# Analysis of HERV-K (HML2) Expression in Colorectal Cancer Samples

## TL;DR

This study examines the expression of HERV-K (HML2) retrotransposons in colorectal cancer and finds links to epigenetic regulation and cancer subtypes.

## Contribution

The study identifies HML-2 expression patterns and their epigenetic regulation in different molecular subtypes of colorectal cancer.

## Key findings

- HML-2 expression is differentially regulated in colorectal cancer molecular subtypes.
- Inverse associations were found between HML-2 locus methylation and expression.
- HML-2 activation correlates with RIG-I-like receptor signaling in certain CRC subtypes.

## Abstract

Background: HML-2 subgroup mobile genetic elements of the HERV-K family were described to participate in carcinogenesis processes, but their expression and epigenetic regulation in molecular subtypes of colorectal cancer (CRC) remain partly characterized. The present study aimed to evaluate the expression of HML-2 elements using RNA-sequencing data in paired tumor and normal intestinal tissue samples from 63 patients with CRC to identify patterns of the retrotransposons’ activity in different molecular subtypes (CMSs). Methods: RNA-sequencing and DNA methylation data were analyzed for paired CRC and normal tissue samples. HERV-K expression was assessed using three bioinformatics tools: Telescope (version 1.0.3), TEtranscripts (version 2.2.3), GeneTEFlow (version 2020). Molecular tumor subtypes were defined using the CMScaller (version 0.99.2) program. The results of the HML-2 loci expression analysis were supplemented with the HML-2 proteins expression data obtained by quantitative RT-PCR. Results: HML-2 expression assessment by GeneTEFlow (version 2020), TECount (version 2.2.3) and Telescope (version 1.0.3) showed high convergence: the Pearson correlation coefficient for each tool exceeded 0.88. Several HML-2 loci were identified as differentially expressed in CRC samples of different CMS. The PCR results confirmed an increase in HML-2 expression in tumor tissues. For all CMSs, an inverse association was detected between differential methylation of CpG sites and differential expression of HML-2 loci. Associations of HML-2 expressions with differentially expressed genes in which they are located were found, and for a number of such genes an inverse relationship between the expression level and the methylation level of their promoters were demonstrated, and data on the involvement in the pathogenesis of CRC were described: CR1, CD48, TTLL3, ABCC2 and ZNF420. Expression signatures associated with the activity of the RIG-I-like receptor signaling cascade were identified in CMS1–3 CRC samples, which may indicate the possible implementation of viral mimicry against the background of HML-2 activation. Conclusions: Analysis of the expression of HML-2 and its association with CpG methylation contributes to a comprehensive interpretation of the CRC pathogenesis mechanisms.