Skip to main content
Premium Trial:

Request an Annual Quote

Direct Analysis of tRNA Abundance, Modifications Enabled by New Nanopore Sequencing Method


NEW YORK – Researchers from the Centre for Genomic Regulation in Barcelona, Spain and their collaborators have developed a new method to study the abundance and modifications of transfer RNA (tRNA) using direct nanopore sequencing.

Described in a proof-of-concept study published last month in Nature Biotechnology, the method, named Nano-tRNAseq, illustrates the potential of nanopore sequencing for cost-effective and high-throughput analysis of tRNA.

Transfer RNAs play a critical role in protein synthesis, serving as a link between messenger RNA (mRNA) and the amino acids that form proteins.

"They are the most abundant RNA molecules in the cell," said Eva Maria Novoa, a group leader at the Centre for Genomic Regulation, who is the corresponding author of the study.

Despite their biological importance, Novoa said, the field still lacks a simple and cost-effective method to systematically investigate tRNA abundance and modifications. While short-read sequencing-based approaches have allowed researchers to study tRNAs in recent years, they require reverse transcription of tRNA into cDNA, which often creates technical difficulties, she added.

For instance, tRNAs are generally very difficult to reverse transcribe due to their structure and modifications. In addition, tRNA modifications are typically erased during the cDNA synthesis step.

To avoid these shortcomings, Novoa's lab turned to nanopore sequencing, which enables direct analysis of RNA molecules. However, her team still had to overcome a few technical hurdles to optimize the Oxford Nanopore platform for tRNA sequencing.

For one, the nanopore platform can produce "very noisy" data when it comes to sequencing short molecules, often leading to a 5' end read drop out of the tRNA. To cope with that, her group ligated RNA adapters at both the 5′ and 3′ ends of the tRNA molecules to ensure a good signal. In addition, the researchers came up with a strategy to linearize the tRNA molecules to prevent their folding structure from inferring with sequencing.

As for bioinformatics, Novoa pointed out that MinKnow, Oxford Nanopore’s proprietary analysis software, was found to misinterpret tRNAs as adapters in its default settings, erroneously discarding most of the tRNA reads. Moreover, the software was also biased toward longer tRNA molecules, skewing the analysis results. "If we would have used the default nanopore parameters, not only would we have thrown away most of our library, but we also would have had biased quantification of tRNAs," Novoa said.

To alleviate these issues, her team devised a custom MinKnow configuration by lowering the strand minimum duration and the adapter maximum duration. This resulted in a 12-fold increase in the number of mapped tRNA reads and accurate recapitulation of tRNA abundance compared to the default parameters.

The researchers applied Nano-tRNAseq to Saccharomyces cerevisiae cultures that were exposed to various environmental conditions. The method allowed them to effectively study cross-talk and interdependencies between different modification types within the same tRNA molecule and changes in response to oxidative stress.

"This is a very good paper," said Tao Pan, a biochemistry professor at the University of Chicago who was not involved in this study. "It opens a new direction, with future applications to follow."

Pan's lab has been developing tRNA sequencing methods using both nanopore and Illumina platforms. While he agreed that nanopore sequencing may have theoretical advantages for studying tRNAs, he believes the technology is still not mature enough for widespread use.

For instance, the current pores used in Oxford Nanopore's platforms are not designed for RNA sequencing but optimized for DNA molecules. In addition, the lack of training data for tRNAs makes the analysis pipeline "suboptimal," he said.

"Nanopore [sequencing] is fun and new and exciting, but it's too early to be used practically for anything [related to tRNA] yet," Pan noted. "For the next few years, we will be living with Illumina sequencing for tRNA."

Previously, Pan's group developed a method named multiplex small RNA-seq library preparation (MSR-seq), an Illumina sequencing-based approach that enables multiplex sequencing of small RNA molecules, including tRNAs. It also allows for the enzymatic and chemical treatment of RNA on beads, thus accommodating the analysis of modifications.

Pan said MSR-seq, which was patented by the University of Chicago, has been licensed to a Cambridge, Massachusetts-based company named HC Bioscience to develop tRNA-based therapeutics. He and his collaborators are also about to launch a separate company to commercialize MSR-seq into tRNA sequencing kits that can be utilized widely by researchers, he said.

According to Novoa, one of the limitations of Nano-tRNAseq is that the algorithm can only label tRNA modifications as basecalling errors at this point because they are not trained into the Oxford Nanopore basecalling model. Similarly, the method cannot directly predict which tRNA modifications are dysregulated unless they are compared with modifications in a reference that has been previously characterized by mass spectrometry-based approaches.

"What we are lacking for nanopore [sequencing] is a basecaller for all the RNA modifications," Novoa said. "Instead of basecalling for four letters, we should be able to basecall many more letters," which is now possible for nanopore DNA sequencing, she added.

Generating enough training data for tRNA basecallers can be a challenge, she said, given that tRNAs have many types of modifications, some of which are not easy to synthesize chemically.

On top of that, Novoa thinks Oxford Nanopore has "lagged a bit behind" in the RNA sequencing realm. Currently, RNA sequencing is only commercially available on the company's flow cells with the older R9 pores, she said, compared to DNA sequencing, which has already moved to a new generation of pores. Even if Oxford Nanopore releases a new pore for RNA sequencing, the existing basecallers will likely have to be retrained based on the new signal profiles, she added.

Still, Nano-tRNAseq can pave the way for efficient and cost-effective tRNA analysis, she said, opening the door to potentially exploiting tRNAs as biomarkers for therapeutics and other applications.

Moving forward, Novoa said her team is interested in trying out Nano-tRNAseq on a broader range of organisms and cell types. Additionally, her lab hopes to expand the method to other small noncoding RNAs beyond tRNAs.

The researchers also continue improving their method and eventually want to bring it to market. They have filed a patent for Nano-tRNAseq and are in discussions on licensing the technology out or possibly establishing a company.

"We really believe in the method," Novoa said. "That's why we're pursuing these [commercial] opportunities."