Meta-analysis of heritability estimates and genome-wide association for tick-borne haemoparasites in African cattle

Abstract

The control of tick-borne haemoparasites in cattle largely relies on the use of acaricide drugs against the tick vectors, with some vaccination also being used against selected pathogens. These interventions can be difficult in Africa, where accessibility and cost of vaccines can be issues, and the increasing resistance of tick vectors to the widely used acaricides is a complication to disease control. A potential complementary control strategy could be the exploitation of any natural host genetic resistance to the pathogens. However, there are currently very few estimates of the extent of host resistance to tick-borne haemoparasites, and a significant contributing factor to this knowledge gap is likely to be the difficulty of collecting appropriate samples and data in the smallholder systems that predominate livestock production in low- and middle-income countries, particularly at scale. In this study, we have estimated the heritability for the presence/absence of several important haemoparasite species (including Anaplasma marginale, Babesia bigemina, Babesia bovis, and Ehrlichia ruminantium), as well as for relevant traits such as body weight and body condition score (BCS), in 1,694 cattle from four African countries (Burkina Faso, Ghana, Nigeria, and Tanzania). Heritability estimates within countries were mostly not significant, ranging from 0.05 to 0.84 across traits and countries, with standard errors between 0.07 and 0.91. However, the weighted mean of heritability estimates was moderate and significant for body weight and BCS (0.40 and 0.49, respectively), with significant heritabilities also observed for the presence of A. marginale (0.16) and E. ruminantium (0.19). In a meta-analysis of genome-wide association studies (GWAS) for these traits, two peaks were identified as reaching the suggestive significance threshold (p < 1.91 × 10^-7 and p < 1.89 × 10^-7, respectively): one on chromosome 24 for BCS and one on chromosome 8 for the E. ruminantium infection status. These findings indicate that there is likely to be a genetic basis that contributes to pathogen presence/absence for tick-borne haemoparasite species, which could potentially be exploited to improve cattle resistance in Africa to the economically important diseases caused by these pathogens.

Keywords: African cattle; genome-wide association studies; heritability; meta-analysis; tick-borne haemoparasites.

FIGURE 1

Map showing the major sampling locations across the four African countries (Burkina Faso, Ghana, Nigeria, and Tanzania). Sampling locations are represented by circles, with size and colour changing according to the numbers of animals.

FIGURE 2

Manhattan plot (A) displaying the meta-analysis results (-log₁₀ (p) of the corresponding p-values) and Q–Q plot (B) of observed p-values against the expected p-values for the body condition score. Genome-wide p < 0.05 (red line; p < 9.52 × 10⁻⁹ corresponding to -log₁₀ (p) of 8.02) and suggestive (blue line; p < 1.91 × 10⁻⁷ corresponding to -log₁₀ (p) of 6.72) significance thresholds are also shown.

FIGURE 3

Manhattan plot (A) displaying the meta-analysis results (-log₁₀ (p) of the corresponding p-values) and Q–Q plot (B) of observed p-values against the expected p-values for the Ehrlichia ruminantium infection status. Genome-wide p < 0.05 (red line; p < 9.42 × 10⁻⁹ corresponding to -log₁₀ (p) of 8.03) and suggestive (blue line; p < 1.89 × 10⁻⁷ corresponding to -log₁₀ (p) of 6.73) significance thresholds are also shown.

FIGURE 4

Manhattan plot of the regions of interest on chromosome 24 for the body condition score (A) and on chromosome 8 for the Ehrlichia ruminantium infection status (B). The purple diamond in the plots represents the variant with the smallest p-value in the areas of interest. Other variants in the area are coloured according to their LD (r ²) with the target variant.