RNA sequencing (RNA-seq) in real time, known as the third-generation sequencing at the single-molecule level, is an important technology that will improve our understanding of the human genome. The development of real-time RNA-seq, however, has been challenging, due to the complexity of RNA in sequence and structure that requires a processive reader with single-nucleotide resolution. While Pacific Biosciences (PacBio) can generate long-reads, the process involves cDNA, which loses the informational content of RNA. The only real- time RNA-seq in the current field that does not involved cDNA is the Oxford Nanopore Technology, which is limited to sensing of 5-7 bases of RNA at a time. We report here enzymatic features of 3Dpol, the RNA- dependent RNA polymerase of poliovirus, that are attractive for developing a new RNA-seq technology. We show that 3Dpol copies the RNA template one base at a time with processivity across highly structured RNA. We also show that 3Dpol prefers a hairpin primer to initiate RNA synthesis, generating a double-stranded (ds)- hairpin RNA that allows sequencing of both the template strand and the complementary strand in a nanopore. We further show that 3Dpol, when placed between two electrodes, displays protein conductance that is sensitive to its conformational transition upon NTP-binding. We hypothesize that these features provide the basis to explore 3Dpol for direct RNA-seq with single-nucleotide resolution at the single-molecule level.
In Aim 1, we will determine the ability and quality of 3Dpol as an enzymatic reader of RNA. We will test 3Dpol to read difficult RNA sequences, including sequences that contain post-transcriptionally modified bases, homopolymers, and repeated sequence motifs. We will determine the quality of RNA reading by 3Dpol using the Nanopore device in a 2D (2-directional) platform that sequences both the template strand and the copied strand with the potential to improve accuracy. These studies will also determine error signatures of 3Dpol that are useful for identification of modified bases in RNA.
In Aim 2, we consider that while the intrinsic error rate of 3Dpol is low (10-5), this quality is masked in Nanopore sequencing, due to the latter?s technical error rate (10-15%). We will thus test the possibility to develop 3Dpol in an electronic device for real-time sequencing of RNA by measuring protein conductance through the polymerase. We will engineer 3Dpol to possess two built-in contacts for stable tethering to two electrodes. We will measure protein conductance of 3Dpol in response to NTP binding using a scanning tunneling microscope (STM). If successful, data of STM measurements will support a technology that will generate long-reads of RNA-seq in a solid-state platform that produces direct electronic readout without the need for dyes or labels. This work is at the forefront of exciting development of a new RNA-seq technology that will broadly impact on RNA research and clinical practice.
We aim to explore 3Dpol as an RNA-dependent RNA polymerase in direct and real-time sequencing of RNA. As a processive polymerase across long and highly structured RNA, with electronic protein conductance that is sensitive to NTP-binding, 3Dpol is uniquely suited as an enzymatic reader of RNA. If this potential of 3Dpol is fully developed, we envision that direct RNA-seq by the polymerase can be realized in the clinic with real-time feedback.
