Tech Advance: direct RNA sequencing of the native flu genome
Indirect sequencing of RNA
Ribonucleic acid sequencing can be used to monitor the RNAs present in a sample. So far, the sequencing of RNA has been mostly indirect, being first reverse transcribed into cDNA, which is then analyzed using DNA sequencing methods. Current sequencing-based transcriptomic analyses (RNA-seq) is based on the high-throughput sequencing of complementary DNA (cDNA) and have enabled us to build a more accurate picture of the active transcriptional patterns within organisms. However, these cDNA strands are amplified by PCR, which can introduce bias, such as reduced complexity of the resulting cDNA library, distortion of relative cDNA abundances, and overlooking some RNA species. Additionally, sequences generated with these methods are subject to the processivity (the enzyme’s ability to catalyze) and error-rate limitations of reverse transcription, and during PCR amplification, any modifications on the RNA are lost.
Directly sequence the RNA genome of the flu virus
Although a few attempts have been made to directly sequence native RNA strands, these have involved degrading one chemical base at a time and produced only short read lengths (between 20 bases and 50 bases long). In Nature’s News in Focus section, Ewen Callaway describes how a novel RNA-seq method has been successfully applied to directly sequence the RNA genome of the flu virus. Publishing in bioRxiv, the group that performed this work describe how they modified a nanopore technology system (originally developed by Oxford Nanopore to sequence messenger RNA) to directly sequence viral RNA.
Nanopores and RNA in the Nineties
Since the late 1990s, RNA has been an early candidate substrate for nanopore technologies, which offer a fundamentally different method for sequencing RNA, allowing direct interrogation of original RNA strands. Using the nanopore approach, full length, strand-specific RNA sequences can be generated in real-time, and the method also permits the detection of nucleotide modifications and analogs in RNA.
Because the RNA is single-stranded, a ligation-based approach had to be used to attach molecules (called adapters) that can trigger the sequencing process. Therefore, starting from the poly-A tail present in most RNAs, the Oxford Nanopore team ligated small dsDNA molecules (the adapter) carrying a T10 overhang designed to hybridize with poly-A mRNA and a 5’ phosphate, which then ligates the RNA to create a DNA-RNA hybrid.
Next, a reverse transcription step is used to generate cDNA molecules. Although cDNA is created in the library prep, this is not what is sequenced: this cDNA synthesis step somehow improves the performance of the RNA sequencing (possibly by reducing the intramolecular secondary structure of the RNA).
The resulting mRNA/cDNA complex is then ligated by motor proteins that, through electric potential, load the RNA toward and into the nanopore. The platform consists of single nanopores embedded in an array of thousands of individual synthetic polymer membranes on a single flowcell. When a single RNA molecule is captured in a pore, it is then ratcheted through the pore at a constant rate by an engineered motor protein. This creates perturbations of the nanopore current, which a recurrent neural network converts into base sequences. As an adaptor-ligated RNA enters the pore, the adaptor oligo is detected first, followed by the poly(A) tail, and then the body of the transcript. Because the initial adaptors were ligated to the poly A tail, the RNA molecule is sequenced in the reverse direction (3′ to 5′). When the transcript eventually exits the pore on the opposite side of the membrane, the nanopore current returns to its original high open-pore level.
Viral RNA sequencing using a novel approach
In their effort to sequence native RNA influenza virus, Keller et al. modified the Oxford Nanopore procedure (which was originally designed to sequence mRNA). Because virus RNAs lack a poly-A tail, Keller et al. customized a reverse transcriptase adapter that hybridizes to the 12 conserved nucleotides at the 3’ end of the influenza A virus genome (rather than using the oligo dT). The specificity of this adapter allows efficient sequencing of influenza virus RNA genomic segments from RNA isolated from purified virus particles (control) or from a crude extract that contains many viral and host (e.g., chicken) RNAs. The authors could successfully sequence the complete influenza virus genome with 100% nucleotide coverage, 99% consensus identity, and 99% of reads mapped to influenza. The next step will be to attempt to identify and quantify splice variants and base modifications, which are not practically measurable with current methods.The success behind this experiment implies that the adapter sequence can be modified to target RNAs from other pathogens, specific viral families, genera or species (by extending the target sequence and or by adding degeneracies). The same system could be applied even to compare (+) and (-) sense mRNAs, present during RNA virus infections (such as for influenza viruses). This could help to dissect viral replication processes, as well as host mRNAs activated during an influenza infection. This quantitative sequencing of viral RNA has the potential to directly detect base modifications, splice variants, and transcriptional changes under different replication conditions, such as in viruses used for vaccine production. This is an advantage over the poly-A-based methods, which have a relatively low signal-to-noise ratio due to the presence of host mRNA.
All that glitters is not gold!
If the complete sequencing of the flu RNA using the nanopores technology reveals the potential of this approach, it also identifies specific challenges for future research. It will be essential to improve two major limitations of this technology: the high read level error rate and high input material requirement. Reducing the error rate is essential to enable more accurate sequence determination and to understand nucleotide polymorphisms and genome sub-populations, particularly in viruses that have significant intra-host diversity, such as influenza.
