NEW YORK – Swedish life sciences firm Olink Proteomics has launched a version of its proximity extension assay (PEA) proteomics technology that uses next-generation sequencing for readout of results.
The NGS approach will allow the company to lower assay costs while increasing its throughput and multiplexing capabilities, said Olink CEO Jon Heimer.
Olink will offer the NGS-based readout product, which it has named the Explore 1536, on Illumina's NovaSeq platform. The company is offering it on a fee-for-service basis this year and plans to begin offering external installations in 2021, Heimer said.
PEA uses pairs of antibodies linked to DNA strands that, upon the antibodies binding to their target, are brought into proximity. They are then extended by a DNA polymerase that creates a new sequence that can be used as a surrogate marker for the target protein.
Because two antibodies must bind for a signal to be generated, the assay significantly reduces background, which ups sensitivity and specificity of detection. This enables it to work with very limited sample volumes and also allows for much higher levels of multiplexing than is possible with conventional immunoassays.
To date, the company has used qPCR as the readout for the DNA sequences that serve as proxies for the proteins being detected, but Heimer said that the combination of falling costs for NGS and NGS's ability to enable more rapid and more highly multiplexed analyses had moved the company to develop the new readout.
Use of NGS for read out of proteomics assays like Olink's that use DNA barcodes have been an area of interest for close to a decade. For instance, as early as 2011, Ulf Landegren, a founder of Olink and a professor at Uppsala University, was exploring the use of NGS for readout of Olink's proximity ligation assay, which, similar to PEA, generates DNA molecules when antibodies bind to target proteins.
At the time, Landegren suggested that the single-molecule counting quality of NGS could provide a more precise readout compared to PCR, though Heimer said this week that this was no longer a significant advantage given Olink's efforts to optimize its PCR-based readouts.
"We've optimized our DNA oligo framework to fit with NGS and maybe we can [tweak] that a bit, but I think it will be a limited improvement [in precision] because we have had a high quality in the past, so it is difficult to make improvements in that," he said. "I think the most important things are the throughput, the accessibility to platforms, and perhaps most importantly, cost."
Regarding the last factor, Heimer noted that the company does not expect costs for PCR to decline rapidly the way they have fallen and continue to fall for NGS.
"Five years back, it wouldn't have been cost effective," Heimer said. He added that NGS continued to see considerable technical advances, and that Olink would be able to piggyback on these developments.
He said that use of the NGS technology would allow Olink to further expand the breadth of its proteomic panels. The company currently offers assays to roughly 1,500 proteins. It plans to expand that number to 3,000 next year and to 4,500 in 2022.
Measuring this number of proteins creates complexities that Olink believes NGS can help address, Heimer said.
"You have sensitivity, specificity, dynamic range… then questions like throughput and precision and reproducibility," he said. "If you think about an experiment like, you want to measure the UK Biobank, 500,000 patients with 4,000 proteins each, that has been so far from reality."
He said Olink began doing proof-of-concept work on the idea 18 months ago. He added that the company had put together five of what he called "showcases" demonstrating the use of the technique in different research areas, one of which was SARS-CoV-2.
Olink has a co-marketing deal with Fluidigm through which it offers its protein assays on Fluidigm's BioMark HD rtPCR system, which allows users to measure up to 92 proteins across 96 samples. Heimer said he expected there would still be demand for this product.
"We see on the market many different types of customers," he said. "Some want to look at thousands of proteins in every experiment and the NGS platform enables that for that segment, but then there are a lot of other segments that prefer looking at more targeted questions, so less proteins and less samples, and qPCR is still the preferred platform there."