Extensive differential DNA methylation between tuberculosis skin test positive and skin test negative cattle.

Bovine tuberculosis (bTB), caused by Mycobacterium bovis (M. bovis), represents a significant problem for the agriculture industry as well as posing a risk for human health. Current diagnostic tests for bTB target the cell-mediated immune (CMI) response to infection with M. bovis, primarily through screening of animals with the tuberculin skin test. Epigenetic modifications have been shown to alter the course of the immune response and differentially methylated regions (DMRs) might also influence the outcome of the skin test in cattle. Whole Genome Bisulphite Sequencing (WGBS) was used to profile DNA methylation levels from peripheral blood of a group of cattle identified as test positive for M. bovis (positive for the single intradermal comparative tuberculin test (SICTT) and/or the interferon-γ release assay compared to a test negative control group [n = 8/group, total of 16 WGBS libraries]. Although global methylation profiles were similar for both groups across the genome, 223 DMRs and 159 Differentially Promoter Methylated Genes (DPMGs) were identified between groups with an excess of hypermethylated sites in SICTT positive cattle (threshold > 15% differential methylation). Genes located within these DMRs included the Interleukin 1 receptor (IL1R1) and MHC related genes (BOLA and BOLA-DQB). KEGG pathway analysis identified enrichment of genes involved in Calcium and MAPK signalling, as well as metabolism pathways. Analysis of DMRs in a subset of SICTT negative cattle that were IFN-γ positive showed differential methylation of genes including Interleukin 10 Receptor, alpha (IL10RA), Interleukin 17 F (IL17F) and host defence peptides (DEFB and BDEF109). This study has identified a number of immune gene loci at which differential methylation is associated with SICTT test results and the degree of methylation could influence effective host immune responses.

The impact of changing the cut-off threshold of the interferon-gamma (IFN-γ) assay for diagnosing bovine tuberculosis in Ireland.

In Ireland, the interferon-gamma (IFN-γ) assay is routinely used as an ancillary test interpreted in parallel with the single intradermal comparative tuberculin test (SICTT) to maximize the detection of bovine tuberculosis (bTB) infected animals. Up until 2018, a positive test result was recorded in the IFN-γ ELISA assay following whole blood stimulation with purified protein derivative (PPD)-bovine (B), PPD-avian (A) and nil sample (N), using the interpretation criteria, B-N > 50 optical density units (OD), B > 100 and B-A > 0. Following a review of available data, the threshold of the B-A component changed to B-A > 80. As predicting the impact of changing the cut-off thresholds for the IFN-γ test de novo is challenging, the aims of this study were to follow animals that initially tested negative using the new IFN-γ assay interpretation criteria and investigate their future risk of disclosure with bTB, with a focus on animals that otherwise would have been removed when using the older interpretation criteria (0 < B-A ≤ 80). Enrolled animals (n = 28,669 cattle from 527 herds) were followed up for two years (2019-2021), or to point of bTB detection or death. At the end of follow-up, 1151 (4.0%) of enrolled animals were bTB cases. The majority of these cases were diagnosed using SICTT (80.5%). The cumulative number of positive animals that would have been removed if the old cut-off (0 < B-A ≤ 80) was used amounted to 1680 cattle (5.9% of the enrolled cohort). Of these, 127 (7.5%) were diagnosed with bTB during follow-up. In contrast, 1024 of the 1151 cattle which subsequently tested positive during the study period following a negative IFN-γ test would not have been identified with the old or new IFN-γ cut-off criteria. Survival analysis showed that animals that would have been removed under the old interpretation criteria were at increased risk of a positive diagnosis with bTB during follow-up compared to other test negative animals. A newly developed risk prediction model (using a Cox proportional hazard model) showed that age, animal number of SICTT tests, number of inconclusive SICTT tests, B-A (IFN-γ assay), B-N (IFN-γ assay), animals from store herds and the percentage of the rest of the herd that were positive during the breakdown were statistically significantly associated with bTB detection. However, inclusion of the IFN-γ OD variables did not show added value in terms of prediction performance of the model.