At A Glance
Name: Thomas Schmittgen
Position: Associate professor, Ohio State University
Background: Associate professor, Washington State University — 1995-2002; Postdoc, University of Southern California — 1992-1995; PhD, pharmaceutics, Ohio State University — 199;2 BS, pharmacy, Ohio State University — 1985
Originally trained as a pharmacist, Thomas Schmittgen returned to academia to pursue other interests with an eye to oncology. Now an associate professor at OSU, where he began his undergraduate studies, Schmittgen is dedicating some of his time to studying the role of microRNAs in cancer development. Recently, Schmittgen spoke with RNAi News about his work.
How did you get involved with microRNAs?
I had an interest in RNAs stemming back to my postdoc days [when] we were looking at the effects of anti-cancer drugs on RNA processing. So I had kind of a remote interest in the field — I had been studying it on the sidelines.
I heard a talk by [Cold Spring Harbor Laboratory’s] Greg Hannon at the American Association for Cancer Research meeting in San Francisco back in 2002; he gave a kind of sunrise session on RNAi, and I got inspired by his talk. It got me to read the literature a little bit more closely — there was quite a bit out there even at the time of his talk.
Part of my training also was in these so-called translational research studies where you look at patient specimens as a way to study cancer as a better model system than things like cell lines. I really got interested in microRNAs, and wanted to know if there were differences in expression levels between cancer and normal tissue. At the time I first got interested in this there wasn’t anything known about [whether] microRNA levels were expressed differently in cancer versus normal [tissue], but there was a lot known about microRNAs involved in development and differentiation, so I had a hunch that there might be something going on there. That’s really how I got started.
Where did this hunch lead?
A lot of my expertise was in real-time quantitative PCR. Everything that was done at the time was using Northern blotting to detect microRNAs, which is a good method because there are these various precursor molecules that get progressively smaller as you get to the active, mature microRNA. Northern blots will allow you to distinguish that, but what they don’t allow you to do is [a] high-throughput, sensitive type of analysis when you’re working on clinical specimens from patients — [in this situation], you don’t have all 30 micrograms or 100 micrograms of RNA to work with. So what I wanted to do was take my expertise in real-time PCR and apply it to microRNA.
That was shortly after this [AACR] meeting and I’ve been working on that.
Can you talk in detail about your work?
Originally what I wanted to do was use PCR to quantify the mature microRNA, and we quickly found out that that was going to be really challenging because it’s only 22 nucleotides long — it’s basically the size of a PCR primer. We have some ideas that we’re still going to try to follow up [on] and we’ve tried some things, but it’s been a real challenge.
So what we decided to do instead of amplifying the mature microRNA is to amplify the precursors. There are two main precursors: There’s something called the primary precursor, which is probably in an intron or maybe transcribed along with the pre-messenger RNA … That’s processed to something called the precursor microRNA, which is a 75-nucleotide hairpin.
We decided to quantify these precursor molecules as a way to predict the mature [microRNA ] levels. There are some challenges with that because these do exist as these imperfect hairpins … and I think most people, when they try to do PCR, generally stay away from those areas. Many people probably believe that it really couldn’t be done, so we designed primers to these hairpin precursors and showed that it was possible to do. We published a paper in Nucleic Acids Research — it came out in February of 2004 — demonstrating the technique and showing some of the things that need to be done in terms of using gene-specific priming and various changes in the reverse-transcription step, which was really what we found to be the critical thing to get this to work. What we also showed in that paper, which I think was a little bit different, is that we were able to take our data and covert it to red-green pseudocolors similar to microRNA data.
Most people would agree that you wouldn’t want to ever replace PCR as a way of screening gene expression, but the nice thing about these microRNAs is that right now there’s 200 known human microRNAs and there’s predicted to be about 250. So, we can screen the precursors using a much more sensitive and specific assay, and get the same type of data that you’d get from a microarray — you could do the same clustering, you’d get the same red-green-black pseudocolors. That’s where we’re at now with this.
How does all this tie in to cancer?
What we were funded for [under a recent National Cancer Institute grant] … is to take this assay and use it as a way to screen microRNA precursor expression in cancer versus normal tissue.
Again, the idea is to get clinical specimens. The grant was funded to look specifically at leukemia and prostate cancer, [but] we also have some other collaborations going. [We’re going] to do screening, see if there is altered expression of these different microRNA precursors in cancer versus normal [tissue]. That can tell you a lot of information.
The most interesting point of this is that for many years people have been looking at coding messenger RNAs as being indicative of being important in cancer — are they going up or down and how does that relate to the development of cancer? The interesting thing about these microRNAs is: First of all, there’re non-coding RNAs, so they’re not present in any of [collections] of microRNA that are out there, even the Affymetrix array with the 30,000 genes. The other thing is that they also would not have been picked up in these cloning experiments when people would pull out polyadenylated mRNA and see if there is altered expression in cancer versus normal [tissue].
What’s known about the mechanism of action of [microRNAs] is that they regulate gene expression at the level of translation, so the microRNA then binds to its target, which is messenger RNA, and causes a decrease in translation of that protein, probably without changing the mRNA transcript level. This would have been another thing that would have been lost during these messenger RNA screens because the changes of the messenger RNA are probably going to be the same, but if you were to go back and look at protein, the protein levels could be reduced.
All of this that I’m talking about now I can’t take the credit for, because there’s been a number of groups — Carlo Croce at Thomas Jefferson [and] Phil Sharp’s group at MIT — have written about some of this in review articles and those types of things.
Where do you see this work headed? What are the implications?
I think there’s a lot of exciting things that could come out of this. What we hope to find are microRNAs that have either increased or decreased expression in cancer versus normal [tissue], and either way it could be interesting. [Genes] can be decreased — through all the common ways you would think a gene would be decreased in cancer; a gene is deleted, a transcription is reduced — or they can be increased — there’s gene amplification or there’s increased expression.
The key here [is] trying to identify the target [of a microRNA]. If the target was normally a growth-promoting messenger RNA, then in the natural setting you might have a microRNA that’s reducing the translation of that protein so there’d be less growth. If that microRNA suddenly has reduced expression — again, either through a deletion or reverse transcription — it’s going to remove the brakes on that messenger RNA and you’re going to see an increase in that protein, which is going to lead to increased growth.
The opposite situation [is] if you had a tumor-suppressor RNA [and] maybe there’s a microRNA binding site for that messenger RNA that is not there for some reason. Now all of a sudden, if that microRNA was to [have] increased expression, it would put the brakes on the tumor suppressor; the tumor suppressor isn’t going to work, and it could cause cancer.
At this point, [the implications] are all the traditional things that one would think of — using it as a diagnostic would certainly be a possibility, [using it] to help us understand cancer formation, and third … using it as a therapeutic.
I think this is a really interesting thing because … since we’re dealing with a microRNA, which is a fairly small molecule compared to an entire gene or protein, it is possible that we could either … add a microRNA back to a cell or … go after its target with siRNAs. It may allow us to identify target genes, target messenger RNAs, in a somewhat different way than the traditional way of screening for the messenger RNA.
Is this therapeutic angle something you’d think about getting involved in directly?
Yes, I think so. It’s definitely something we’re interested in. At this point, we’re at the discovery and screening [stage].
Is there anything else you’re working on?
We’re trying to come up with better delivery methods, better ways to stabilize siRNAs, just like a lot of people, but we don’t have a lot of data at this point. We’re trying to think of the next step if we come up with something that looks interesting. We’d like to first validate the target, if it’s a suspected target of [a] microRNA, and then try to reintroduce it or block it.
What about diseases beyond cancer?
I’ve been working in the cancer [field] for most of my career, but … I’ve gotten a lot of interest in our Nucleic Acids Research paper. Sometimes people [say to me], “I’m working on this,” and there’s been more than one person that’s mentioned fat metabolism, so I suspect that’s something that others are working on.
The other interesting area is the neural science area. The greatest diversity of microRNA expression is in the brain, and I would suspect that the microRNA is important to the development and differentiation of different tissues, so therefore it would certainly be a candidate for something that’s involved in disease — but nobody has shown anything like that.
I don’t want it to sound like I’m the only one doing this because that’s not true … but the advantage of our assay is that PCR is very sensitive and specific, and many of these microRNAs are in families of isoforms of very similar sequence. We’re going to be able to quantify those different isoforms. For example, the let-7 family has 14 members that [have] very closely related sequences. The advantage of our assays is that we can do the high-throughput screening, we get very quantitative data, it’s very sensitive and specific.
The disadvantage of our assay is we’re not able to quantify the mature microRNA, which is the active [form]. One of the things we’re trying to do is … understand the level of the precursor versus the mature — are the precursors truly predictive of the mature [microRNA]? If that’s the case, we have a really good method. If it’s not the case, then ultimately the arrays may be a better way to go.