Sequence diversity in cancer-causing viruses

From the Mullins lab, University of Washington and the Vaccine and Infectious Disease Division

Kaposi sarcoma-associated herpesvirus (KSHV), also known as Human Herpesvirus 8, is a gamma herpesvirus that is often harmless, but causes a cancer called Kaposi sarcoma (KS) in a subset of infections. People infected with HIV, especially those who are not treated with antiretroviral therapy, are disproportionately affected by KS. Although immune suppression is associated with KS, the specific factors that determine KSHV progression to cancer are poorly understood.

A putative driver of differential KS outcomes is viral genetic variation. KSHV encodes oncogenes that can dysregulate cell cycle, cell-to-cell adhesion, and inflammation; therefore, it is plausible that de novo mutations in these regions may cause some KSHV strains to be more cancer-inducing than others. However, there is currently insufficient whole genome sequencing data available to accurately investigate the relationship between KSHV sequence variation and KS disease severity. Likewise, the dynamics of intra-host KSHV diversity is unresolved, as previous research has reported conflicting results regarding the diversity of KSHV strains within an individual. Additionally, PCR artifacts can be introduced during sequencing, leading to over-estimation of KSHV genetic diversity and further confounding the field. To clearly resolve the level of KSHV diversity that exists both between and within KS patients, more sensitive sequencing technologies are needed.

Clement Santiago (Mullins lab, University of Washington) and Dr. Jason Goldman (University of Washington, Fred Hutch Vaccine and Infectious Disease Division) investigated KSHV genetic diversity by applying an existing, highly accurate short-read sequencing method called “duplex sequencing” to KSHV whole-genome sequencing. Duplex sequencing incorporates duplexed unique molecular identifiers (dUMI)— double-stranded strings of random base pairs that serve as barcodes on individual DNA molecules during PCR—to create dUMI-consensus reads of each DNA molecule that significantly reduce PCR-associated errors, allowing the true diversity of KSHV strains to be better illuminated. After validating this technique on a KSHV lab strain, the authors obtained KSHV samples from tumors and oral swabs from nine participants with HIV-associated KS. They recently published this work in PLoS Pathogens.

The dUMI-consensus reads revealed very few point mutations in KSHV sequences in both tumors and oral swabs, demonstrating that intra-tumor KSHV diversity is low and providing no evidence of KSHV quasispecies, suggesting that KSHV infections result from single strains and do not evolve extensively within an infection. However, in some cases, the genomic structure of KSHV differed markedly between tissues. Aberrant genomes structures and unique point mutations in certain genes were found in tumors but not in the oral swabs from the same individual, suggesting that mutated KSHV genomes may be associated with cancerous lesions.  Likewise, the aberrant sequences made up most of the sequences found within a tumor, suggesting that they may have non-randomly propagated to high copy numbers and exacerbated tumor growth. Because KSHV encodes immunomodulatory and anti-apoptotic factors, these mutations could contribute to KS cancer progression, and these findings support the idea that KSHV genetic variation may drive cancerous progression. However, it is also possible that these changes occurred subsequent to tumor growth.

Duplex sequencing pipeline for KSHV genomes.
Duplex sequencing pipeline for KSHV genomes. Figure from publication.

Interestingly, the authors found the same KSHV genome aberration in multiple tumor lesions within the same individual. This result led the authors to hypothesize that aberrant KSHV genomes may possibly be spread by metastasis, a phenomenon that has not been previously reported in KS. Further research is needed to determine if infected tumor cells can seed distal KS lesions and exacerbate cancerous progression.

This work demonstrates that duplex sequencing is a useful whole genome sequencing pipeline for KSHV that eliminates PCR-associated errors and provides more accurate views of intrahost viral populations. "We found that all KSHV genomes from oral swabs and tumors of the same person were virtually identical at the point mutational level, but that tumors can harbor KSHV genomes with structural variations,” explained the authors. “Interestingly, we found certain viral genes and KSHV genome regions are commonly impacted by intra-host mutations, and we also found that multiple tumors from the same person can have the same rearranged KSHV genome," the authors summarized. Further research with larger KS cohorts, is now ongoing to determine the prevalence of these mutations and if they are associated with KS cancer progression.

This work was supported by the National Institutes of Health, including funding from the University of Washington Centers for AIDS Research Retroviruses and Molecular Data Sciences Core.

UW/Fred Hutch Cancer Consortium members James I. Mullins, Warren Phipps and Corey Casper contributed to this work.

Santiago JC, Goldman JD, Zhao H, Pankow AP, Okuku F, Schmitt MW, Chen LH, Hill CA, Casper C, Phipps WT, Mullins JI. Intra-host changes in Kaposi sarcoma-associated herpesvirus genomes in Ugandan adults with Kaposi sarcoma. PLoS Pathogens. 2021 Jan 19;17(1):e1008594. doi: 10.1371/journal.ppat.1008594.