Michael Dean & Nicole Rossi
Analysis of Human Papillomavirus integration sites and gene expression in cervical cancer using long-read sequencing technology
Michael explained how ‘cervical cancer is an international tragedy’, with nearly a thousand women dying each day from the disease. Living in a very low-income country is the most significant risk factor for cervical cancer mortality – suggesting that such mortality is preventable.
Cervical cancer is caused almost exclusively by high-risk human papillomavirus (HPV). The HPV genome encodes two oncogenes, E6 and E7, and via mechanisms that promote host genome instability, the virus integrates into the host cells’ genomes. His team are using long-read nanopore sequencing to study cervical cancer because: these HPV integrations can be complex; the host HLA genes are often somatically mutated in cervical cancer, and this region, and any variants, are difficult to resolve with short-read technologies; and lastly, HPV transcription and splicing are incredibly complex.
Ultimately, their goal is to improve the understanding of cervical cancer, leading to better treatment and accessible treatment in low and middle-income countries (LMICs).
Establishment and validation of whole-genome sequencing and CRISPR-based targeted sequencing
Nicole introduced her work investigating the role of HPV in cervical cancer, using cervical cancer cell lines (SiHa and CaSki) as validation models. She performed low-coverage whole-genome sequencing of SiHa cells using the MinION, obtaining reads up to 100 kb in length, to identify HPV integration in the human genome. Targeted sequencing was then performed using the Oxford Nanopore Cas9-based enrichment protocol, using probes flanking HPV integration sites. With this method, they achieved several hundred-fold coverage at the integration sites, and validated the presence of two identical copies of the HPV genome in the host cell genome.
With this knowledge, they then ‘tackled the far more complicated CaSki cell line, with up to 800 copies of HPV at over 40 genomic locations’. Previous research has shown that CaSki cells contain both full-length and truncated copies of HPV, of unknown size and structure. Nicole stated that ‘our long-read sequences allowed these structures to be viewed for the very first time’, revealing complex arrays of full-length and truncated HPV genomes. Overall, they revealed 31 new human-HPV integration sites. Using their Cas9 enrichment method to perform a ‘truly insane experiment’, which involved cutting CaSki DNA inside all 800 copies of the HPV genome, confirmed human-HPV junctions and detected novel junctions.
Long-read transcriptome analysis of SiHa and CaSki HPV transcripts
Michael introduced his team’s transcriptome sequencing and splicing analysis of their two cervical cancer cell lines: ‘Full-length transcripts would allow us to better understand and quantify these different [HPV splice] products’. Nanopore direct cDNA sequencing was used to investigate this. Michael explains how they were ‘astonished’ to see in CaSki cells that 99% of their HPV-containing reads also contained the same 90 bp segment of human chromosome 6 – suggesting that, despite the large number of integration sites in the human genome, only 1 of them is transcriptionally active. From their transcriptome sequencing data, they resolved full-length HPV transcripts – stating that ‘to our knowledge, this is the first time this has been accomplished in an HPV-positive cell line’. All transcripts were predicted to produce the E7 oncogene, and only 1% expected to produce full-length E6.
The team have a panel of 15 additional cancer cell lines to sequence and analyse, containing 5 different HPV types. They also plan to investigate methylation differences in the cell lines, and perform chromosome conformation capture techniques to investigate changes in chromatin architecture.
Michael concluded his talk by saying that the most exciting part of this research has been being able to perform the sequencing in the lab, thanks to the MinION, meaning that students can carry out and see the whole sequencing process as it happens, in real time.
And finally, he challenges the Nanopore Community to develop a $2-3 HPV test – a cost that would really enable HPV screening to be widely implemented in LMICs in future.