Search

Additional file 5: Phylogenetic tree of all HML-2 proviruses that have complete env sequences and that integrated since the human-chimpanzee divergence (~ 5 million years ago).

Additional file 1: Figure showing interruptions of ORFs among HML-2 proviruses in the human reference genome that integrated since the human-chimpanzee divergence.

The cell-surface attachment protein (Env) of the HERV-K(HML-2) lineage of endogenous retroviruses is a potentially attractive tumour-associated antigen for anti-cancer immunotherapy. The human genome contains around 100 integrated copies (called proviruses or loci) of the HERV-K(HML-2) virus and we argue that it is important for therapy development to know which and how many of these contribute to protein expression, and how this varies across tissues. We measured relative provirus expression in HERV-K(HML-2), using enriched RNA-Seq analysis with both short- and long-read sequencing, in three Mantle Cell Lymphoma cell lines (JVM2, Granta519 and REC1). We also confirmed expression of the Env protein in two of our cell lines using Western blotting, and analysed provirus expression data from all other relevant published studies. Firstly, in both our and other reanalysed studies, approximately 10% of the transcripts mapping to HERV-K(HML-2) came from Env-encoding proviruses. Secondly, in one cell line the majority of the protein expression appears to come from one provirus (12q14.1). Thirdly, we find a strong tissue-specific pattern of provirus expression. A possible dependency of Env expression on a single provirus, combined with the earlier observation that this provirus is not present in all individuals and a general pattern of tissue-specific expression among proviruses, has serious implications for future HERV-K(HML-2)-targeted immunotherapy. Further research into HERV-K(HML-2) as a possible tumour-associated antigen in blood cancers requires a more targeted, proteome-based, screening protocol that will consider these polymorphisms within HERV-K(HML-2). We include a plan (and necessary alignments) for such work.

Additional file 10: Comparison of results of three different methods for measuring the relative provirus abundance in the JVM2_growth2 FASTQ file.

Additional file 2: Comparison of our results to those from whole transcriptome datasets in the Short Read Archive (SRA).

Additional file 7: Guidance on future proteomic work.

Additional file 10: Comparison of results of three different methods for measuring the relative provirus abundance in the JVM2_growth2 FASTQ file.

Additional file 4: Figure illustrating problem in mapping reads to proviruses that have similar sequences.

Additional file 3: Ion Torrent run summaries.

Additional file 6: Further details of other published RNA expression studies and further discussion of ours.

Additional file 2: Comparison of our results to those from whole transcriptome datasets in the Short Read Archive (SRA).

Additional file 6: Further details of other published RNA expression studies and further discussion of ours.

Additional file 3: Ion Torrent run summaries.

Additional file 7: Guidance on future proteomic work.

Additional file 4: Figure illustrating problem in mapping reads to proviruses that have similar sequences.