At A Glance
Name: Vivek Mittal
Position: Assistant professor, Cold Spring Harbor Laboratory
Background: BS, chemistry/biology, St. Joesph’s College, Darjeeling — 1984; PhD, molecular parasitology, Jawaharlal Nehru University — 1994; Postdoc, Cold Spring Harbor Laboratory —1994-1999
Despite receiving degrees while in India, Vivek Mittal decided to continue his research on Long Island at Cold Spring Harbor Laboratory because of its strength in basic molecular biology. After completing his post-doctoral studies there, he got the opportunity to stick around as an assistant professor.
Recently, Mittal spoke with RNAi News about his work in cancer and where RNAi comes in.
How did you get involved with RNAi?
When I started my laboratory, I wanted to study how tumors neovascularize — how they set up vasculature. To do that, we were using gene expression profiling [with] DNA microarrays. We obtained some very significant targets, which are involved in making blood vessels, specifically for the tumor. Now, we’re interested in inhibiting these targets to see if lost gene function can phenocopy the vasculature formation.
RNAi has been there for a little while, but the ability of RNA to suppress gene expression in mammalian systems was not there at the time [we were doing expression profiling work] because of the interferon response.
Greg Hannon, who is my colleague here at Cold Spring Harbor [Laboratory] — his lab and other laboratories around the country, have made a significant impact on understanding how RNAi works in mammalian systems. And because of the biochemistry they’ve done, the genetics they’ve done, we are now able to very, very effectively suppress any gene you want to — this is a very powerful method to do that, and that’s the kind of method we wanted because it was not just looking for a knockout mouse or using very slow, traditional gene silencing methods.
We have worked with a lot of genes at the same time, a lot of targets at the same time because we do a lot of DNA microarrays, so you get a lot of targets coming out. [In order] to be able to suppress each of these targets efficiently, we thought RNAi was the way to go.
And so, we dabbled in RNAi a little bit and we got some good hits. Then the [problem] was: When we’re trying to suppress some of the genes, it was lethal for the cell — these were essential genes for cell survival or proliferation or development. So I thought: ‘We need something called inducible RNAi.’
We know that the short hairpin is transcribed by a pol III promoter. I had worked on a pol III promoter during my postdoc … so I knew the structure really well. What I did was, I used that knowledge and hence, we were able to modify the U6 promoter and make it interactive.
This is in cell culture right now, but I think it can be adapted to a mouse model pretty easily.
Can you talk a bit about how inducible RNAi works?
There are various kinds of inducible [RNAi] for gene expression, which have existed for a long time. The most common ones are tetracycline-inducible systems and the ecdysone-inducible systems. I have worked with the ecdysone system before to express a gene and … we and a group of other people found that the tetracycline system was leaky in some cell lines. We found that the ecdysone system is pretty tight, and that’s what you need if you want to have a very robust inducible system. So I used the ecdyson system.
The way it works is there are two transcription factors called RXR — it’s a retinoid x receptor — and the other is VgEcR. They are considerably expressed in a cell line, [but] to be an active transcription factor, they have to heterodimerize. The heterodimerization is carried out by the addition of an inducer molecule, which is called muristerone — it’s an analog of ecdysone. So when you [add] muristerone, these two transcription factors, which are inactive otherwise, heterodimerize and become active. They bind to the ecdysone-responsive element … and transcribe whatever you want them to transcribe.
In our case, we transcribed [an] activator, which is … a GAL4-DNA-binding domain fused to a pol III activator called Oct-2. This activator, in turn, binds to the U6 promoter, which we have modified. [For] the U6 promoter, we ha[d] to move the natural enhancers and replace them with GAL4-binding sites.
Now, the [GAL4-Oct2] binds to the [GAL4-DNA-binding sites and activates the U6 promoter] and transcribes the hairpin RNA.
Based on how much inducer you add, we’ve shown that [the gene silencing effect] is dose-dependent — if you increase the amount of inducer you get more hairpin production and more gene silencing. It’s also time-course dependent.
[These] are all hallmarks of an inducible system. It’s tight — in the absence of an inducer there’s no leakiness. It’s time-course dependent — with time you see more gene silencing. It’s also dose-dependent — [the effect varies] depending on the amount of inducer you add.
This was really where we showed [inducible RNAi] in both human and mouse cell lines. In the human cell line, we showed that if you use a p53 hairpin you can suppress p53 expression really well. In the mouse cell line, we suppressed expression of MyoD. We also used a couple of other genes in the lab, which we have not published before.
And you think you can get this inducible RNAi silencing to work in mice?
We haven’t done it, but what you need to do is make a mouse which expresses the receptors.
Is this something you’re working on?
There is a lot of interest in the field. There are people who have approached me — these people do a lot of mouse work and want to do it as a collaboration.
It’s not a major [focus]. We are getting into some other things in the lab now, but definitely we will do that in collaboration with others.
These are academics?
What other sorts of things are you doing with RNAi right now?
We’re trying to suppress genes in vivo. So, we are using stem cells and things like that.
The retroviral vectors we use are not very good at infecting pluripotent cells. So what we [have done] is generate lentiviral vectors. We will transduce those ex vivo and then transplant them in the mouse and see how they function in vivo.
What genes are you looking at?
Some of the targets we discovered using gene chips. Some of the genes are … pro-angiogenic genes, which are involved in supporting angiogenesis.
What’s the goal of all this? Therapeutics?
First, we want to understand the mechanism in vivo — how does this really work? First, the goal is to dissect the signal transduction pathway … or the signaling cascade involved in angiogenesis.
The second is: Can we deliver [RNAi molecules] as therapeutics? Can we suppress genes?
So far, people are limited to targeting cell surface receptors by small drug molecules. Now with RNAi, you really can suppress any gene you want to, even if they’re inside the cell.
What are some of the key difficulties to this?
One of the key hurdles is definitely going to be delivery, right? The question is: How is it going to [work] in vivo? You can use different retroviral vectors and stuff, but for humans … using viruses to deliver hairpins, there [might be] some problems.
Remember, there was this case where people tried to cure [an immunodeficiency] using [retroviral gene therapy]. What they found was that two kids developed leukemia. So, there are some potential dangers.
But people have not tried lentiviruses, and because lentiviruses don’t integrate into promoters, insertion mutagenesis might be a little less [of a risk]. You can also deliver hairpins using adenoviruses, but the thing is that adenoviruses don’t degrade, therefore the problem is that you have to administer adenoviruses from time to time.
Have you gotten any interest in your work from anybody in industry?
For the inducible system, we have. It’s generated a lot of interest. We’re getting overrun with requests for reagents. Four different big companies want to license the technology.
Lots of academic people want to license it, [too].
The advantage of the system is that you could deliver tissue-specific promoters. Our system induces things indirectly — you can just change the promoter and make it cell specific.
Are you in negotiations to license to any of those companies?
Cold Spring Harbor has been dealing with three or four different companies right now.
I don’t know if I can tell you, but probably Stratagene.
What about interest in your cancer research?
We are devoted to understanding the mechanisms of tumor angiogenesis. I think soon we’ll have some interesting results. So far, we have gene expression profiling resolved, and now we’ve moved on to combine gene expression profiling and RNAi together to be able to dissect the pathways.
It’s a little bit early, but things are moving.