• Senior investigator, Immune Disease Institute, Harvard Medical School
• Professor, pediatrics, Harvard Medical School
• Director, division of AIDS, Harvard Medical School
• Associate professor, pediatrics, Harvard Medical School — 1998-2004
• Assistant professor, pediatrics, Harvard Medical School — 1996-1998
• Assistant physician, hematology/oncology, Tufts-New England Medical Center — 1988-1995
• MD, Harvard Medical School/Massachusetts Institute of Technology — 1981
• PhD, theoretical physics, Rockefeller University — 1974
• AB, physics, Harvard University — 1969
Harvard Medical School's Judy Lieberman and colleagues recently published data in Cell Host & Microbe demonstrating that intravaginal administration of cholesterol-conjugated siRNAs targeting viral and host genes could prevent herpes simplex virus-2 infection in mice for up to one week.
This week, RNAi News spoke with Lieberman about the findings and her efforts to develop siRNA-based prophylactics for both herpes and HIV.
You had been developing an siRNA-based microbicide for HSV before, but you were finding that the viral protection was rather short-lived.
Yes. First of all, [in our earlier work] we found that the [protection] was short-lived. In trying to overcome that, we increased the amount of siRNA we administered, [but] we didn't find any improvement. We became concerned that the transfection lipid might actually be causing some problems.
In the original Nature paper, we used lipoplexes — we mixed the siRNAs with a transfection lipid — and administered it that way. What we found was that the transfection lipid by itself facilitated viral transmission. We think it transfected in the virus … [by] causing the viral envelope to fuse with the cell membrane. So we needed to find something else.
We found that naked siRNAs didn't get into epithelial cells, but if we used cholesterol-conjugated siRNAs, they got in almost as well as the transfected siRNAs. And they worked even if we gave the siRNAs after the viral challenge or up to a week beforehand.
With the cholesterol-conjugation approach, were you finding that you could use the original dose of siRNAs?
We stepped it up a little [to] four-fold what we used before. So we had to go a little higher, but it was still well within the range of what would be clinically feasible.
Can you talk a little bit about the targets of the siRNAs?
Previously, we had targeted two viral genes: a gene that encodes for a glycoprotein and [one] for a DNA-binding protein. The one that worked the best was [one for] the DNA-binding protein, which is called UL29. So we continued to use that.
One of our hypotheses was that it might be that one of the reasons that the [viral-protection] response was short-lived [in the earlier work] was because maybe you need to express a target gene in order to have long-term silencing. We came to that notion based on some of our work with HIV in vitro, but that hypothesis didn't turn out to be the case.
What we used in this [latest] paper is a combination of an siRNA targeting the viral gene UL29 and an siRNA targeting … [an HSV] receptor called nectin-1, which turns out to be probably the most important receptor.
[ pagebreak ]
One of our notions was that if we targeted [a] host gene, it would be expressed and therefore the silencing would last longer. And in fact, the antiviral protection of silencing nectin-1 was more durable that the silencing of the viral gene. Our study suggests that the reason for that is that the nectin-1 siRNA was probably incorporated into RISC much more efficiently than the viral siRNA to start with. It has a similar half-life within the cell, but it was taken up much better at the beginning so [its effect] lasted longer.
Do you anticipate that this dual-targeting approach is going to be the most effective approach?
When you silence a host gene like a receptor, it takes a few days for that to work, whereas if you silence the viral gene, it works immediately.
So by silencing the viral gene, you get immediate protection but it's not as long lasting?
The viral gene works immediately, even if you delay giving the siRNA until after you're challenged with the virus. It's not clear to me yet whether you have to target a host gene, or if we improved on the viral siRNAs, they would also be taken up very efficiently and provide durable protection.
In all of our studies, we try a few siRNAs and take the ones that look the best. We don't do a systematic survey for the best siRNA for each gene target, which is what you'd do if you were developing this as a drug.
Is there the potential for adverse events when you target a host gene?
With any host gene you would worry. It turns out there is a mutation in nectin-1 that has been associated with cleft palate. [Since] people with cleft palate don't have any other medical problems, I think nectin-1 may be a gene required during development … and not in a mature individual. However, that means that you might have some concerns that silencing nectin-1 might have an adverse effect on the fetus in pregnant women, for instance. So it's definitely something you would want to look at.
In the paper, were you finding any unintended effects with treatment such as toxicities?
No, not at the doses we used. We looked pretty carefully and didn't find any evidence for inflammation, induction of interferon genes, interferon-response genes, or the genes that would be activated by TLR3 engagement. That was all good.
There was one thing that we found that would have to be borne in mind -- when we did a dose response, we didn't get protection at the highest dose we looked at.
So there was protection for all the increasing levels …
Except at the very highest dose. What that suggests to me is that there is the potential for some dose-related toxicity. I don't know what the toxicity would be — whether it's off-target gene silencing or induction of interferons; it's just something you'd have to look at carefully of you were developing this [for use in humans].
With the most-effective dose, how long was the protection lasting for?
We only tested it out for a week. Our gene silencing suggested that it might last up to two weeks, but probably not longer than that.
With these data in hand, are there plans to move forward with this effort?
I think it could be a very effective prophylactic or even treatment for people latently infected with herpes … [and my collaboration with] Baxter is an attempt to think about how you might formulate an siRNA-based microbicide (see RNAi News, 10/30/2008).
These studies are in an acute infection model, and we're basically looking at whether we can block transmission. But we're [also] trying to develop a model where we can actually look to see whether we can treat latent infection, as well. In women who are infected with herpes, even though they don't have symptoms … they are shedding virus continuously … [and] putting their sexual partners at risk.
[ pagebreak ]
One possible use of this would be to treat women who have already become infected to protect their partners from sexual transmission of herpes and to protect them from re-activation.
[Additionally], one of my main interests is applying this [approach] to HIV (see RNAi News, 2/6/2004). To extend this [work] to HIV, you have to transduce different cells, and one thing we're looking at very carefully is in which cells we're getting silencing in the genital tract.
We have some encouraging results that the gatekeeper cells in the genital tract for HIV infection, which are the Langerhans cells or epithelial dendritic cells, are being targeted.
You mention Baxter, but the Cell Host & Microbe paper mentions Alnylam collaborators.
I have a large program-project grant that is funded by the [National Institutes of Health] to develop an HIV microbicide, and Alnylam is part of that grant. So, I'm working with them to do everything to develop an HIV microbicide, and we're using the herpes model as a sort of stepping stone.
That's one [industry] collaboration. The other is with Baxter, and that is more focused on how you would formulate the siRNAs in a way that would be useful clinically. But they are really independent [of each other], and I don't think either Baxter or Alnylam is committed to the clinical development that would be required to make this into a herpes treatment/prophylactic or an HIV microbicide.
Basically, there has never been a microbicide that has gone to market. That scares off a lot of people.
So the paths to market for these kinds of drugs aren't exactly clear.
Right. I'm pursuing this because I think we need to do everything we possibly can to develop a way of preventing the transmission of HIV and other sexually transmitted viruses. That's a commitment of mine. I'm pursing it with NIH support because I haven't found anyone on the commercial side who is interested in developing it.
You might have to start your own company.
[Laughs] I have contemplated that.