At A Glance
Name: Phillip Zamore
Position: Associate professor, biochemistry and molecular pharma- cology, University of Massachusetts Medical School
Background: Postdoc, Whitehead Institute of the Massachusetts Institute of Technology — 1993-1999; PhD, biochemistry and molecular biology, Harvard University — 1992; AB, biochemistry and molecular biology, Harvard University —1986
As part of the team of researchers who published a landmark paper in 2001 describing how 21-nucleotide siRNA duplexes suppress gene expression in mammalian cells, Phillip Zamore became one of the most important figures in the RNA interference field. (See Nature May 24, 2001; 411: 6836, 494-498).
In 2002, he went on to co-found RNAi therapeutics firm Alnylam Pharmaceuticals, and has since continued to work in academia as a member of the faculty of UMass Medical School.
Zamore took time before the holidays to speak with RNAi News about his groundbreaking work in RNAi, as well as his current projects.
How did you get involved with RNAi?
It was an amazing set of circumstances. I was a postdoc with Dave Bartel, I had something like six or nine months left on my postdoc, and had accepted a faculty position here at the University of Massachusetts Medical School.
Tom Tuschl was also a postdoc with Dave — he was a postdoc jointly with Dave and Phil Sharp. Tom was in the same situation as I was in, but he was almost exactly six months ahead of me: he had also accepted a faculty position at the University of Gottingen, Germany, at the Max Planck Institute.
Neither of us worked on RNAi. We both got our jobs and other postdoctoral [positions] that had nothing to do with RNAi.
The way our group meetings [at MIT] were structured was, the first component was a journal club, followed by an experimental progress report from one of the lab members. I was giving the journal club that week and I presented the Fire and Mello paper in which the molecular explanation for what triggers RNAi was revealed to be double-stranded RNA.
Tom immediately got up and said: “Wow, what a great introduction for what I’m going to present. My project has come to a logical conclusion and I want to start something new in the sort of protected space that one has after you’ve accepted a job but before you actually leave your postdoc.”
He proposed to develop an RNAi in vitro system that would allow you to study RNAi biochemically. He made a very clear case for why he would want to do this, and I of course was primed to hear this, having just read and been stunned by the brilliance of the Fire and Mello paper.
He then proposed his strategy, which I hated. I very smugly proposed an alternative strategy, which he suggested we pursue jointly. And we did.
What was his original strategy?
That I’m not going to tell you. We know what works; there’s no use talking about what might have worked. I do think that Greg Hannon tried the proposed strategy and it didn’t work.
You recently co-authored a paper in Cell discussing asymmetry and the assembly of RISC. Could you discuss the findings?
Well, we started out with the belief that siRNAs have two strands and therefore would be symmetric, and that … most of the bias in the literature — in terms of language talking about the antisense strand of an siRNA guiding the destruction of a sense mRNA—was an artifact of our worldview in which we value more highly mRNAs in cells than artificial RNA sequences in test tubes.
We thought that that if you looked with antisense mRNAs as the potential RNAi target in a cell-free extract, we’d find that half the time they got cut and the other half of the time the mRNA target got cut, because you would find that the two strands of the siRNA had a 50/50 chance of going into the RISC.
What we found is that’s not true at all, and that the two strands rarely have the same probability of going into the RISC, and that one can design siRNAs in which the probability of one strand going into the RISC is 100 percent and the probability of the other stand going into the RISC is practically zero.
[It had] nothing to do with which strand you called sense and which [you called] antisense, and cells didn’t use the information of the target to pick the strand that went into RISC. Rather, the structural relationship of the two strands of the siRNA to each other actually determined which strand ends up being incorporated into the RISC and the fate of the other strand is to be destroyed.
Once we saw that one strand goes into the RISC and the other is destroyed, we realized that that also explains why microRNAs are single stranded — that the mechanistic basis of those two phenomena are one and the same.
I take great pleasure in the fact that two other groups — one at Roche and one at Amgen — essentially came to the identical conclusion by completely independent means.
What about microRNAs? There seem to be these questions lingering about what they are doing and what they target. Do you have any input?
I think I take a somewhat extreme view in the field in believing that there is only a single pathway that transfers one strand of a small RNA duplex into the RISC. Whether that small RNA duplex is the precursor to a microRNA or an siRNA, the pathway doesn’t know the difference.
That RISC complex has all possible functional capacities either associated with it or it can recruit factors to it that will provide all possible functional capacity. Which function is supposed to be the outcome of the experiment is determined by the degree of complementarity to the target. So, the right mismatches will produce translational regulation [and] perfect complementarity will produce cleavage.
MicroRNAs and siRNAs are equally capable of both and are therefore interchangeable.
My view is that we will find microRNAs that regulate translation, but we will also find microRNAs that cleave their target in animals. Already, we’re finding both classes in plants.
What kind of projects do you have going on?
We’re interested in the biochemical basis of translational repression by microRNAs. We’re also interested in microRNA biogenesis. …
We’re very interested in the proteins that are involved in translational regulation that are not involved in cleavage and vice versa. So, we’re using genetics to try to dissect those two pathways apart.
We’re also trying to set up in vitro assays for translational repression.
We’re very interested in understanding the biochemical roles of the different parts of the siRNA — the 5’ nucleotides, the middle nucleotides, the 3’ nucleotides. Do they all have the same role in the enzymology of the RNAi pathway?
We’re starting to learn that the answer is clearly, no. Different parts of the siRNA play different enzymological roles — some contribute to binding, some contribute to catalytically permissible geometries.
We’re also extremely interested in the connection between animal development and the RNAi pathway. In particular, we’re trying to understand the biochemical basis for developmentally interesting proteins that are required for RNAi, especially in Drosophila, where the system is extremely tractable.
And, we’ve continued the next logical steps from our published work: Trying to understand if siRNAs can be asymmetric, what proteins determine the asymmetry and how?
What about things that you may not have started working on but you may want to tackle at some point?
Two big questions for the field that I’m very interested in are: How does this pathway change the structure of chromatin? Clearly, the RNAi pathway has a role in the initiation of hetrochromatin assembly, and I’d like to understand the biochemical basis for that.
I’m also tremendously interested in how RNA-dependent RNA polymerases amplify the RNAi pathway in plants and nematodes — something that, unfortunately, we can’t study in Drosophila or mammalian cells because that part of the pathway doesn’t seem to function.
I’m fascinated, although we’re not doing anything about it, by this paradox in plants that certain kinds of transgenes can elicit an RNAi response that spreads across the sequence of the transgene, and yet endogenous genes seem incapable of that kind of spreading.
Do you have any theories as to what’s going on?
There is almost nothing about which I don’t have a theory, but I don’t have a good theory.
Do you have any predictions on where you see the RNAi field going, even in terms of therapeutics, looking at five years or so?
I think five years from now there’ll be a completely validated set of siRNAs against every human gene that’s been identified. And so, there’ll be no questions of siRNA specificity or design for the practitioner.
As for therapeutics, I think that siRNAs have tremendous promise. There’s a lot of work to be done, but I think the signs are good.
A viable method?
I think so. We have enough hints that it is.