Retroviruses, like human immunodeficiency virus type-1 (HIV-1) and human T-cell leukemia virus type-1 (HTLV-1), integrate DNA copies of their genomes directly into the host cells’ genomic DNA, forming a provirus. Integration allows these proviruses to coopt the molecular machinery of host cells to produce more virus, but the proviruses are subject to the same epigenetic factors that regulate non-viral genes on that chromosome. Understanding the regulation of transcription of these proviruses is valuable to both basic research and treatment of retroviruses that cause life-threatening illness.
Chromatin immunoprecipitation followed by massively parallel sequencing (ChIP-seq) has the potential to be an excellent tool for understanding the regulation of provirus transcription. However, these proviruses are only ~9 kb, integrated within the ~3.0 billion bp human genome. Performing detailed expression analysis on such a small proportion of the total DNA requires further enrichment of the target sequences beyond the ChIP enrichment. This secondary enrichment must be done without introducing bias into the ChIP-seq data. It must also be customizable to account for the variety of provirus strains and sufficiently robust to sequence variations within the provirus genome.
Using IDT xGen® Lockdown® Probes, Miyazato et al. designed biotinylated target capture probes against the most frequent HIV-1 and HTLV-1 subtypes prevalent in Japan. Probes of 120 bp were used to create probe panels with 60 bp tiling. The HIV-1 probe set contained 161 probes, and the HTLV-1 set contained 148 probes.
Prior to provirus sequence enrichment, chromatin immunoprecipitation (ChIP) was performed in HIV-1 and HTLV-1 samples, using anti-CTCF, anti-H3K4me, anti-H3K9ac, and anti-H3K36me antibodies. DNA libraries were then prepared for high-throughput sequencing on Illumina® MiSeq® or NextSeq® systems.
NGS library samples used for targeted enrichment were then blocked from unwanted hybridizations using Cot-1 DNA (Thermo Fisher) and xGen Blocking Oligos (IDT). Blocked samples were resuspended in hybridization buffer and enriched using the relevant HIV-1 or HTLV-1 probe sets. The researchers compared results from target enriched samples to results obtained from sequencing non-enriched ChIP libraries.
When analysis was performed on non-enriched ChIP samples, >99.99% of the reads were of the host genome and not useful for provirus analysis. However, probe-based, target enrichment increased provirus reads several hundred to several thousand fold over non-enriched samples. The researchers also found that the data for the enriched ChIP libraries correlated well with the non-enriched samples, indicating that any bias in the data was not from probe-based enrichment.
Important to the study of proviruses, the authors also found that the xGen Lockdown Probes tolerated a level of mismatched bases. They found that mismatches of ≤7 bp in probe sequences from the HTLV-1 probe set, and of ≤8 bp in the HIV-1 probe set, did not significantly affect the capture process.
The authors concluded that this target capture method improved analysis of provirus sequences within the genomic context and that the method was easily customizable for the provirus sub-types they studied. They also noted that they were able to order enough of each probe panel to perform ~100 capture reactions at an affordable price, and that their method of probe-based retroviral enrichment after ChIP enrichment should be useful for other applications.