Evidence for Disease Susceptibility Differences in Sheep
Like cattle, reported flock-level prevalence of MAP infection varies by breed and location. For an in-depth review of estimated flock-level prevalence of MAP infection in sheep from countries across continental Europe see Nielsen and Toft (2009) and Chapter 1 of this volume.
A retrospective study in India compared the imported Dorset to the indigenous Nilagiri and Sandyno breeds, and reported that the native breeds were more susceptible to infection and had higher mortality rates than Dorset or Dorset crosses (Hemalatha et al., 2013). A more recent study in Australia also reported variability in disease susceptibility between Merino, Poll Dorset, Border Leicester and Suffolk breeds, where the Dorset was also more resistant to MAP than the other breeds (Begg et al., 2017).
4.3.1 Heritability estimates
Few studies have reported heritability for MAP infection in sheep. A single study in 2003 reported heritability estimates of 0.07 for Romney and 0.18 for Merino (Hickey et al., 2003). These estimates are similar to those reported for Holstein and Jersey cattle.
4.3.2 Genome-wide association studies
In contrast to cattle, few GWAS have been published investigating susceptibility to MAP infection in sheep. A recent study in Sarda sheep using the Ovine SNP50K BeadChip, which used serum ELISA to detect MAP infection, identified 32 loci (P < 6 ? 10-4) associated with resistance to MAP infection (Moioli et al., 2016a; Table 4.1). These loci were located across 18 ovine chromosomes and harboured 30 positional candidate genes (Moioli et al., 2016a). The positional candidate genes had common functions pertaining to transcription, metabolism and cell growth. Five of the 30 candidate genes (CD109, ITFG2, PCP4, PRDM2 and SEMA3D) have a direct involvement in the immune system (Moioli et al., 2016a).
4.3.3 Candidate gene studies
While few GWAS have been conducted in sheep, there have been multiple studies investigating candidate genes.
As in cattle, the association of SLC11A1, TRL2 and TLR4 with MAP infection have been explored in sheep. Candidate genes investigated in sheep that haven't been investigated in cattle include CD109 and the major histocompatibility complex (MHC) (Table 4.2).The CD109 gene encodes a protein that binds to and negatively regulates transforming growth factor β (TGFB) (Bizet et al., 2012). During MAP infection, upregulation of TGFB interferes with macrophage activation resulting in less effective macrophage-killing mechanisms (Khalifeh and Stabel, 2004). The coding regions of CD109 were further investigated by Moioli et al. (2016b) in Sarda sheep to identify missense mutations as potential causal variants
in LD with SNPs previously associated with MAP infection. Using this method, a single missense mutation within CD109 (XM_004011463.1) was found to be in LD (P < 0.0001) with the OvineSNP50 BeadChip SNP OAR 8_2 70360.1 (LD = 0.74) previously associated with MAP infection. The authors then determined that LD was maintained between these two mutations across 33 other breeds (P < 0.001), and suggested that OAR8_2 70360.1 could be used to determine sheep susceptibility/resistance to MAP infection through genomic selection (Moioli et al., 2016b).
The MHC class II molecules function to present antigens to the host immune cells. The presentation of the antigens stimulates the immune response and activates various signalling pathways (Holling et al., 2004). Mutations within these molecules have been previously associated with multiple mycobacterial infections in humans including M. tuberculosis (Holling et al., 2004), Mycobacterum leprae (Hashimoto et al., 2002) and Mycobacterium avium (LeBlanc et al., 2000). In sheep, a 2005 study investigated mutations within MHC molecules for association with MAP infection (Reddacliff et al., 2005). This study identified a microsatellite within the MHC locus in a population of Merino sheep that was associated with MAP infection (P < 0.05) when infection status was determined by culturing tissues and faeces (Table 4.2).
The MHC genes have also been associated with MAP infection in cattle, as Weiss et al. (2001) found that MAP infection resulted in downregulation of MHC molecule on the surface of infected macrophages.or multibacillary forms of MAP infection. Paucibacillary infection is when few mycobacteria are present within the gut and only T cells are present within the lamina propria, whereas sheep with multibacillary infections have a high level of bacteria present in the gut and macrophage and B cells are present within the lamina propria (Smeed et al., 2007). The Gossner et al. (2017) study reported that more than 20 pathways enriched for both the pauci- bacillary and multibacillary forms had functions related to inflammation and tissue repair. As would be expected, sheep with multibacil- lary infections had pathways enriched for T cells while sheep with paucibacillary infections did not.
A second IPA also compared healthy sheep with paucibacillary or multibacillary MAP infection. Purdie et al. (2019) used a population of Merino sheep that were experimentally exposed to MAP and used the transcriptome from white blood cells isolated from peripheral blood. This analysis reported that the most common pathways associated for both forms of MAP infection were related to cell growth and proliferation, and lipid metabolism (Purdie et al., 2019). Lipids are an integral component of plasma membranes and they play an important role in phagocytosis of mycobacteria by host macrophages (Gatfield and Pieters, 2000). Supportive of the finding that lipid metabolism is important to MAP infection was a study of Holstein and Red cattle from Australia that also identified a lipid metabolism pathway as important to MAP pathogenesis (Thirunavukkarasu et al., 2014).
4.3.4 Pathway analysis
While GSEA-SNP has not been used to identify gene sets or leading edge genes associated with susceptibility to MAP infection in sheep, several pathway analyses have been conducted using IPA. Gossner et al. (2017) performed a pathway analysis to investigate genes and pathways associated with naturally MAP infected Scottish Blackface/Blackface cross ewes. In this study, the transcriptome of ileocecal lymph nodes from healthy sheep were compared with those from sheep classified as having paucibacillary
4.4