Research into RNA interference (RNAi) and its emerging use as a tool to explore gene function has taken the research community by storm. Can you tell us how you first became interested in RNAi?

Dr. Hannon: It’s sort of interesting actually. I was at a Pew Scholars Meeting, it was my first year.

Craig Mello was also a Pew Scholar, and he gave a small "chalk talk" at the meeting - although we were in the bowels of Mexico somewhere, so there was no chalk. He presented this really interesting phenomenon. It was either just before or just after the first RNAi paper was published. It really excited me and we chatted a couple times about it at the meeting. But we didn’t really do anything about it, not being C. elegans biologists. His theory just sort of percolated for a year.

Then the next year, again at the Pew Scholars Meeting, Rich Carthew showed that RNAi worked in Drosophila. He was doing this by embryo injection. And that’s what really pulled me in. Here was something that was not just a biological oddity of C.elegans (I was unaware of the plant work at that time so I hadn’t thought about it very deeply). The notion that this phenomenon was going to be universal really captivated me. This was partly because I was spending a lot of effort trying to do forward genetics in cultured mammalian cells using things like retroviral libraries. I saw the RNAi phenomenon as, one day long down the road, something that could complement over expression approaches, and that would give us the loss-of-function tool that mammalian systems have always lacked.

Wow, "…one day long down the road…", I bet you were surprised at how quickly this field has progressed.

Q. So how did your interest in using RNAi to study gene function morph into your work focusing on the mechanism of RNAi?

Dr. Hannon: We got hooked on this technique in part because I have a background in RNA processing - I worked on trans-splicing with Tim Nilsen as a graduate student - so there was the possibility that we would have just run with the whole notion of doing gene function in S2 cells. As it turns out, for most of the things that we were interested in - cell cycle control, and such — S2 cells are a terrible system. But we said, okay, since nobody understands much about the RNAi mechanism yet, let’s play for a little while and see if we can generate an in vitro system from cultured cells that we can use to try to figure out how this all works. And we sort of got sucked down the mechanism path from there.

Q. Last year, your lab, Mello’s lab, and others published research indicating that Dicer is the nuclease that digests long dsRNA into siRNAs which in turn mediate RNAi. Can you describe some of the experiments that led you to this conclusion?

Dr. Hannon: We published Dicer about the beginning of last year, January, I think. We had approached this purely from a biochemical standpoint, where we had taken a "candidate gene approach". We tagged all of these proteins, did IPs (immunoprecipitations), and looked for activity. It led to a beautiful biochemical correlation between Dicer activity and siRNA. In other words, we had an enzyme that did basically what it was supposed to do. But any time you have a biochemical result, you have to be somewhat concerned about whether or not the story you’ve derived based on in vitro experiments has any basis in reality. To really know that something is involved in any kind of specific pathway you need genetics. Ultimately the proof is always in the genetics. And we had done a really crappy experiment, which was to use RNAi to knock out Dicer - that’s sort of become de rigueur in the field - so I don’t know if I should still be embarrassed about or not. But our primary motivation was just to verify the hypothesis that Dicer was the initiating enzyme…..

Q. Now we know that Dicer is also important in a gene regulatory pathway involving short temporal RNA (stRNA) and your lab’s work helped to demonstrate that. How did you tie Dicer to the stRNA pathway?

Dr. Hannon: Shortly after Emily left, Plasterk’s lab finally found a C.elegans Dicer mutant. It was initially somewhat worrying that there weren’t somatic RNAi defects. Ronald tried to be reassuring and talked about maternal effects. But ultimately when they looked at germ line transgenes, silencing of germ line transgenes was defective. And this was sort of a nice confirmation. I think at least now there’s no doubt that Dicer really is the initiating enzyme for the process.

Q. What do you think will be the single greatest outcome of research into RNA silencing?

Dr. Hannon: It’s really an impossible question to answer. One of the things that makes this field very hot at the moment is that in a way it’s all things to all people. You have very interesting biology. For example, scientists are studying transposon silencing, the evolution of the relationship between repetitive and mobile genetic elements, DNA parasites and their hosts, and the interaction between viruses and their hosts. That group is interested because RNAi is involved, at least in plants. And there is also a definitive relationship in C.elegans. RNAi seems to be involved in somehow controlling these kinds of nucleic acid parasites. But there are also people looking at these micro RNAs and, at least the N=2 for endogenous gene regulation, and saying, "Look, here’s a whole new regulatory paradigm where there are hundreds of, what you might call orphan hairpins, running around out there. We don’t know what they do, we don’t know who they regulate, we don’t know how prevalent this is, how general it is, or even really at what regulatory levels these different things act. We know about two that act at the level of protein synthesis, but there’s nothing to say that others don’t act at the level of message stability, or even at the level of directing the modification of chromosome structure. Here we may find that this is something that is as important as the discovery of enhancer sequences, in terms of controlling gene expression itself.

You’ve also got a group of people who are interested in the basic mechanics of RNAi and what it means. Everybody is intrigued about a mechanism where a worm can eat a piece of RNA and knock out a gene in its progeny. There’s something intrinsically appealing about that - understanding the mechanism of that bit of biology. And there’s a whole group of people who want to understand the biological ramifications of the system - that it may regulate development in plants, maybe stem cell identity, or maintenance of stem cell character in both plants and animals.

And then, a much broader group of people, who don’t really care that much about mechanism, are just interested in harnessing this phenomenon as a tool. So, it’s really impossible to predict the single greatest outcome of this research. And I think if you ask that question of ten different people, you’d get ten different answers. The reason I can’t give you just one is because we’re interested in all of them.

Q. Recently, your lab and other labs have developed expression vectors that express siRNA long term. Do you see the use of siRNA expression vectors as a replacement to transfection of siRNAs for inducing RNAi, or are these techniques complementary?

Dr. Hannon: Oh I think they’re complementary. Very much so. It’s still early in terms of trying to understand the power of each of these technologies. What seems to be true, at least this early on, is that siRNAs can get into cells at very low concentrations to provoke a very good effect. I think the jury is still out on whether it’s easier to get an effect with an siRNA on a sort of per cell or whole population basis. We don’t really have that much information on it. And I suspect eventually these things will run even because of advances in different transfection technologies. But in terms of ease of use, nothing beats typing in a sequence, having a couple oligos show up, and then dumping them onto cells, right? If you’re looking for a quick answer, and you only have one or two genes that you want to look at, nothing is going to beat the idea that you can chemically synthesize these things, just in terms of ease.

Q. What are some of the advantages of expression vectors over chemically synthesized siRNAs?

Dr. Hannon: The expression constructs are going to be powerful for a different reason. And maybe more appropriate for specific sorts of experiments that involve much more long term analysis of phenotype, or biochemical studies that involve larger cell numbers that might be more difficult or more expensive to do by transfecting each cell that you want to analyze.

Q. Cell logistic problems, right?

Dr. Hannon: Right. There are a lot of phenotypes that you want to look at over long time scales or in mosaic animals. That’s where the power of these kinds of expression constructs are going to be. Now, another advantage of the expression constructs is the fact that they are propagatable - you make one and validate it and you have it forever. You never have to remake it. You never have to reorder it - it sits in the freezer and you can trot it out and use it any time you want. We are finding that you can marry these sorts of modular cassettes with pretty much any gene transfer technology that you want to talk about - viruses, etc. They will be useful with model systems like tissue slices, where it might be more difficult to get siRNAs [inside cells] in good numbers.

Q. I’d like to ask you about your laboratory. How many people do you have working with you currently and how are they split between post docs, grad students, etc?

Dr. Hannon: We are at around 16 at the moment; 6 graduate students, a couple of visitors, about 4 or 5 post docs, me, a few technicians, and a research associate who is semi-independent, who’s also working with me.

Q. We know that you’re the founder of the biotech company, Genetica. And you’re obviously busy as a professor at the Cold Spring Harbor labs, too. Do you still find time to work at the bench? And, if so, do you think that’s unusual for someone in your position?

Dr. Hannon: Yeah, oh definitely. I try to work at the bench every day. I’m not sure that I do anything useful, but I try to work. I think that Cold Spring Harbor has, not necessarily a tradition. But a lot of the faculty here do work at the bench. And, if you think about it, it’s sort of what got us involved in this whole RNAi work in the beginning. I find that it keeps me engaged much more in the day to day activities of the lab. It also keeps me grounded in the reality of doing experiments and makes me much more understanding about students’ ligations failing occasionally, because mine fail right alongside theirs. And I think that the work moves faster because I’m there. I’m available. We talk about things more - everything from the biology of RNAi to the nitty gritty details of lab work, like, "…gee, this isn’t working", and "…maybe I’ve seen that before..". That helps us troubleshoot and move things along a little bit more quickly. I like being in the lab, and I like interacting with the people in my lab.

Q. What’s next for Gregory Hannon?

Dr. Hannon: I’m going to go do mini preps for one of my students, that’s what’s next.