Using a target amplification method it developed, Dutch startup Cergentis promises to sequence any genomic locus of interest in full, identifying not only point mutations but also structural variants.
The company was founded in the spring of 2012 as a spinout from the Hubrecht Institute, a research institute of the Royal Netherlands Academy of Arts and Sciences, to develop and commercialize the technology, called targeted locus amplification, or TLA.
According to Max van Min, the company's CEO and one of its co-founders, Cergentis developed the patent-pending technology but was "inspired by" work previously done at the Hubrecht Institute, which is also affiliated with the University Medical Center Utrecht.
"We immediately realized it has such an important range of applications in genetic research and diagnostics, [so we] felt it was justified to set up a company, get funding, and work on commercializing and developing the technology," van Min said.
The firm, which has 10 full-time and part-time employees, is based in an incubator at Utrecht University. So far, it has raised an undisclosed amount of funding from private investors — both individuals and institutions — and has received a number of grants from Dutch government programs, including the Netherlands Genomics Initiative and the Cancer Genomics Center.
The current funding is "more than sufficient for us to be able to develop and commercialize the technology," said van Min, who gave a presentation about TLA two weeks ago at Hanson Wade's NGS for Cancer Drug Development conference in Boston.
TLA employs crosslinking to connect DNA areas that are in close proximity to each other and then uses a single locus-specific primer pair to pull out and amplify the target gene.
The company’s current protocol starts with cells as input material — it has not yet determined the minimum amount of material required. A crosslinking agent is then added to link areas of genomic DNA that are adjacent to each other.
After that, the DNA is either sheared or cut with a restriction enzyme, generating what van Min referred to as "hairballs" of crosslinked DNA fragments, where most of the DNA originates from the same locus. After diluting the hairballs, the DNA fragments are ligated, resulting in DNA circles.
A locus-specific primer pair targeting the region of interest is then added to amplify those circles that contain the target DNA, followed by sequencing. The target sequence is then reconstructed from the reads.
Because the crosslinking and ligation steps connect DNA strands at random that happen to be next to each other, each DNA circle is a little different. But DNA that is located close to the primer sequence is much more likely to end up in the circle than DNA that is farther away, so the target will have better read coverage than more distant regions.
For sequencing, Cergentis has mostly used Illumina's MiSeq so far, although it has also sequenced on "various Life Tech instruments" as well as the HiSeq. It does not have its own sequencing instrument right now but has been using those of customers and partners.
"The fundamental importance of the technology is that it provides complete sequence information of any region of interest in a hypothesis-neutral manner, meaning that irrespective of what the point mutations or structural variations in the regions of interest might be, the TLA technology always allows you to amplify it," van Min explained.
This, he said, sets TLA apart from other capture or multiplexed PCR methods, where "you capture what you know to be there, but anything that you were not aware of will simply not be amplified or captured." In addition, TLA, unlike other methods, enables haplotyping, he said.
While researchers wanting to use TLA still need to know what genes or regions to target — and at least one unique sequence in there — they will get out the entire sequence, including both SNPs and insertions or deletions.
Van Min said the technology is very suitable for the analysis of a few and up to tens of candidate genes, and the company has worked on methods to automate the process to accommodate multiplexing and large sample numbers.
While the technology has "a very broad range of applications" in areas ranging from plant genomics to microbial genomics to genetic diagnostics, the ability to detect both point mutations and structural variations might be especially useful for applications in cancer, where "a considerable percentage of clinically relevant mutations occur as a result of structural variants that conventional [capture] techniques don't pick up," van Min said.
In collaboration with a number of research institutions, for example, Cergentis has sequenced the BRCA1 gene, which is about 81 kilobases long, from patient-derived xenograft tumor samples, using a single unique primer pair. They were able to identify structural variations in the gene that relate to variations in drug response, he said.
In addition, Cergentis is applying TLA to leukemia research to detect gene fusions, for example in the MLL gene. In this area, TLA could become an alternative to fluorescence in situ hybridization assays, he said, enabling the detection of several gene fusions in a single assay.
Another application of TLA, which Cergentis did not initially have in mind, is the characterization of transgenes. Once a transgene has been inserted into an organism, "it can be difficult to identify where it has ended up, and to ascertain whether it's still intact," he explained, and TLA is able to provide that information.
A few months ago, Cergentis started offering TLA as a service and has customers at academic institutions as well as pharmaceutical companies, including in the Netherlands, the UK, Belgium, and the US. The cost of the service is "comparable with other approaches to do targeted sequencing," van Min said.
However, the company does not intend to become a large service provider and is currently in discussions with other service providers to include its technology in their portfolio.
Cergentis is also "in the final stages" of automating its technology, he said, using conventional liquid handling instrumentation. This will allow the firm to sell reagent kits, which it is currently beta-testing and plans to launch in the near future. Kits will be for defined set of genes, and there will be both catalog and customized versions, van Min said.
In collaboration with several research institutes, the company is about to submit a paper for publication describing its methods in detail, and it plans to publish a study relating structural variations in genes to drug response, and another one on using TLA to sequence the HIV genome.
Going forward, Cergentis is exploring new applications for TLA, including working with formalin-fixed paraffin-embedded samples and combining it with bisulfite sequencing for targeted methylation analysis.