In this work we have first estimated the heritability of gastrointestinal nematode burden in the Dutch Holstein-Friesian dairy cattle population. Parasite load was quantified using faecal egg counts (EPG) in animals of at least two years of age. In a first data set, h2
estimates for EPG obtained were relatively high, of the order of 0.20. Stratifying the parasites by species using faecal larval counts did not increase h2
, with the noticeable exception for Haemonchus
sp. Although Haemonchus
sp. can infect cattle (e.g., [50
]), this high h2
is puzzling, because until now evidence for a genetic component of host resistance to this ovine parasite has not been found. In general, herd effects were of the same magnitude as corresponding h2
, including for Haemonchus
sp larval counts. The significant herd effect in case of Haemonchus
sp can in part be explained by the fact that this species (H. contortus
) only occurs in cattle if sheep are present on the farm. In a second data set, however, h2
for EPG, estimated using both an animal model and a half-sib design (e.g. [41
]), were less than 0.10. This drop may be due to the use of a faster but less sensitive egg counting procedure in this cohort.
We then conducted a whole genome scan to identify QTL affecting faecal egg count. The study was certainly amongst the largest conducted to date as it involved the phenotyping of over 4,000 animals and genotyping of more than 750 within-family extremes. It is worthwhile noting, however, that other QTL mapping studies targeting parasite resistance typically used experimentally challenged animals and/or relied on repeated phenotypic measurements of, for instance, faecal egg counts, which is expected to increase the heritability of the analyzed traits (e.g. [27
]). The genome scan revealed two chromosome segments with strong evidence for QTL, respectively on BTA9 and BTA19. Although these findings certainly await independent confirmation in cattle, it is noteworthy that in sheep, Crawford et al. [29
] detected suggestive QTL influencing faecal egg counts at the orthologous positions on OAR8 (BTA9) and OAR11 (BTA19), while Beh et al. [27
] detected a suggestive QTL for faecal egg count at the orthologous OAR11 position. These coincident findings considerably strengthen the support for the detected QTL.
In addition we report six loci with suggestive within-family evidence for QTL affecting faecal egg count. This information might be useful for comparison with results of future studies. Note that because our experimental design was primarily based on linkage analysis in paternal half-sib families, we have not included markers on the X chromosome in this study [51
We have attempted to refine the map position of the BTA19 QTL using a denser map and by exploiting LD. This effort did not result in substantial gains in terms of signal strength and resolution. This is possibly due to the fact that the marker density was still insufficient to effectively capture an LD signal (Fig. ). In particular, this might be the case if the segregating QTL alleles are old and have therefore had ample opportunity to recombine with flanking markers. Alternatively the QTL may be characterized by multiple, individually rare causative alleles that would be more difficult to detect by association mapping.
Genome-wide SNP arrays with a marker density that is at least one order of magnitude higher than what was used in the present study (e.g. [52
]) are becoming available in livestock including cattle. It may be worthwhile to genotype the pedigree material used in the present study and attempt to find additional QTL by performing genome-wide association studies. One advantage of association-based designs is that they extract information from the maternal chromosomes whereas the daughter design [32
] only uses information from the paternal chromosomes albeit in a very robust manner. LD-based approaches thus have the potential to double the amount of usable information.
The strongest signal for the BTA19 QTL was obtained in an interval flanked by markers rs29027283 and rs29013747. The corresponding chromosome segment measures ~3.3 Mb in the bovine (Btau4.0). Ninety-five genes are annotated to the orthologous genome segment in human (Additional file 3
). At least one of these is worth mentioning: ITGAE
. This gene codes for an integrin alpha chain that is preferentially expressed on the surface of intestinal intraepithelial T lymphocytes, which are thought to be important for immune surveillance and immune responses to mucosal pathogens [53
It is noteworthy that BTA19 harbours several QTL influencing a range of phenotypes of economic importance in dairy cattle, including milk yield and composition and fertility (Coppieters et al., unpublished observations). The impact of selecting for increased parasite resistance on other traits will thus have to be evaluated carefully prior to any MAS attempt.