Quantitative evaluation and multivariate analysis-based ranking of factors influencing blood proviral load in breeding cattle infected with bovine leukemia virus.

Bovine leukemia virus (BLV) infection in cattle results in variable proviral load (PVL), a key indicator of viral transmission and disease progression. However, the relative contributions of host and viral factors to PVL are not fully understood. The present study evaluated four potential determinants of PVL in BLV-infected cattle: two host-related factors (BoLA-DRB3 alleles and age) and two viral factors (Tax amino acid 233 (TAX-233) and 5'LTR nucleotide 175 (LTR-175)) in 1673 Japanese Black (JB) and 979 Holstein Friesian (HF) breeding cows. The contribution of each factor to PVL was quantified using Quantification Theory Type I. BoLA-DRB3 and age accounted for the greatest variation in both JB (55.1 % and 32.4 %) and HF (65.6 % and 18.2 %) cattle, while TAX-233 and LTR-175 made smaller contributions. Notably, the L233 and 175C variants-linked to increased tumorigenicity and transmissibility-were significantly associated with higher PVL in HF but not JB cattle, suggesting breed-specific viral effects. Despite their clinical relevance, these viral markers explained PVL variation to a lesser degree than host factors. These findings underscore the limitations of PVL-based culling strategies, since low-PVL animals may still harbor high-risk viral genotypes. We recommend the integration of viral genotyping into risk assessment frameworks to improve control measures. Although our multivariable models explained only a modest portion of PVL variance (R2 = 5.5-11.7 %), the independent contributions of each factor were confirmed. Future studies incorporating additional host, viral, and environmental variables will be essential to refine predictive models for BLV pathogenesis and transmission.