Columbia Genome Center and Department of Chemical Engineering
Columbia University researchers Jingyue Ju and Nicholas Turro on April 13 published a paper, "Design and Synthesis of a 3'-O-allyl Photocleavable Fluorescent Nucleotide as a Reversible Terminator for DNA Sequencing by Synthesis," in the Proceedings of the National Academy of Sciences.
This development of this method, which utilizes chip technology, is significant in light of the decline of the economic prospects for Applied Biosystems' flagship DNA-sequencing technology (see BCW 4/14/2005) and the thrust to find less costly methods for sequencing DNA. This method is one of several next-generation sequencing methods in development that would be of interest to companies engaged in that sector. And, in fact, the researchers say it has already generated inquiries to the university's tech-transfer office.
The publication comes after October's announcement that Ju and his team received a $1.8-million three-year grant from the National Human Genome Research Institute as part of its "$100,000 Genome" awards to promote the development of a technology that can sequence a mammalian genome at that cost.
BioCommerce Week this week spoke with Ju and Turro, professors of chemical engineering at Columbia, to find out about their plans to further develop and possibly commercialize this technology.
The NIH has put out a lot of grant money out to encourage multi-disciplinary collaborations, but you have created a collaboration organically. How long have you worked together?
Nick Turro: I was the chairman of the Department of Chemical Engineering a few years back, and we had the good fortune to have a position opening up in the Genome Center at the Medical School. We hired Jing and he had a number of projects that involved photochemistry and fluorescence, which is my specialty. And, I have access to graduate students who were interested in these sorts of subjects. So, we began collaborating and writing proposals together. It's quite natural, we admire each other's work, we get along well and we don't live too far from each other in New Jersey, so we can bum a ride home from each other. When NIH reviewed the [Genome] center, they said one of the most interesting things was that everybody seemed to get along, and enjoyed their collaborations.
When did the work first begin on the sequencing technology?
Jingyue Ju: The idea was there when I first came to Columbia and Professor Turro recruited me. The science already connected us to each other. Once I came on board from California, we got together and figured that we could use photochemistry and organic synthesis to solve our problem. The sequencing by synthesis concept has been around a long time. Basically, it's an application to read the sequence out in the context of the polymerase reaction, which is a very efficient reaction to drive cell replication. The cell replication of DNA is literally sequenced multiple times at very high accuracy. We thought about how we could use the synthetic chemistry to dissect this polymerase reaction and make it possible to read the sequence.
Would this method be similar to the methodology involved in microarrays?
Jingyue Ju: Yes, that's right. If you do individual reactions, the throughput is low, but [not] if you match the scale of this polymerase reaction onto the scale of the chip. Where Affymetrix potentially puts millions of individual DNAs onto the surface of a chip, [with this process] you can sequence these simultaneously millions of templates. That is why, potentially, this can give you very high throughput in terms of data generation, but also in terms of accuracy. There is no separation involved; everything is carried out on the surface of the chip.
To get the technology to this point, what has it cost in funds, and labor?
Jingyue Ju: We have trained several post-docs and at least four graduate students, and we have a couple of post-docs that have gone on to take positions at leading universities. Typically, genomic technology development is a long process. Once you have the concept, you validate it, and demonstrate the feasibility. The next level is to refine it to a robust system, involving engineering and so forth. At this point, we are putting our efforts into the molecular level, to address the high-risk aspects of the platform, to make sure that the molecules are working with the enzymes and that the fluorescent dyes can be detected with high sensitivity and can be cleaved with high efficiency. We have addressed these problems, the potential issues that are involved in this platform.
What are the next steps?
Jingyue Ju: After we have worked out the issues at the molecular level, then we need engineering for full-color detections, excitations. We might need industry partners to participate because we have one patent issued, and a couple of patents pending. There are lots of inquiries for the Columbia [tech transfer office] on this particular technology. We can't disclose the details but there is a lot of interest.
How does it compare to other approaches?
Jingyue Ju: The molecular tools we have worked out can be technically used for single-molecule detection because we are dealing with fluorescent molecule labeled nucleotides.
Nick Turro: The notion here is that in the long run, if this works out, you have the potential for doing something that could become a household product in the sense that you could do an analysis on the chip, take advantage of all the chip technology, and take huge amounts of information that would be connected to some computer and go to some database and then come back. That is a concept of how one could do it in the final product five years, 10 years, from this point.
Are you looking to the commercialization of the technology?
Jingyue Ju: We are focused on our research. The Columbia [tech transfer office], has a lot of connections. We let them handle that aspect. We are dealing with the research.
Nick Turro: Columbia is bringing in $100 million a year [from tech transfer]. We have no interest in that. It's not like we are running around with a piece of paper and asking: 'Are you interested?' It's a bit early, but there are folks that can see the potential and want to get involved at some level early on and those are whom our [tech transfer] people are handling. Once you get the patent, you are in good shape to call the shots.
What do you think of the technology?
Jingyue Ju: In terms of the feasibility of this sequencing by synthesis, using this four-color fluorescence detection, we are able to detect the four bases continuously and we have established the feasibility if you put a blocking group on the 3' [end] of the nucleotide and then the polymerase can still incorporate. Subsequently, you can remove it and let the polymerase reaction continue. That is the key to detect this repetitive DNA sequence in the templates.
What would the cost be?
Jingyue Ju: At this point, the cost would be more relevant when you have the whole package ready to decipher the genome. At this point, we don't have an estimate of what the cost is. We know that this can be scaled up to the level of the chip, where you have potentially millions of these different DNA templates. Each cycle you are able to decipher is on the order of something like a million base pairs. You just continue for a few base pairs, 10 base pairs, or 100 base pairs, and you have enormous information generated, just on the chip.
Nick Turro: The view has got to be that looking over the last 20 years what people have been able to do with respect to this sort of project and reducing cost. The final price is so speculative that we don't know how to deal with it. Go back and look at computers. Today, if they aren't running, you can almost throw them out. At that level of scale, one can hope that engineering, volume, and production can bring us to the goal of reducing costs. With the chip format you can do that kind of thing. With other methods, you can't see how to reduce it to that level. We are using light, and spots that are identifiable through standard laser techniques. None of that has to be invented. Now, it is a question of getting the chemistry down right. When that is done, one can suppose that the previous engineering will be part of that.