© MAX S. GERBER As a student at Case Western University in the 1980s, Greg Hannon got his first taste of honest-to-goodness bucket biochemistry. “I was purifying something called Bence Jones protein from huge buckets of urine,” he says. “It involved running these giant, six-foot-tall sizing columns—a job that was not particularly pleasant, so perfect for an undergraduate.”
But Hannon was even more productive outside the cold room. “I had learned a little computer programming in high school—back when people were still writing code in Basic,” he says. “So, using the coordinates from the protein data bank, I wrote what was probably the first 3-D protein structure visualization program on a PC.” The effort allowed his advisor, Joyce Jentoft, to take a closer look at the Bence Jones protein—and a lot of other things. “For awhile she partly funded her lab distributing that program at 25 bucks a pop.”
The project also earned Hannon his first publication—in Computer Applications in the Biosciences—and ignited his passion for technology development. “Whether it’s writing thousands of lines of code or manipulating oligos on a microarray, I like solving a puzzle. That’s what first appealed to me. And I guess it still does.”
“I try to get my students to write a paper early, so they know that everything’s going to be OK. Then they can be much more adventurous and take more risks.”
Over the years, Hannon has put this penchant for solving technological puzzles to good use—probing the molecular underpinnings of RNA interference (RNAi), exploring how the mechanism protects the genome from the potentially disruptive movements of genetic elements, and building tools that allow investigators to exploit RNAi to study gene function in mammals. Here he discusses how drowning his sorrows changed his career, why it pays to publish early, and how to smoke your own bacon.
Hannon Up and At ’Em
Redirection. Hannon stayed at Case Western for his graduate training, transferring to the laboratory of Tim Nilsen, a molecular biologist who was studying gene expression in parasites. Hannon had already published a paper on ribosomal RNA processing when one of his labmates got roundly scooped. Fellow student Adrienne Takacs had made a cDNA library from a parasitic nematode called Brugia malayi and found that all the 5’ ends had a common sequence. She was still puzzling over this observation at the 1987 RNA processing meeting in Cold Spring Harbor when other investigators reported the occurrence of trans-splicing in the nematode C. elegans—a process that would explain her results. “I can remember it like it was yesterday: sitting on the steps outside the bar, drowning our sorrows and thinking about what to do next,” says Hannon. That’s when Dick Davis, now at the University of Colorado, Denver, mentioned a giant parasitic worm called Ascaris. “It had some interesting properties: you could harvest lots of eggs, which would develop synchronously”—an attribute that could provide the copious amounts of material needed to study how worms carry out trans-splicing, which unites RNA fragments snipped from different transcripts. “Tim got all excited and hunted down some guy in Michigan who worked with Ascaris.” He immediately flew back to Ohio, jumped in his car, and went to pick some up. “He said that by the time he got back, we should have figured out what to do with them.”
Making a model. Figuring out what to do with Ascaris required a bit of research. “Pat Maroney and I had to scurry around reading these papers from the 1940s that were all in German, which neither of us spoke. But the whole project turned out to be such an amazing learning experience.” Three years and 10 papers later, Hannon says, “we ended up with what I think was the first faithful in vitro transcription system for small nuclear RNAs. We had cis-splicing extracts and trans-splicing extracts, and we figured out the splice leader. We basically took this boutique organism that people had used for developmental studies and made it a fertile and interesting biochemical system. When you’ve done that once, it kind of makes you fearless in terms of developing models and systems in the future.”
Coming full cycle. Hannon went to Cold Spring Harbor full-time as a postdoctoral fellow in 1992. The day he showed up for his interview with David Beach “happened to be the day that David and Chuck Sherr figured out that they had both cloned the gene for this new, D-type cyclin. I remember sitting outside David’s office, chatting with the other postdocs, as he and Chuck were faxing each other one amino acid at a time to see if they had the same gene. And I thought, ‘If this place is this exciting every day, it’s going to be really cool to be here.’”
Mexico and the magic RNAs. Hannon first heard about RNAi at a meeting for Pew Scholars in Puerto Vallarta in 1997. “Craig Mello gave a talk, but they knew nothing about the mechanism. It was like you just sprinkle this double-stranded RNA on worms like magic fairy dust and it would silence genes in a specific way. I was totally fascinated.” But it wasn’t until Richard Carthew showed up at the next Pew meeting and reported using RNAi to shut down genes in fly embryos that Hannon made the leap. “I literally walked out and called the lab—my giant lab of three people—and said, ‘We have got to start working on this.’ Within a week or so we had shown we could use double-stranded RNA to knock down expression of a LacZ reporter gene [in a cell line derived from Drosophila]. Then we got bold and knocked down an endogenous cyclin gene. And we were off and running.”
RISC-y business. Hannon’s biochemical know-how came in handy when it came time to dissect the RNAi machinery. “We took our S2 cells, did some crude fractionation, fed the extracts radiolabeled transcripts, and then isolated a nuclease activity that would specifically recognize and degrade target RNAs that shared homology with the double-stranded RNA we put in,” he says. That nuclease activity localized to a complex called RISC—a protein machine that uses small RNA fragments as guides to direct the destruction of target mRNAs, thus silencing the associated genes. The initial paper—published in Nature in 2000—“said that RISC existed and described its basic properties,” says Hannon. “Further biochemical examination led us to one protein after another”—including Dicer, which minces the magic double-stranded RNAs to produce RISC’s RNA guides, and Argonaute, the nuclease that ultimately destroys the RNA targets.
Sharing your toys. With the help of HHMI investigators Steve Elledge of Harvard Medical School and Scott Lowe, currently at the Memorial Sloan-Kettering Cancer Center, Hannon has generated collections of “silencing triggers”—small, synthetic RNAs that can be used to silence genes in mice or humans. “We’ve been constantly trying to improve that technology and get it into people’s hands”—including their own. “That’s always been a problem for us. The guys in my lab who spend all their time developing and building these tools complain that everybody else gets to play with their toys before they do. After all, we develop the tools because we have biological problems we want to solve.”
Early to publish. As a graduate student, Hannon had enough material for a thesis by the end of his third year. “When we started the Ascaris stuff, Tim said, ‘You’ve already got your paper, so you know you’re going to graduate. Maybe you should stay another year and take a chance on this.’ Then things really started to roll, so I stayed even longer. We were having a ball. Every day was something new. Now I try to get my own students to write a paper early, so they know—from the end of their first year or so—that everything’s going to be OK. Then they can be much more adventurous and take more risks.”
Out in four. At CSHL, the goal is for students to complete their graduate training in 4 years. “We have a 6-year absolute limit. At 6 years, if you haven’t defended, you get a master’s and you’re done. It’s tough love.” Former director James Watson championed the abbreviated program. “Jim felt that the length of graduate education was getting completely out of control. The way he put it, ‘You’re old by the time you get out of graduate school, and your most productive, exciting, and creative years are over.’ I don’t know what the average for my lab is, but it’s probably not much over four. I’ve had four students out in three and a half.”
Caffeinated conclusions. “When I write a grant, I’ll sit down in the morning with a goal that I’m not going to get out of my chair until I’ve written a section. I do have to exercise some bladder control to work that way. So it’s a race against the coffee, but it’s the only way I can get it done.”
Spoiled by sequencing. “Next-gen sequencing is incredibly expensive. A student can walk up to the MiSeq machine, press a button, and spend $1,000. But it’s so powerful. We were very early adopters. Back then, you would get 100,000 reads and think, ‘I have so much data, what am I going to do with it?” Today’s machines, by contrast, can generate millions of data points. “Now if students get 100,000 reads, they just throw it away. It’s like, ‘What could I possibly do with such a small amount of information?’”
Bench envy. “I miss working in the lab. I kept my bench for quite awhile. Then one day I went down and all my stuff was on a cart. I was sad about that, but at least I could wheel the cart into the lab and pretend to do experiments. I knew it was really over when I went down and found that somebody had cleaned off the cart. Because they needed the cart more than they needed me doing things in the lab.”
Too many meetings. “I could probably go to a different meeting on small RNAs every week. There’s not really a ‘home’ meeting, even for the noncoding RNA field.” To attempt to remedy that, CSHL has teamed up with the European Molecular Biology Laboratory to create a series that would become what Hannon calls “the meeting people feel like they should go to.” Success, Hannon suspects, will require “an evolutionary process of education and guilt. But once we hit a tipping point, then everybody who wants to do a postdoc with a leader in the field will show up to access the big guys—and the big guys will come to see their friends. Are the days of that kind of meeting gone? I guess we’ll find out.”
Hammer time. “We do have an official Hannon lab hammer. It’s called the ‘ball buster’ because of its history making RNA from testes. But it works for all sorts of material. When you freeze something on dry ice and then powder it, the extraction process is much more efficient.” The use of a hammer to prepare RNA may have been mentioned in some of Hannon’s papers. “There are a few similarly interesting things buried in our materials and methods sections. You have to have a little fun.”
Hannon at Home
Pedal power. Last summer Hannon spent a week biking in Croatia with his 17-year-old son. “We biked around the islands off the Dalmatian coast. The terrain was essentially mountains sticking out of the ocean. I thought we’d be biking around the edges, but that’s not how they built the roads there. The roads go straight up and then straight down. It was tough but fantastic. You’d go up in the morning, to try to beat a little of the heat, and get back down around dinner time so you could lie in the ocean and let various body parts recover.”
Mountain retreat. “My hobbies are as broad as my science. I have a little pottery shed and a wood shop at my place in the Catskills.” The entire lab heads to the region for its annual retreat. “We found a place big enough to hold us all—as long as people aren’t bothered about sleeping on the floor. We go up and basically have a few days away from the lab to talk about stuff and ride bikes and play paintball. I was skeptical about the paintball at first,” Hannon says. But gunning for each other “helps build a sense of cohesion.”
Holy smoke. “I probably shouldn’t admit this, but I actually make my own bacon. It’s like homegrown tomatoes or fresh baked bread. Some things are just worth doing yourself. My assistant’s husband is a butcher. He gets me really good pork belly. I cure it for a week and a half or so and then cold smoke it. And, yes, I built my own cold smoker. It looks like a very small outhouse with a long tube that’s connected to a smoke-generating device—which, in my case, happens to be a hot smoker. The smokers are connected in series and then, just like running stuff through a still, you run the smoke through metal ductwork to cool it down.” The process smokes the meat without cooking it—perfect for making bacon or even salmon. “I’m still trying to master the cold-smoked salmon. I’ve wasted hundreds of dollars and I think I’ve gotten it right once. It’s extremely challenging, but fun to do, like an experiment you can eat.”
• As a graduate student, exploited the parasite Ascaris to unravel the mechanisms by which worms can splice together RNAs from different transcripts.
• Identified and characterized key components of the machinery that lies at the heart of RNA interference, including Dicer, Argonaute, and the RNA-bearing RISC.
• Discovered piRNAs, a class of small RNAs that protect germ cells from transposition-induced damage by directing the destruction
of transposon mRNAs.
• Devised a technique for capturing and sequencing large collections of DNA, a tool that was used, in collaboration with Svante Pääbo, to compare the coding sequences of human and Neanderthal.
• Designed synthetic RNAs that can silence mammalian genes. Working with Stephen Elledge and Scott Lowe, generated comprehensive libraries of these RNAs, representing every gene in human, mouse, and rat.