This article has been updated with additional information from Genia.
Genia Technologies is collaborating with scientists at Columbia University and Harvard University to develop a commercial single-molecule sequencer. The company has licensed a nanopore sequencing-by-synthesis technology developed by researchers at Columbia and the National Institute of Standards and Technology, which it plans to integrate with its nanopore chip platform, and is using polymerase fusion proteins developed at Harvard.
Genia plans to ship its first nanopore sequencing device to beta customers by the end of next year, and to bring a commercial product to market in 2014.
The technology licensed by Genia, called Nano-SBS or NanoTag sequencing, was recently described in a proof-of-concept paper published online in Scientific Reports by the teams of Jingyue Ju at Columbia and John Kasianowicz at NIST.
The approach uses four 5'-phosphate-modified nucleotides that carry tags of different sizes, which are released when a polymerase incorporates a nucleotide into a growing DNA strand. The tags are then fed into a nanopore in the order they were released, where they each produce a unique ion current blockage signal that identifies the DNA base.
Genia's CEO, Stefan Roever, told In Sequence that his firm took an exclusive license to four patents from Columbia and NIST, all of them relating to NanoTag sequencing.
"We plan to offer this chemistry as our preferred sequencing approach on our scalable [integrated circuit]," he said, and to sell chips and reagents supporting it.
The company is also collaborating with the group of George Church, a professor of genetics at Harvard, who will provide fusion protein constructs to couple the polymerase to the entrance of the nanopore.
Genia is currently using a first version of its complementary metal oxide semiconductor nanopore chip to further develop and test the chemistry.
While it will look into alternative approaches for non-sequencing applications of its platform, it has settled on the NanoTag chemistry to develop a DNA sequencer.
Earlier this year, Genia said it is developing a single-molecule electrical detecting platform that uses arrays of alpha-hemolysin nanopores on a semiconductor chip and can work with different chemistries (IS 1/17/2012).
At the time, the firm said that an alpha version of its chip has several hundred nanopores, a number that will increase to tens of thousands in a beta version, and eventually to a million.
The paper published by Ju, a professor of engineering, chemical engineering, and pharmacology at Columbia, and his colleagues at NIST shows proof of principle for several parts of the technology, but not yet for generating sequence data.
The key to the approach is to detect large molecular tags released from the nucleotides during DNA synthesis instead of detecting the nucleotides themselves, which are hard to discriminate by a nanopore because their structure is so similar, Ju explained. Using the tags, "we are no longer constrained by the small differences of the natural nucleotides" because there are "drastic differences in terms of nanopore current blockage caused by the four tags, compared to the native A, C, G, and T."
For their study, researchers in Ju's group attached four polyethylene glycol-coumarin tags of different lengths to the terminal phosphate of 2'-deoxyguanosine-5'-tetraphosphate and tested these nucleotide analogs in a polymerase extension reaction, using the Therminator gamma DNA polymerase. They found that all four G analogs – they did not test analogs of A, C, or T – were incorporated into the DNA with 100 percent efficiency.
Polymerases tolerate modifications on the terminal phosphate of a nucleotide well, Ju said, recognizing and incorporating such analogs readily. The newly synthesized DNA, he pointed out, only contains natural nucleotides, so the reaction can proceed for a long time.
In a separate step, researchers in Kasianowicz's lab analyzed synthetic versions of the released coumarin-PEG tags for their nanopore current blockage effects, after treating them with alkaline phosphatase to reduce the complexity of their charge.
They measured current blockages of a mixture of the four tags passing through a single alpha-hemolysin nanopore and found that each tag produced a blockage that is characteristic of its size. Thus, if each of the four nucleotides carried a different-sized tag, they could be discriminated by the nanopore with high accuracy.
In order to enable actual DNA sequencing with this approach, the researchers will need to couple the DNA synthesis step with nanopore detection, and coupling these elements into an integrated system will be the greatest challenge going forward, Ju said.
They also want to scale up from a single nanopore to many, and Genia's CMOS nanopore array seemed an ideal match for their sequencing chemistry, he said.
According to a nanopore sequencing expert, one problem with the approach, as demonstrated in the paper, is that the nanopore detection experiment was carried out at very high salt concentration, which almost no polymerase can tolerate, so the signal shown "is unrealistic." The polymerase synthesis experiment, on the other hand, was conducted at much lower salt concentration. "Personally, I would have liked to see more of the system working," said the expert, who requested to remain anonymous because Ju might be a reviewer of his work in the future.
The Nano-Tag approach appears similar in principle to the exonuclease nanopore sequencing method championed by Hagan Bayley at the University of Oxford, which Oxford Nanopore Technologies started to develop but later dropped in favor of nanopore strand sequencing. For that, a modified alpha-hemolysin nanopore was to detect nucleotides clipped off from the DNA by an exonuclease.
The problem with that approach was that too many nucleotides escaped the nanopore because they diffused away. "I think Ju's approach will have the same issue, even though he claims that the coumarin-PEG-HN2 [molecules] drift away more slowly since they are a bit bigger," the expert said. "Unfortunately there is no data or estimate on this."
Ju said that Nano-SBS is "drastically different" from exonuclease nanopore sequencing, both because the large tags enter a nanopore more easily and are easier to discriminate than natural nucleotides and because it uses the DNA polymerase reaction, which is used in all currently commercialized next-gen sequencers and has high accuracy and efficiency.
The PEG-tag method is not the only sequencing chemistry from Ju's lab that was licensed by a company for commercialization. He previously developed an SBS chemistry using cleavable fluorescent nucleotide reversible terminators that was licensed by Intelligent Bio-Systems, which was recently acquired by Qiagen (IS 6/26/2012).