've told you that telomerase is essential for telomere maintenance telomere shortening leads to cell death or senescence after many cell divisions short telomeres inhibit tumor growth through either apoptosis or senescence and this suggests that telomerase inhibitors may be effective in certain kinds of cancer therapy I've shown you that apoptosis for telomerase causes telomere shortening short telomeres limit cell growth even in the presence of telomerase short telomeres may limit the long-term cell division potential inheritance of short telomeres causes phenotype even in wild type animals and these are our wild type star or so-called occult genetic disease short telomeres may cause loss of tissue renewal in normal aging population telomere lengths may predict the onset of certain age-related disease and I just want to give credit to the people in my life in my life, yes also in my life in my lab who've done all of these experiments David Felzer did the experiments with the emu-mic mice showing the both senescence and apoptosis in the cancer cells Margaret Strong has bred all of the castanius mice and Tammy Morris is working with the mice the emu-mic mice to understand the role of telomere dysfunction and our collaborators at Hopkins Mary Armanios has a lab in the oncology department and in pathology to look at the effects of the absence of telomerase we collaborate with Buck Harman, David Huso so I'll stop there, thanks I'm sure Carol will take a few questions but people will go to microphones we have microphones in each of the aisles congratulations for pioneering work so is there a difference in the telomere length in different tissues obviously this is an issue mainly for the brain and other issues where the turnover of the cells is very low yes there clearly in situations where it's been looked at closely if you just look at different tissues in mice you can find that there are different telomere lengths another question though might be what is the rate of telomere shortening in different tissues is it the same rate of telomere shortening and we don't know that there may not be tissue specific amounts of shortening per cell division and also what is the set point at which a short telomere causes these phenotypes these are very important questions but I think they're going to be essential to ask questions about some of these disease situations for the same cell type you have stem cell as well as differentiated cells is there a way to see the difference in those two groups for the same cell type so we haven't specifically taken a stem cell to ask does the stem cell have longer telomeres than the immediate differentiation partners we've looked in a more sort of gross way at the progressive loss through different generations one of the things about these assays is that there is such heterogeneity in telomere length because what happens is a little bit of shortening a little bit of lengthening a little bit of shortening and so to get a precise measurement is fairly difficult so you have to have fairly large differences to see a change thank you as we get closer to induced protein stem cells have you had a chance to look at what the telomeres look like in the induced pluripotent stem cells and I guess the real question is is there a way after the telomeres have undergone shortening during some of these generational shortening events can they be re elongated I think that's a very interesting question and there have been a couple of papers that have published looking at IPS cells but I think that you have it exactly right that they start off with short telomeres and one of the things that we found is that it certainly takes a large number of generations to reestablish a wild type telomere length so I think that it's something that people should be paying attention to because if you start off with something that has short telomeres you can't assume that as soon as you put it into a wild type setting with normal levels of telomerase that the telomeres will then be reestablished back up to the correct level so I think that that's very important that over a period of time you may expect it to be reestablished at the correct equilibrium but it won't necessarily be that right away so I think that nobody is really taking a look at the telomere length and the cells that they're looking at and it could be very highly variable Thank you for a beautiful talk Is there a point of no return in the shortening of the telomeres because the wild type mice you get from head head mating still have short telomeres but the presence of normal levels of telomerase is not enough to rescue this phenotype so is there anything that is known on that? Well in these studies we're starting with mice that are already alive and are at the level where they can reproduce certainly what we would expect is that when a telomere gets to be too short and it's signaling this DNA damage response that at some point those cells won't survive anymore we do know that there is a signaling mechanism by which the short telomeres specifically recruit telomerase so the shortest telomere gets re-longated much more efficiently than a medium length telomere does so there are inherent rescue mechanisms built into the feedback system in terms of the telomere length maintenance but I can't really address that question because we're dealing with the animals that have to at least still be breeding in order to do that experiment That was wonderful I've read that stress or depression or cardiovascular disease or lots of things can cause shortening and I just wondered if you could expand a little on your last point that you made about there's a genetic starting point and then there's other things in the environment and I think we maybe think of environment in certain ways but maybe we should be thinking in other ways about environment and telomeres Yeah I think that there's a large number of things that one can think about that may contribute to telomere shortening we've really been taking initially the genetic approach and asking in these families what sort of diseases do you see and what do we see in the mice so that we can get a question about cause and effect and then we can put telomerase back and know whether or not it was really the short telomeres that caused that there certainly is a large literature of correlation of things that correlate with telomere length but as we know things that correlate aren't always causative so I think that the door is really open for a number of different diseases but we'd like to take the genetic approach and really ask the question to ask the causality kind of question but I really think it's a tip of the iceberg if you start thinking about the number of age related diseases that may have to do with tissue turnover and I think that using this genetic approach will get us to know more of those diseases that are going to be associated with telomere shortening Hi, any thoughts on what the mechanism is that's sensing telomere equilibrium and what's gone so haywire in C57 black 6 mice in particular? So do we know about the mechanism that senses short telomeres? No we don't but we do know that it's the short telomeres not the telomere length equilibrium. We were able to do a genetic cross where we can bring in just a few short telomeres and those mice have the phenotype and it's not relative to average telomere length. So again that's what brings us back to the DNA damage response that the short telomeres actually trigger something and it's not measuring for instance you can imagine it was measuring telomere binding protein and the amount of telomere binding protein that was floating around and then it would be average telomere length that would be causing the effect. Now why the black 6 mice have such long telomeres really is a mystery to me we actually looked at a number of recently derived wild mice caught wild mice and we got from www.wildmice.com wild mice and they have telomere lengths that are very similar to humans so you will hear people say oh mice don't pay attention to them their telomeres are so long that's 1, 2, 9, black 6, DBA those laboratory strains that people tend to use have these long telomeres and my only conclusion is they've lost their normal telomere length equilibrium and somehow the telomeres have been allowed to get long under whatever breeding conditions they were bred under I think that's very interesting but clearly the normal established wild type telomere length is a very tight telomere length and in order to get at those genetic components if they are genetic components you would need to do a very large breeding and we haven't really been able to follow that QTL analysis to ask about the differences but I think that it would be an interesting question. Once again I want to thank Carol I forgot to mention in my introduction this is actually the first seminar she's given since getting the Nobel Prize which is really pretty special that this all worked out and as a small token of our appreciation we would like to share with her this small gift that we give to each of our lectures each year and Kathy if you'll come up on stage we're just going to do one photo op and then we're all done