At A Glance
Name: Premlata Shankar
Position: Assistant professor, Harvard Medical School
Background: Junior investigator, Harvard Medical School — 1995-2001; Postdoc, Tufts University — 1990-1995; Postdoc, Fred Hutchinson Cancer Research Center — 1988-1989; Postdoc, All India Institute of Medical Sciences — 1987-1988; Postdoc, Pasteur Institute — 1984-1986; MD, Banaris Hindu University — 1980
After coming to the US from her native India, Premlata Shankar has established herself at Harvard Medical School where she is combining RNAi with her work with viruses, namely HIV. Recently, she spoke with RNAi News about her research.
How did you get involved in RNA interference?
Phillip Sharp’s group [at the Massachusetts Institute of Technology] was very much interested in RNAi, and we were doing a lot of HIV work. My major interest was in the cellular-immune response in HIV infection — why the patients go on to develop AIDS in spite of an apparently robust immune response. RNAi had just been discovered to work in mammalian cells and they wanted to use it in a viral setting.
So, they approached us and that’s how I became involved. In fact, [for] the first paper that Dr. Sharp and I were corresponding authors on, a lot of the HIV work was done in my lab.
Could you talk a bit about that paper?
It was published in Nature Medicine, and this was one of the first papers where we had applied RNAi as an antiviral strategy. We used synthetic siRNAs against HIV … in cell lines, as well as primary cells, and showed that [RNAi] was highly effective and even in established infections we could suppress viral replication.
That was one of the first demonstrations that RNAi works against HIV and that it works in primary CD4 cells.
Where does that lead us now?
That [paper] was published in Nature Medicine along with other [papers from] other groups —there was [John] Rossi’s group [at the Beckman Research Institute] and there was one by Mario Stevenson [from the University of Massachusetts Medical Center]. These were all the first few papers that came out [on RNAi]. Since then, I’ve been interested in using [RNAi as a therapeutic strategy] against viruses.
First of all, delivery, of course, is a major bottleneck [and] is something we are interested in, so we are developing strategies for [targeted] delivery to the cells. One more thing I’m interested in: In HIV so far, whatever has been published has been proof of principle-kind of studies. My interest is to find certain sequences that would be useful against multiple strains of HIV, because there are multiple clades of HIV [that exist] in different geographic locations — even in the same patient you have multiple quasi-species. So, from highly conserved regions we wanted to first identify [RNAi target] sequences and then, within those sequences, see what the tolerance is to mismatches.
That’s the work we are writing up now. We have identified some sequences … and because we want more stable inhibition than [that provided by] synthetic siRNA, we’ve expressed these sequences as small-hairpin RNAs in lentiviruses. We’ve gotten some very nice results.
You mentioned that you’re working on delivery strategies. Could you touch on how you’re approaching this very difficult issue?
There are some things that we’d rather not talk about now. [However,] I am working in another viral system other than HIV. [As part of] a program project … I am working on using RNAi against the mosquito-borne flaviviruses … dengue, [Japanese encephalitis], and West Nile. [Dengue can] cause hemorrhagic fever, [while] some flaviviruses cause encephalitis like West Nile and Japanese encephalitis. We wanted to use the same technique [that we are with HIV] and identify siRNA sequences that are common to the three flaviviruses.
We got some sequences that are shared. Then we wanted to see if they would inhibit all three viruses, and we [were able to] identify some sequences [that could do so]. Here, what we want to see is [for example] with an encephalitis, to see if we can get neuron-specific delivery. What we’ve done is [taken] our lentiviruses and pseudo-typed [them] with rabies virus, because rabies is known to attack the central nervous system.
What we’ve done is use the rabies envelope to pseudo-type the lentivirus. That gives us a kind of targeted delivery.
In terms of difficulties you’ve encountered, can you talk about that?
Delivery into primary CD4 cells with synthetic siRNAs [has been] very difficult. With the lentiviruses, it’s been easy with activated primary cells, but when we try to do it with resting CD4 cells … [we’ve had problems]. Without activation, even with lentiviruses, we’re not really finding that we get a very good transduction. Although lentiviruses are supposed to infect all cell types, with resting cells, it’s been a problem for us.
Those are the areas in which we’re looking for strategies for targeting resting cells.
What about difficulties in finding sequences that target multiple strains of HIV?
For 19-nucleotide sequences, we haven’t found any that are totally conserved. For example, the best one we have also has a few sequence mismatches. In spite of that, it seems to work, which means that RNAi is tolerant [of mismatches at some positions] — earlier it was thought that even a single nucleotide mismatch would make the RNAi non-functional. But it is agreed upon now by most people that there are some sequence changes that will be tolerated.
Most people when they want to look at what the tolerance [of RNAi] is to mismatches, take the siRNAs and introduce mismatches and then look for whether the target recognition still remains or not. We thought that the better way for HIV is to take a lot of primary viruses and lab-strain viruses and see whether this siRNA we have is able to recognize [the target]. Then, what we did was take the viruses that are recognized and the ones that are not and sequence them [to] see whether non-recognition was associated with certain sequences being mutated. That’s how we went about it.
We found that there were certain sequence mutations that were not tolerated — [mutations in] the central sequences where the target mRNA makes contact with the siRNA, those were not tolerated. [With] … the best candidate [siRNAs we’ve identified,] we’ve found that although they have single nucleotide mutations toward the 3’ region, they are effective against HIV strains from multiple clades.
Since the ultimate goal [with this work] is to develop an RNAi-based therapy for HIV, what’s your take on the feasibility of RNAi as a therapeutic?
I think there are two major hurdles. One, like I told you, is getting resting cells transfected. I don’t think we have an answer for that now.
The other thing is the bio-safety issue of lentiviruses, which are the only kind of transduction [vectors] that seem to work best. That has to be resolved, but they have become safer and safer.
What about down the road? Where would you like to apply RNAi later on?
My major interest is dysfunction in HIV. One of the things I’d like to do is look at this issue and see why the CD8 cells, which are the primary responding [immune] cells in HIV, fail to protect. For example … if it is that these cells are more prone to apoptosis, I would like to use RNAi to see whether I could reverse this process. Or [I’d like to examine] one of the co-stimulatory pathways [and] use RNAi to dissect out what exactly makes them non-functional or dysfunctional.
Is there any other aspect of your research that we didn’t touch on but you’d be interested in talking about?
I published … an interesting piece … in [the Journal of Virology] regarding the longevity of RNAi silencing in primary macrophages, which constitute an important reservoir of HIV in vivo.
We could show that unlike in dividing cells, where the silencing lasts only for three to seven days, presumably because of siRNA dilution with cell division, in non-dividing cells such as macrophages, synthetic siRNA retains its silencing ability for up to three weeks.
We targeted CCR5, the major HIV-1 co-receptor in macrophages, and the viral structural gene, p24, either singly or in combination. Both CCR5 and p24 siRNAs effectively reduced HIV-1 infection for the entire 15-day period of observation. The longer sustenance of siRNAs in macrophages was also confirmed by detection of internalized siRNA in modified Northern blot analysis. More importantly, co-transfection with both siRNAs was able to abrogate HIV-1 infection throughout the 15-day period of observation, suggesting that combined treatment with siRNAs targeting CCR5 and a viral gene may not only reduce viral entry but also destroy genomic RNA of virus that slips through because of passive viral adsorption or incomplete silencing of CCR5.
By analogy to antiviral drug cocktails currently used to disrupt different proteins required for HIV-1 infectivity, it may be possible to use siRNA cocktails targeting both host cell and viral genes to interrupt the viral life cycle at multiple points.
[This is] important because it is the longest lasting silencing effect demonstrated for synthetic siRNA.