Nobel Prize lecture: Gary Ruvkun, Nobel Prize in Physiology or Medicine

I now have the great pleasure to introduce this year's Nobel Laureate in Physiology or Medicine, Dr. Gary Ravkan, our second star. Gary Ravkan was born in Berkeley, California. In 1973, he received a Bachelor of Arts with a major in biophysics from the University of California, Berkeley. In 1982, he received his PhD from Harvard University after studying bacterial nitrogen fixation genes in the laboratory of Frederick Ausubel. Dr. Ravkan then started his postdoctoral research jointly between the laboratories of Walter Gilbert of Harvard University and Robert Horvitz at the Massachusetts Institute of Technology, remaining there until 1985. He then became an assistant professor at Harvard University and started his own laboratory at Massachusetts General Hospital. Gary Ravkan is now at Massachusetts General Hospital and professor of genetics at Harvard Medical School in Boston, Massachusetts, USA. Dr. Ravkan, we are very much looking forward to your Nobel lecture entitled, A Vast and Ancient Hidden World of Micro-RNAs Across the Eukaryotes. Please join me in welcoming Dr. Ravkan onto the stage. Am I on? Yes, I thought I might have unplugged. It's really wonderful to get this kind of reception. It's been a wonderful week and the phone call from Stockholm was lovely. Thank you. So, I'm going to talk about our work on the vast hidden world of micro-RNAs and it's many of my students and postdocs over many years. I just want to remind you for a second that what we're talking about is watching these lineages in wild type. They make a transition and in a Lin-14 dominant mutant that I'm going to be describing or the Lin-4 micro-RNA mutant, you suddenly fail to make this switch and you keep going L1, L1, L1, L1. Whereas if you don't have Lin-14 protein, you never get to L1 and you go straight to L2. This was sort of the worm's blessing and curse was to use lineages as a way to assess what a cell is wanting to be and trying to be. So, these patterns of division tells you what the cell thinks it is in terms of time or space. As developmental biologists, talking about lineage was very strange to the developmental biology community. So, we would go and present these sorts of lineage mutations at developmental biology meetings. All the other people who were working on cell-cell interactions would think we're very much interested in ancestry and who your parents are. Developmental biology was all who your neighbors are. Worms were sort of an outlier in biology, but we kind of liked it. The worm community was always a little bit separate. So, the micro-RNA story is a bit of that where we sort of think different. It works to think different sometimes. The gain-of-function mutant was really interesting to study. Using the same kind of chromosome walking that Victor described, we got to the Lin-14 gene. The interesting part of it was not the protein coding segment, which is again a little bit strange. Usually, what people are interested in is the protein. It's homologous to something that's related to a disease. Proteins have a lot more letters to them. There's 20 amino acids in which it matches. It's significant. DNA is harder to match. These gain-of-function mutations turned out to be in the 3' untranslated region. So, this is downstream. It's a 3.5 KB message. About half of it is 3' untranslated. There were two different activating mutations. So, it gave us statistics of two, which is way better than one. One of them was a 600-base deletion that activates the gene. The other was a translocation that removes this back end of the gene. If we look at the protein products, we made an antibody. This is the work of Bruce Whiteman and Il-Ho Ha. Making an antibody to Lin-14, it turned out to be a nuclear protein. At the larval stage 1, it was abundant. At larval stage 2, it was unabundant. It sort of fit with what the genetics was telling us, that the gene is active at the L1 stage and inactive at the L2 stage. Then, the gain-of-function mutant failed to turn itself off. It continued to have this nuclear protein at the L2 stage. This nuclear protein didn't have any homology until a year or two ago when AlphaFold was used to say that Lin-14 folds in the same way as a couple of different things that don't have amino acid homology. AlphaFold has saved our butts on having some homology. So, we're going to the lecture tomorrow. Thank you, AlphaFold. We can see this at a Western blot. Now, since Victor mentioned Southerns, it's time for an educational moment. A Southern blot is named after Ed Southern. Do you think a Western blot is named after Fred Western? No, it's a joke, right? A stupid molecular biology joke. Northern blots aren't named after it's all. Aren't we funny? I know. There's a lot of room to be better than that. So, here's the Western blot showing the protein going down by a factor of 50 from L1 stage to L2 stage, but in a Lin-4 microRNA mutant, we didn't know it was a microRNA when we started. It doesn't go down nearly as much, so it fails in that down regulation. And again, this is Bruce and Ilho did this experiment. Now, to prove that it was a 3' UTR, this is in the days before GFP, right? 1993, GFP is discovered. What is it, 97? You guys probably know, but there was no GFP, but LacZ fusions were the coolest thing then. So, you could take beta-galactosidase from E. coli and put it in front of the Lin-14 3' UTR. So, this is an animal that carries a collagen promoter, LacZ, the Lin-14 UTR. It gets turned off. Oh, I'm sorry. That should say L4. Oops. It gets turned off at a late stage, but in a Lin-4 mutant, it fails to get turned off, or if we switch the UTR to have a non-temporally regulated muscle 3' UTR, it fails to get turned off. So, the Lin-14 UTR has something that turns it off and is responsive to the Lin-4 mutant, and this is the paper that was backed back with the Lin-4 microRNA paper. Here's the Lin-4 stem loop, and this is the microRNA itself. That, Victor's lab showed, is upregulated at the L2. The way that upregulation works is there's these seven sites. You can see where they are relative to the Lin-14 gain-of-function mutant. So, this one is a weaker gain-of-function than this gain-of-function that gets rid of all of them. Here's site one that has Lin-4 on the bottom, Lin-14 on the top, and it's Watson-Crick base pairing with this loop, loop, loop, loop, loop, and that keeps it from getting cleaved. Here's the model that emerges that Lin-4 microRNA gets upregulated, and that downregulates the Lin-14 protein, the translation of the Lin-14 protein. Again, it was done by Bruce. Here he is today, and Ilho. That's where we were in the pathway, but then Let-7 emerged as having heterochronic defects as well, and so now I'm going to tell you the story of Let-7 and Lin-41 that happened a few years later. The Let-7 phenotype is affecting a transition that takes place later, and what we're watching is the pattern of divisions of these cells along the lateral hypodermis of the animal. These are cells that have different temporal patterns, and the last pattern is to not divide and to form these sort of racing stripes on the animal. The wild type has a perfect racing stripe, and the Let-7 mutant fails to develop those racing stripes because it keeps going L-4, L-4, L-4. It never gets to adulthood, which some of us can understand very well. Getting to the Let-7 gene was the work of Brenda and Frank, and this was in collaboration with Bob Horvitz and Michael Basson, who did much of the genetics on Let-7. They had to walk into the region and figure out that the gene corresponded to a microRNA, and we actually didn't expect it to be a microRNA, even though there was another microRNA. Lin-4 had already been discovered five years earlier, but we were imagining it was going to be the regulatory cascade, another set of transcription factors or something. It turned out that it was a point mutation, and the point mutation helped guide us because it said, well, that's the active site of whatever it is we're working on. When they did a northern blot, again, funny joke, northern, not southern, this microRNA is 22 long. It's not expressed at early stages, and then suddenly at the L-4 stage, when we know it functions from genetics, it gets turned up. This is a lethal mutant, and so you can then do a genetic screen. Now, how do you work with a lethal mutant? Well, it's temperature-sensitive lethal. It's lethal at 25 degrees, but if you grow it at 15 degrees, which is fine for a worm, you can grow it up, do a mutagenesis, wait two generations, which is the glorious thing about a worm is you're talking a week. It's not six months to get two generations. It's a week. For humans, it would be 100 years. A week is better than 100 years. You could ask for suppressors out, and out of that, this is the work of Brenda and Frank, and they're in the audience too, and out of that, the suppressor turned out to be the target gene LIN41. LIN41 is a conserved protein. It has lots of paralogs, both in the worm. These are all the worm genes, but these are the mammalian homologs, and it's a ring finger protein, probably a translational control factor. When we looked in the three-prime UTR of LIN41, we saw conserved complementary. Here's the LIN41 sequence going that way. This is the let-7. This is one let-7 microRNA, a second let-7 microRNA. Similarly, bulgy, not the same kind of bulge. This is more of a two-base bulge instead of a five-base bulge in the case of LIN4. We don't know why that is. We have now a case of a microRNA here, the let-7 here, and in fact, there's paralogs of let-7 that I won't talk about that do these transitions here. You can see in the lineage of Sulcinate, there's these waves of division. There's one wave of generation of cells here, a second wave of generations here, a third wave, a fourth wave. Those are all in between molts. At this point where there's no divisions is when the animal sheds its cuticle. This idea of having molting is not unique to C. elegans. There's a whole large clade of animals called the ectisozoans because they do what's called ectisis, which is to shed your cuticle. That's how insects, you go from larval stage to the larval stage as well. There's something about microRNAs and larval staging for sure that comes out of C. elegans. Then we could look with the let-7 sequence and look for homologs, but here we're aided by the emerging genome sequences in 2001. At that moment, there's a human genome that's available. In fact, our hero, John Sulston, who did all this lineage work, he did this working almost alone, maybe two or three people, including Bob Horvitz, working on this, but it wasn't a giant team. Then he aids us, as Victor said, in chromosome walking. I have to tell you that I was important in this because Sulston came and saw me doing chromosome walking and said, what a waste of a brain. You're a pretty smart guy and it's so boring. I could walk from 100 places. You're walking from one place. I'll do it from 100. That's why Victor said we would communicate with them and they'll help us. They did it with everybody, putting together the Genome Project as a gift to the field. There's such a thing as thinking with your hands. He realized how genomics worked with he and Alan Coulson, basically doing cosmic preps and fingerprinting, very labor-intensive stuff. He's not sitting in a room thinking. He's thinking with his hands. At that point, he ramps up to do the Genome Project. He goes from two or three people to 100 people. In fact, he's a third of the human genome, doing a third of the human genome. When we wanted to look for a homologue, we could just take this very short DNA sequence. It was before BLAST, but you would just say, well, is there anything that looks like this in the third of the human genome? That was done with a modem from home. I'm sitting at my desk and you get the data and it comes back and it says, yeah, there's a perfect homologue in C. Briggsia, of course. Well, that's not a big reach, but zebrafish, human. At this point, all there was was human. Zebrafish and all these other chimps, et cetera, they show up in the next five years. Now there's 20,000 eukaryotic genomes. There's another half million bacterial genomes. I can't begin to ... You all maybe know how wonderful it is to be a molecular biologist today, that you have all this data that you can just look at and like, oh, I think I'll blast this. I mean, it's all I do. There's Amy. Of course, guess what that is? That's the LET7 sequence. She lives in California, so that's how people are in California. I'm from California, I know. Here's Brenda again, working on this project as well. Both of them are here today. Then how about proving it? That was a database search modeming into a third of the human genome. Part of this project was me being charming and begging for RNAs from people. Mitzi, who's also in the audience, sent us fly RNA, staged fly RNA. Amy did northern blots. I won't belabor the joke. You can see that there's nothing early. Then it comes on at the equivalent of the L4 stage of a worm, if you don't know too much biology. You can see that this is a late stage regulatory event. It doesn't prove that this does anything. It just says that it's regulated temporally in the same way. In the Lin-4 case, you don't find a homologue in the fly. We could also do it in zebrafish with RNA that Mark Fishman sent. Here's LET7, not there in the early fish, but later in the development of a vertebrate. That's getting closer to a human. It's still a fish, but it's temporally regulated. It validates what the human genome comparison said, but it adds to it that it's temporally regulated. Then we can look in human tissues. You can see that it's in the brain or kidney, liver, etc., and less in cells that are more totipotent or dividing and not so differentiated. Surely, what it did say is that it was expressed as a 22-nucleotide RNA as well. Summarizing here all the organisms we looked at. Again, people sent us RNAs from just the weirdest organisms. Super fun. Some of them came from Australia. Plus means we could see a band. We didn't get developmental series from everybody, but almost every place we looked that's in the bilaterians, and these are the dysozoans, but almost all of them have it, and almost all of them are developmentally regulated. We did not detect LUT7 in these more basal eukaryotes, but other microRNAs have been detected in these, and so LUT7 didn't happen to be conserved. This told us that microRNAs were going to be in many different organisms of the eukaryotes. Then in the end, we have hundreds in the vertebrates, almost a couple hundred in the invertebrates, a few in sponges. There are no microRNAs in the yeast, but plants have also hundreds, and none of the plant microRNAs are equivalent to the animal microRNAs. What it tells you is that the common ancestor of animals and plants had microRNAs, and I'll tell you about Argonaut in a minute. They use the same PIWI protein, the Argonaut, to present those microRNAs, but they've evolved different microRNAs, and it's either because evolution is too fast, but it's really only a factor of two between plants and animals. It might tell you that there's something different, and what I think, and I'll probably get to it, is that the time scales for how it works in a plant versus an animal is very different. We have microsecond things going on in our nervous systems, and maybe I'm just an animal-centric view of the universe. Maybe plants don't have millisecond-type things going on. Their sort of temporal sequence of responses is slightly more slow, and that might be what's different, and I'll get to that in a bit. So, what are some of the other conserved microRNAs? It's not just LET7. Another really highly conserved one is called MIR1, most likely because it was the first one. I don't remember who discovered MIR1. Probably Victor. Oh, there it is. Yes. Sorry. Sorry, Victor. It's hard to keep track. So, this is really highly conserved, and I bet it wasn't the first one you discovered. You just put it top of your list, and actually, Victor's lab shows that it's involved in muscle development, and it does something in our heart muscle in a mouse, and we did some work on it as well. One of the ways to study these microRNAs is to take the gene for it, and even though this is coding for a tiny RNA that folds back, etc., you can still slap GFP onto it. We had evolved by 2001 from lacZ to GFP because Marty Chalfie had made GFP what everybody uses, and you can see now that this is expressed in body wall muscle and the muscle of the pharynx of the worm, so it's sort of a muscle-specific microRNA. In fact, if you look at who the targets might be, you can see good conserved base pairing that the MIR1 is conserved, as is the targeting of an acetylcholine receptor called UNC29. UNC is used to refer to worms that don't move. They act uncoordinated, and in this case, the uncoordinated is due to a muscle defect, a response to a motor neurotransmitter, and these acetylcholine receptors are multiple transmembrane domains with multiple domains, multiple protein domains that heterodimerize, and in fact, in collaboration with Josh Kaplan's lab, we can see that the UNC29 protein is expressed at a lower level in wild type, but when you get rid of this microRNA, it now is expressed at a much higher level, and what was strange about this is that we then expected the microRNA mutant to make more of that protein to therefore be more resistant to levamisol, and we got the opposite result out of that. We don't know why that is. One view is that there's other acetylcholine receptors that are also repressed, and there might be a relative mixture of the two that leads to the levamisol sensitivity. We also looked for more microRNAs. John Kim did this in 2004 in collaboration with Ken Kosick's lab with Anna Krzyzewski, pulling out microRNAs from cortex, and we got miR-124 as one of the abundant things that came out of that, and there again, we could fuse in GFP. This is very highly conserved between mouse and worms, and with a GFP, we could see that it's also neuronal specific, and in fact, miR-124 in a mouse is in the cortex and cerebellum very particular to the brain, and when we looked at where is that microRNA on a polysome prep, so this is where monosomes are here. Polysomes are where translation is really cranking out the product because multiple ribosomes have glommed onto the RNA and are gliding along producing protein, multiple copies of a protein from a single mRNA. We can see that the microRNA is actually in the polysome fraction on a sucrose gradient, saying maybe it inhibits translational elongation and not translational initiation, which is something I'm really interested in today. I wanted to do two minutes on plant micros because they've been really interesting, so at the same time that the Bartell lab and the Ambrose lab and the Tushel lab were cloning microRNAs, this is a Bartell paper, and notice Brenda Reinhart's on this one too, so Brenda's kind of the star of the show here. She was involved in this plant micro work, and one of them called miR-156 from a plant is very highly conserved in plants all the way from Arabidopsis to rice, and rice is a long way from Arabidopsis. It's monocot versus dicot. These are like as far apart as you can be on plant phylogeny, and it's still conserved, and if you look at where it is, some of the complementary sites to the microRNA are in the 3' UTR. That's the light box here, but a bunch of them are actually in the protein coding region of the plant, and it's all perfectly conserved, and the base pairing is perfect base pairing. It's not the base pairing that we see with the loopy, bulgy base pairing in animals, and that's a blessing in terms of finding targets because 4 to the 22nd power is a lot more statistically significant than 4 to the 15th power, and so it's easy to find targets, and if you look at who the targets are, there are a transcription factor called SBP, and this is the coding region of it, and that's what it codes for, and so you can see that the same thing is coded. Here are the ones that are in the 3' UTR. It's again probably that's driven by a selection to target this coding region, and what's interesting about the plant micros is that they tend to work in a much more RNAi-like fashion of leading to degradation. So again, at the time that this micro explosion was happening was the explosion of RNAi, and sort of a key event was Craig Mello's discovery that there's a protein that it's called a peewee protein that mediates the presentation of these 22 nucleotide sRNAs in the RNAi pathway, and so the question was do micro RNAs use anything similar, and it's almost numerological reasoning, right, because we're saying, well, this is 22 long, and this is 22 long, and don't you think they're the same? 22 is the number of letters in the Hebrew language, and so I used to bring people in and say, well, it's a magic number, and there's a whole ideology around 22 in sort of ancient Judaism. The people in my lab were like, is he crazy? And the answer is yes, and so here's DICER. It's sort of, of course it should interact with RNA. It's got an RNase 3 domain, and those trim off from the 3 prime end of a message, and so that's involved in production of sRNAs, and then this is what Craig discovered, RDE1, and then it turned out that RDE1, so this is a program that we developed in my lab. I'll tell you about it in a second, where we take up the protein domains of RDE1, and we do a blast p against everything in the animal, many things in the animal kingdom, many things in yeast, many plants, many protists, sort of 500 across, so blue means there's a homologue and a good one. This isn't weak homology. Bright blue is like, yeah, you'd believe it if you looked at it, and so you can see that the RDE1 peewee is found all across the animals. There's a couple of weird guys that don't have it, and then what was really important in the history of our work is that these proteins, the peewee proteins, of which then these are all the worm paralogs, so these are different peewee proteins. These are the ones Victor mentioned, ALG1, ALG2. These are the ones that mediate how microRNAs work. These other ones mediate other RNAi pathways, and the worms are sort of lunatics about peewee proteins. They have 24 of them. We humans have eight. Plants have 12, 15 sometimes. Worms are really crazy on them, but what was really important here is that Saccharomyces cerevisiae doesn't have this, and that saved our asses because if the yeast geneticists had been able to work on RNAi, they would have eaten our lunch. They're just so damn good. Thank God. It's also interesting. Why did these clades of yeast choose to delete it? I won't get into it, but they dumped 99% of their introns. They have 100 introns left, and we have 100,000 introns. It's part of their trimming down of their genome, whereas Saccharomyces pombi has argonaut proteins and does RNAi. It's not a coincidence that RNAi was discovered almost coincidentally in plants that have the argonaut proteins and fungi that have it, as well as in nematodes. This is the program. If you want to use it, you can just type into Google DepCod, and this was developed by Yuval in my lab and Ruslan, who's a collaborator at Mass General Hospital. Now we're going to go into using feeding RNAi. The discovery of RNAi by Mellow and Fire was super important, but the discovery of feeding RNAi made it easy for all the rest of us to do it. Instead of having to inject double-stranded RNA, Julie Erringer's group at the Medical Research Council figured out—and it was Lisa Timmons in Andy Fire's group who actually figured this out—that you could put in any worm gene between a T7 RNA polymerase promoter, so you make double-stranded RNA in a bacteria and induce by IPTG. You have to be really old to know what IPTG is, but some of you know. That will induce this double-stranded RNA in the E. coli. The worm eats it, and the double-stranded RNAi goes from the gut to the rest of the animal, and RNAi works. If you propose this in a grant, they would kick your ass off. No way, but it works really well. They made these libraries and sent it to us in around 2004, and so we could then just feed. I'll show you what it looks like in a bit. We did feeding RNAi, and this was first with Craig Mello's lab, where we could take Dicer and do RNAi on Dicer, and it looks just like a Let-7, or we could take Alge1 or Alge2 and do RNAi. It looks just like a Let-7. That's not enough to say it's the same, but we could also look at what happens to the Lin-4 microRNA. This is what it looks like in wild type if you're not feeding it anything, but if you get rid of Dicer, now you have less of the mature, and you have the precursor accumulating. The same thing with Argonaut1, and all the other PIWI proteins here don't have anything to do with microRNAs. Just these two have something to do with microRNAs, and it worked on a second microRNA. It's not just Lin-4. We could use it on any of them, and they would accumulate a precursor. One last thing to talk about RNAi is that there's a weirdism of RNAi in some organisms. The RNAi that was studied in Drosophila or in humans, like us, is an RNAi that uses Dicer and produces siRNAs that are the same length as the microRNA, but then there's a secondary amplification process called RNA-dependent RNA polymerase. That takes a primary siRNA that's produced by this pathway, uses it as a primer in the same way that a viral replication works by priming, in the case of a virus, from a tRNA, but this is an RDRP that makes copies, and you end up with a thousand-fold more siRNAs if you have an RDRP. Worms turn out to have four of those, and one of them came out of a screen for enhanced RNAi that I'll tell you more about, and the other ones are important for making RNAi work so that if you inactivate this one with a mutation—we didn't do this, it was Ron Plasterk's lab that did it—they showed that this was important for somatic RNAi and the other for germline RNAi. This is looking at the pattern of who has it and who doesn't. This is taking the RDRPs, and what DEPCOD does is it says, okay, I get a pattern of blue and white. Who else has that pattern of blue and white? All the RDRPs look the same, and then ammonium transporters are sort of cousins, and maybe they act in the same pathway, and phosphoglycerate mutase looks similar, and maybe that acts in the same pathway. I'm not going to go over that. What I'm going to go over here is blue is who has the RDRP, so this bright blue is the worm finding itself, and all these other blues are animals that have RDRPs. I've told you that plants have them, and they use them to extend from the microRNA, but a whole bunch of animals like us don't have the RDRPs, and it's, again, not a coincidence that RNAi was discovered in the organisms that have this two-stage. It was discovered in plants and fungi and animals, but what are the animals that have it? This is an arachnid. This is a spider. I'll show you a list. Again, this is our DEPCOD way of looking at it, but you can also just do classic BLASTp, and most people don't know, but when you do BLASTp, if you go to the taxonomy tab, it will tell you who has it in order of how homologous it is, so the top hit, and here I'm excluding worms, so the top hit are these sort of mud-dwelling protochordates, so they have the best homolog, and then we get down to scorpions and spiders and mites, and it's like you're 13 years old, and it's just what you want to read about is spiders and ticks and mites, and they're badass creatures, and so what it means is that our worms are a lot more badass than we realized, right, and that we should respect their badassness, and of the animals, it's not, it's, you know, again, horseshoe crabs, and what do they have in common? They tend to go hypoxic a lot, so it's like the bivalves, when the tide goes out, they close up, and they're hypoxic during that time, and plants have this, and plants spend a lot of time hypoxic because they get flooded when you water them, right, and I don't have no idea why that has anything to do with antiviral defense, but I'll bet it does, maybe, or not. Okay, so I'm gonna end with about 10 minutes on, and how am I doing for time? Oh yeah, I got six. Okay, that's approximately the same. This is a screen that Sylvia Fisher did. Now, I've told you that worms are really good at RNAi. We can do feeding RNAi. What a gift, but we still thought we could do screens to make RNAi better. Can we get mutants that have RNAi that's even more than normal, which is kind of loony because it's so good, but another view of this is that RNAi is an antiviral defense, and maybe animals sort of induce or don't induce their antiviral defense depending on whether they feel like they're under a viral attack, and so maybe if we go for mutants that are enhanced for RNAi, we're putting the animal in an artificial state that would normally be very transient. They will only happen if there's a viral infection, but now because of a mutant, we're putting them in a state much more consistent all the time, and so that's why it's sort of not stupid to go for things that are improved. You know, the normal way you learn genetics is it's, oh, this is all the ways to break an organism, but really the power of genetics is it's all the ways to switch things on that shouldn't be on or switch things off that shouldn't be off, and that's what physiology is all about. It's all about, you know, when do things get induced and not induced, and so these screening for enhanced RNAi is sort of trying to get at that, and I'll tell you why and how and why. So, we start with a transgene that's a collagen, and it makes the animals roll on a plate, so they crawl around, and they do these little circles, and it's a, and if you don't do RNAi, this strain is rolling just fine, and if you have a mutant that makes RNAi work better, and this happened to be an ARY6-7 that was one of the hits, now the animals crawl away, and so you could just do a screen for things that aren't rolling, and so out of that screen, Sylvia did it, and these are the hits, and she gets sort of 70 hits out of this, and this is in order of who's best, and the ones I'm going to tell you about are Lin35, Lin9, Lin54, but there's others as well, but these are the ones that are sort of fun, and she's now at Harvard Medical School studying these too. So, these are some of the hits, and for example, these are the, called the SYNMOV-Bs, and I'll tell you about these in a bit, so if you take a particular double-stranded RNA that doesn't work very well in a wild type, it barely causes lethality, we don't know why, but we know that in these mutants that are in the, what are called the SYNMOV-Bs, now RNAi works way better, and we get the lethal phenotype almost 100%, and what are the SYNMOV-Bs? So, these are mutations that were originally identified in the Horvitz lab, I'll tell you a little bit about them, and so a wild type animal will make this vulval protrusion that's involved in mating with males to then fertilize the eggs, and that's normally made with one, and the way it makes one is by an anchor cell that's adjacent to this hypoderm sending a signal, it's a growth factor signal, and so he was, they were interested in this because this is a RAS signaling pathway, but the mutants that they got, and these were got, these mutants were presented in, by Chip Ferguson in Horvitz lab group meetings when Victor and I were graduate students, and we would hear all about SYNMOV-As and SYNMOV-Bs, because the original mutant that was isolated was a SYNMOV-A, SYNMOV-B double, and it caused this extra MOV, and then the Horvitz lab did amazing work over about 10 years, 10 different papers, showing that you need to have both of them knocked out, and what that leads to is too much of this growth factor from the hypoderm, not from the anchor cell, and it, you know, it's kind of just weird stuff, but then we get all the SYNMOV-Bs are enhanced for RNAi, and what does that have to do with this? We'll get to that in a minute, and there's Bob, and you know, this is our Victor and my mentor, and he's not, this SYNMOV work is sort of number eight on his list of, you know, what people will say are major discoveries, but it's utterly awesome stuff, wonderful papers, and of course Bob has been important in our early work, but it's coming back, like, you know, this is all in the last two years, right? So what should I work on next, Bob? Please. So what do these turn out to be? So LIN35 is the gene that's sort of the most sexy, and that's retinoblastoma, and that's an oncogene that's involved in retinal tumors, one of the very early oncogenes, but you know, there's human homologs of LIN9, LIN37, all these things have homologs, and what these do is they sort of assemble onto what's called a dream complex, and I had originally kind of resisted working on this because it's sort of chromatin regulation. I was really worried that I will never make sense of chromatin, me. Others can, but I wouldn't, but we've gotten more and interested, and I'll show you why, and this was the experiment that really made it fun. So we started with a LIN8 mutation, so that's a SYNMOV A. It has no phenotype on its own, and normally, in order to get it to show a MOV phenotype, you have to do something to knock down a SYNMOV B, and we decided, well, gee, you know, we know that if we knock down a SYNMOV B, the animal is enhanced for RNAi, and so maybe if we do a virus infection, so the Orsay virus was identified by Marianne Felix from an orchard in France, and so you could infect a worm with Orsay and ask, if we take a LIN8 mutant and infect it with Orsay, do we get a MOV phenotype? And so if we take that LIN8 mutant and don't infect it, nobody's MOV, but if we take that LIN8 mutant and infect it in three separate replicates, we basically could get 10 percent, 10 percent, 10 percent, and so that says that a viral infection essentially down-regulates SYNMOV B somewhere, and this is what a SYNMOV A, SYNMOV B double. It's 100 percent MOV, and so the idea is during a viral infection, this SYNMOV is being turned down, and that helps the animal resist the viral infection because RNAi is enhanced, and that's what you want during a viral infection. You want to be resistant to a virus, and in fact, we could see it when we look at the amount of RNA by RT-PCR, so a wild type has sort of this much, one, and a LIN35 mutant that's constitutively airy resists the infection much better. It's about 20-fold better, and a LIN15B, which is another SYNMOV B, resisted about 100-fold, so it says that this is a natural state, and the reason they're airy is because we're constitutively putting the animal in an antiviral state, and antiviral states are important. It would have been better during COVID if humans were in an antiviral state. The animal loss of the RDRP seems like a bit of a mistake. Maybe it's because we have a nervous system because it seems like plants didn't lose their RNA-dependent RNA polymerases. Okay, I'm going to skip this last bit, and I want to go to talking about the people in my lab, so this is the lab today at our house, and the work I talked about just now is from Ashwin and Himani, and there's other projects I didn't go over, and this is the list of people in the lab over the years, and I want to mention, of course, there's a lot of names. I won't go through all of them, but one of the things I want to mention is lots of times I go and I give a talk, and they say, talk to us about mentoring, and my lab has about 800 person years of people, so most of these people were in the lab for six years approximately, and so they multiply six times the number of people, and it's 800 person years. I'm 40 person years, and even if I say, oh, I'm really good, so maybe I'm worth 80 of those 800, that's still 10% of the lab, and when people leave my lab, there's a hole, right? We lose talent and knowledge, and the people who spent time in my lab remember each other a lot and are friends forever, and they write papers together 20 years later. It's just the most wonderful thing to see, so the mentoring in a lab is the lab, and yes, I'm an important part of it, but I'm not at all the most important part, and I want to mention Diane Cicchetti, who has been the soul of the lab for 26 years and has been really important in all the people. She cares a lot about all the people, and many of you might have seen the famous ice cream truck scene that's on Twitter, and I've got a million hits of me, the day of the Nobel announcement, serving ice cream. It's because her husband, Frank, who's the family friend, has an ice cream truck, and of course, he came on October 7th with the announcement of the Nobel, and we served ice to everybody at Massachusetts General Hospital. Yeah, super fun, and this is the home. There's Natasha, my wife, at the Rolling Stones concert, at the famous Food and Books, and she's been the key part of the lab, of my life, and my lab has always had a theory that she's the brains of the operation, and it's true. I learn a lot from Natasha. Our daughter, Victoria, there she is in medical school, and it turns out there's an instinct to be excited when your kid's a doctor. It's really exciting, and Victoria's been the light of our life, and there she is getting her degree from Amherst, and Natasha was able to give her her degree, because Natasha teaches at Amherst College History of Art, and there's Victoria with our dog Barnaby, who we talk to more than anyone in the house, in a high voice, a lot. I wanted to thank my parents. My dad went to the University of California, which is why I went to the University of California, as did almost all my cousins, and he was an engineer for many years, and that really taught me how to think about technology, and I sort of became, I didn't become an engineer, but I could have, should have, and my mother did not go directly to college. She came out of high school and went with her sister to the Young Hebrew Women's, like the equivalent of the YWCA in the old days, and that was where a virtuous girl at age 17, in her case, and she went to the San Francisco Young Hebrew Women's, whatever it was called, and met my father's sister, who met my father, and one thing led to another, and we did our homework together, my mother and I. That's me at a science fair project. I blame my parents for the terrible orbital mechanics of all the planets being in one orbit. This is inexcusable, but I was only six, and this is my early work was becoming an electronics thing. That's what nerds did before the internet, before computer programming, and I learned electronics, and it really taught me a lot, and then just a quick sort of, you know, yeah, this is going backwards in time. Lab and home, important. Harvard and MIT, important. I spent a year on the road in South America. That taught me about surprises and gave me a hundred stories to regale people with, and they might, it seems like after six years, they get tired of them and leave, but it works for five or so, and then I lived in my van for a year at Berkeley. I did physics, what that taught me were scientific revolutions, and not that I participated in them, but I got to read about them, and physics is taught historically, which is really important because it teaches you sort of the temporal sequence of how things are learned, and it sort of prepares you for revolutions, and when I got to Harvard in 1976, it was obvious that the recombinant DNA was a revolution, and now with genomics, you know, that's the result of that revolution, and it has been a revolution for me, for sure, and many of us, and it really all started with space program. You know, that was propaganda to tell the Soviet Union that we were going to win or send missiles towards them, but what it really did was inspire a whole bunch of nerds of my generation to become scientists, and that's why we have Silicon Valley in the U.S., and why we now have a good pharma industry and good biotech, and a whole lot of us were inspired by that. Lastly, you know, as a kid, you sort of say, well, who am I like, and so I sort of bonded with all the people on TV, and so the early TV and radio was very Jewish, and as a Jewish kid, I said, oh, that's maybe that's like me, and so I wanted to be funny, and so I try, but I think funny is not just for comedy. I think funny is a way to look at the world at a slightly different angle, and that's very scientific, and so I actually think one should use that as a criteria for selecting postdocs. You know, how funny are you, and I think that's it. Oh, sorry, Fred. Asibel is, I want to thank, that's how I started in molecular biology. That was my PhD advisor. He worked next door to me at Massachusetts General Hospital for many years, and Scott Kennedy, who was a postdoc, gave me a deer head that he shot, and Fred was in the office behind me, and I tried like hell for many years to find a deer butt to put in his office, because I thought it was appropriate for somebody who's your former boss, but you can't do it with a junior faculty member. That's abusive, but I never did find a deer butt, but I always looked. Wally Gilbert was just this luminous figure in the Harvard BioLabs that sort of a role model of how to be a scientist, and there were many people like him at Harvard that really inspired me, and over the years sort of taught many of us how to be scientists, and lastly, my department was funded by Hook's Pharmaceutical Company because of the work of Hans-Hermann Schoene and Ernst Winniker. I'm actually the Hans-Hermann Schoene investigator at Mass General Hospital, and that was because Howard Goodman had cloned insulin. They produced insulin, and then they thought maybe they could support it, and it was really because Dr. Schoener had a lot of pro-American feelings after World War II, where he had been a 15-year-old soldier in the army in Germany and was treated well by the U.S. during the post-war period, and so I was a recipient of—World War II didn't end at the time that I was working in it, and it led to wonderful support in my department, so thank you very much.