Associate professor, chemistry/biochemistry, City College of New York
• Assistant professor, chemistry/biochemistry, City College of New York — 2002-2009
• Postdoc, Columbia University — 1996-2003
• PhD, organic chemistry, University of Rochester — 1996
• MS, organic chemistry, University of Rochester — 1993
Last month, a team from City College of New York reported on the development of circularized DNA strands that can act as RNA polymerase III templates for small RNA generation in the absence of a promoter. Lacking the potential hazards of viral vectors, such oligos “maintain small RNA information in a stable form that RNAP III can access in a cellular context with, in some cases, near promoter-like precision and biologically relevant efficiency,” according to the researchers.
Recently, Gene Silencing News spoke with the study's senior author, Kevin Ryan, about the work.
Let's start with background on you and your research interests.
We are in the chemistry department at City College, and I'm in the biochemistry part of the faculty. We primarily study RNA — anything involving RNA is of interest to us, although we also have another program in which we study the biochemistry of the sense of smell. But mostly we do a lot of RNA stuff.
For this project, we were thinking about the purpose for most DNA, [which] is to serve as a template for making RNA. You boil it down, that's what DNA does. But in fact, only one strand of the double-stranded DNA carries out this function: the template strand. The basic finding of our paper is that when we synthesize this one template strand, which encodes an RNA hairpin … [and put it in a] circularized [form], this DNA oligonucleotide can enable human RNA polymerase III to make an RNA copy, to transcribe it.
What's astounding is that there is no promoter. In fact, there can't be a promoter; this is single-stranded DNA. The dogma is that you need a transcriptional promoter to precisely start transcription at a specific point in the genome. We don't have one of those, yet somehow RNA polymerase III seems to have an affinity for certain sequences and secondary structures in the context of a circularized oligonucleotide. And we call these “coligos” for “circularized oligos.”
How did you come up with the idea to circularize these things?
When I was a graduate student in 1993 for Eric Kool, who is now at Stanford [University] but was then at the University of Rochester, Eric and I found that bacteriophage RNA polymerases would, at low frequencies, grab onto a circularized oligonucleotide. We found that it would transcribe around and around; it can't run off the end and it has no termination sequence. It would be like an old-fashioned mimeograph, and we called this rolling circle transcription.
This was before the discovery of microRNAs, before the discovery of RNA interference. These circles were small and have very little information, and they weren't really useful. But once microRNAs and siRNAs were discovered, I wondered what would happen if a human RNA polymerase [was used since] these are much more complicated, more highly regulated enzymes than bacterial enzymes.
I thought I'd make a human cell extract and just expose the circularized DNA, and I made a circular DNA that encodes a pre-microRNA. It was microRNA-122 that we started with. We did not see any rolling circle transcription, and figured that it's not going to work. But what we did see was that some enzyme — at the time we did not know which of the polymerases did this — made a precise transcript. It appeared to go around once.
We were surprised because there was no promoter, [and asked] what enzyme it was and why was it doing this. And could it be used as a tool to make RNA in human cells. DNA has a lot of favorable attributes as a potential therapeutic: it's chemically stable, it's more stable than RNA in human bodily fluids, and if you circularize it, it is even more stable because exonucleases can't get started because there is no 5' or 3' end. It's [also] simpler to synthesize, and definitely less costly to synthesize [than RNA]. During the 1990s, the antisense research companies perfected making kilogram amounts of synthetic DNA.
So DNA itself can carry the information, and that's one of the reasons we're so interested in this. We have no way of delivering it — I don't want to mislead, delivery is still a problem — but solutions that deliver RNA might be applicable to our circularized DNA oligos.
Is there additional work going on in your lab now to take these findings farther? Is that delivery issue something you're exploring?
Delivery is something we want to try, but we have cases where we have very precise transcription. We have other cases where it is less precise. So we have not yet learned how to completely control the starting and stopping. We know it's general, but the generality with respect to the precision [is not yet worked out]. We think those rules exist, and we are trying to get funding to continue this project.
We want to see if we can use it to knock down mRNAs on a general basis. Can we replace something like viral vectors? Just imagine, instead of using a virus or plasmid to generate an shRNA or an siRNA, we would use a circularized DNA to carry that information into a cell, then allow a cell's own RNA polymerase to do the difficult step of making the RNA. We have not studied its use in RNAi yet, but if we make the coligos carefully, we should be able to make a substrate for further processing [by] Dicer.
It's a new way of generating small RNAs in human cells. Most companies are making formulations of RNA with some chemical modifications to increase stability. But, at the heart of it, is the information. You need the information for the sequence that you want in a cell, and DNA is just fine with coding for that information. We found a way to unlock it from chemically synthesized DNA.
We're definitely hoping for therapeutic applications, but that is in the future. We need funding to pursue that.