A Targeted Deep Sequencing Method to Quantify Endogenous Retrovirus Gag Sequence Variants and Open Reading Frames Expressed in Nonobese Diabetic Mice

Endogenous retroviruses (ERVs) are involved in autoimmune diseases such as type 1 diabetes (T1D). ERV gene products homologous to murine leukemia retroviruses are expressed in the pancreatic islets of NOD mice, a model of T1D. One ERV gene, Gag, with partial or complete open reading frames (ORFs), is detected in the islets, and it contains many sequence variants. An amplicon deep sequencing analysis was established by targeting a conserved region within the Gag gene to compare NOD with T1D-resistant mice or different ages of prediabetic NOD mice. We observed that the numbers of different Gag variants and ORFs are linked to T1D susceptibility. More importantly, these numbers change during the course of diabetes development and can be quantified to calculate the levels of disease progression. Sequence alignment analysis led to identification of additional markers, including nucleotide mismatching and amino acid consensus at specific positions that can distinguish the early and late stages, before diabetes onset. Therefore, the expression of sequence variants and ORFs of ERV genes, particularly Gag, can be quantified as biomarkers to estimate T1D susceptibility and disease progression.