NEW YORK – Researchers from the Children's Hospital of Philadelphia and the University of Pennsylvania are considering multiple options for the commercialization of a new, cheaper method for targeted long-read RNA sequencing.
It's still early days for TEQUILA-seq (Transcript Enrichment and Quantification Utilizing Isothermally Linear-Amplified probes in conjunction with long-read sequencing), which can help create panels for interrogating full-length RNAs. After publishing a proof-of-concept paper in August in Nature Communications, the inventors are exploring both a spinout and licensing opportunities.
"All those options are on the table," said Lan Lin, an RNA researcher at CHOP Research Institute, cosenior author of the paper, and a coinventor on a patent application covering the method. "We're open for any opportunity."
"This is a very powerful technology with a lot of utility," said bioinformatician Yi Xing, cosenior author of the paper and coinventor on the patent, noting that interest in the method since the paper has come out has been "very high."
"The protocol will have a lot of nice applications for tissues with complex isoform usage, like neurological and cancer samples, as well as better mapping for developmental questions and the impact of a treatment," such as chemotherapy, said Chris Mason, a sequencing expert at Weill Cornell Medicine who was not involved with the study.
TEQUILA-seq is not the first method for targeted long-read RNA-seq, but it reduces the cost of sequencing only RNAs of interest. Commercially available capture probes can cost $800 per sample for even a modest 150 gene panel, with a minimum cost of $13,000 for 16 reactions. And other academic protocols are labor intensive.
Using unmodified oligos, TEQUILA-seq is able to generate large amounts of capture probes. Specifically, it uses single-stranded oligo pools that tile across the target transcript and Klenow fragment DNA polymerase to create probes with biotinylated uridines. Those probes then hybridize to the target cDNA and help capture them. The captured cDNA molecules are also amplified before sequencing. "For the same 150-gene panel, with a few thousand dollars for the initial template oligo pool and reagents we can generate 10,000 reactions," Xing said. "So, you can make the per-reaction cost less than $1."
TEQUILA-seq was developed with Oxford Nanopore Technologies' method in mind, but could also work with Pacific Biosciences' long-read sequencing. With ONT technology, the total cost of sequencing RNA could be as low as $100, Lin said. Most of that comes from library prep; multiplexing samples and running them on a higher-throughput PromethIon sequencer can further reduce costs.
Mason suggested that costs could be higher than presented in the paper, "but it's still a pretty good deal for what you get," he said.
In their paper, the authors compared the efficiency and enrichment of TEQUILA-seq to xGen Lockdown probes from Integrated DNA Technologies in a panel of 10 human genes known to express long transcripts. Both methods achieved on-target rate of 85 percent with similar fold enrichment and "nearly identical fold enrichment for each target genes," they wrote.
TEQUILA-seq could be used in current applications of long-read RNA-seq, such as research into isoforms, RNA splicing, and translation of RNA into protein. "For the studies we want to do, we want the flexibility of trying different gene panels for different projects," Lin said. "16 reactions is really nothing and can be used up in a few weeks."
The largest panel Lin's lab has created covers approximately 3,500 genes, while the smallest was about 10 genes.
Moreover, the process could also be used for any targeted sequencing application, long- or short-read; DNA or RNA. "It's not limited [to] long-read sequencing, that's just a usage we're interested in," Lin said.
In the paper, the researchers used TEQUILA-seq to identify isoforms associated with certain breast cancer subtypes, including HER2-enriched tumors. In this study, they looked at so-called "tumor aberrant" isoforms — molecules with an elevated presence in some, but not all breast cancer cell lines. They identified 635 such isoforms in 256 genes, about two-thirds of which were novel. Complex or combinatorial alternative splicing was the cause for nearly 70 percent of these tumor-aberrant isoforms, they reported.
"There are specific splicing events that are really complex for certain genes," Lin explained. "With short reads, you won't be able to see those."
TEQUILA-seq could help open up a world of isoforms that has been simply too expensive to explore. Lin pointed to a paper published in 2020 that looked at complex isoforms of a single gene, the ion channel CACNA1C. "Those are usually large transcripts with lots of exons," Lin said. The study used long-range PCR to enrich and sequence 241 novel transcripts for that single gene.
"If you're interested in larger genes, especially with lots of exons, you're basically guaranteed to have a lot of alternative splicing events," she said.