Name: Aimee Jackson
Position: Research fellow, Rosetta Inpharmatics
Background: Senior research biologist, Rosetta Inpharmatics 2001-2004
Acting instructor, pathology, University of Washington, Seattle 2000-2001
Postdoc, University of Washington, Seattle 1994-2000
PhD, biophysics/genetics, University of Colorado Health Sciences Center 1994
BS, genetic engineering/biology, Cedar Crest College 1989
In 2003, Aimee Jackson and colleagues from Merck subsidiary Rosetta Inpharmatics published a paper in Nature Biotechnology detailing how a handful of siRNAs regulated genes other than the ones they had been designed to target. The findings confirmed suspicions that siRNAs might not be as specific as previously hoped and made the existence of off-target effects a legitimate concern.
Last month, the journal RNA published online two new papers from Jackson and her colleagues, one following up on the Nature Biotechnology paper and another outlining an siRNA modification capable of skirting most unintended silencing.
This week, Jackson spoke with RNAi News about the papers and what else is happening at Rosetta.
Could you first talk about the two RNA papers?
The first paper is an extension of our previous work on the off-target effects of siRNAs, in which we used expression profiling to characterize the regulation of unintended transcripts that shared partial sequence complementarity to the siRNA. In this paper, we extended those findings to a larger number of siRNAs. [We] also extended that work to look at the regulation of unintended transcripts by shRNAs, as well as to look at the effects on the corresponding proteins as a way to show that the off-target effects are probably causing bona fide phenotypes.
The second paper is on the chemical modification of siRNAs. Basically, the first paper is a further characterization of the off-target effects and how they are generated by siRNAs, and the second paper tells us what we might be able to do about them. In this case, the [second] paper shows that the 2'-O methyl modification of a single position in the guide strand of the siRNA reduces about 80 percent of the off-target activity.
In the first paper, did anything come as a surprise given the earlier work you had done?
Nope. Basically, it confirmed everything we had seen in that earlier paper and extended the findings in terms of showing that the region of the guide strand of the siRNA that corresponds to the seed region of a microRNA is really … the key to driving the off-target activity. We show that all of the transcripts regulated by the siRNA show regions in the 3' UTR that contain complementarity to the seed region of the siRNA. So it shows that the unintended transcript regulations of siRNAs are occurring through a mechanism analogous to that of transcript regulation of microRNAs. It's a fundamental aspect of RNAi silencing through the RNAi pathway.
We [also] extended [the work to include] shRNAs, which … demonstrates that you see the same effect with shRNAs, so it's independent of the concentration of the regulatory RNA and it's independent of the delivery method. So this is not just a transfection-mediated effect, it's not a concentration-mediated effect it's a fundamental aspect of siRNA biology.
In the chemical modification paper, what findings do you think were most important?
Because we are able to see such a dramatic effect of reducing the off-target activity without impacting silencing of the intended target, we really are able to distinguish between on- and off-target silencing using the [2'-O methyl] modification. The other thing that was somewhat surprising and unexpected was that you could modify only a single position in the guide strand and achieve this effect. And it really is limited to a single position position 2 of the seed region. We thought it was quite striking that this single position in the siRNA is having such a dramatic effect on the functionality of the siRNA.
Do these findings raise any questions that you're getting into now?
It certainly raises the question of why that particular position is so important and what role the modification of that position is playing in the ability to distinguish between intended transcripts and unintended transcripts. We make some speculations about possible mechanisms and the role of a three-dimensional confirmation of the modified siRNA and RISC and the ability to form a stable complex in the context of a modified siRNA. It would be really interesting to follow those up with some actual structural experiments with modified siRNAs to see if we can visualize those structural changes in the complex.
What's on the fire now for Rosetta on the RNAi front?
We're still going the way we were going. We're still using RNAi for target identification and validation.
Do you think the findings should change at all people's perception of how siRNAs can be used therapeutically?
It could. We haven't looked at that specifically, but [our recent findings] suggest that if you could incorporate this type of chemical modification into siRNAs to be delivered in vivo, it may provide an improvement in terms of more specific transcript silencing.
From your interactions with the community, do you still see a level of [reluctance to accept] the possibility of off-target effects?
I think that that is absolutely the case. When our first findings came out, there was a lot of disbelief, a reluctance to believe that that was real biology. Now I think most people have accepted that off-target activity is real, but a lot of people still question the relevance. When we first published those results on off-target activity, because of the magnitude of the effects is somewhat small, people argued that therefore you wouldn't see a biological effect.
However, the magnitude of the effect we see on the off-target transcripts with the siRNAs is the same as what we see with bona fide microRNAs. So if you argue that the regulation by microRNAs is biologically relevant, then you can't argue that the regulation by the siRNAs is not relevant you can't have it both ways especially since we showed that they're functioning through the same mechanism in terms of binding of the seed region of the guide strand to the 3' UTR of the transcript, and regulating both the transcript and the protein levels.
I don't understand the reluctance in the field to accept the off-target effects except that perhaps it makes life more complicated. We're trying to use RNAi as a tool for functional genomics, but it's important to be aware of [its limitations] and use some of the tools and technologies that are available to reduce the impact of those potentially misleading phenotypic results.
Are microRNAs something that you guys are looking into?
In what capacity?
We are interested in basically trying to understand how microRNAs function, identifying what transcripts they actually regulate, what their targets are, and seeing if we can use that information to predict function. We're also interested in microRNAs as potential biomarkers for disease progression or response to therapeutics.