Genome-wide local ancestry and evidence for mitonuclear coadaptation in African hybrid cattle populations

Abstract

The phenotypic diversity of African cattle reflects adaptation to a wide range of agroecological conditions, human-mediated selection preferences, and complex patterns of admixture between the humpless Bos taurus (taurine) and humped Bos indicus (zebu) subspecies, which diverged 150-500 thousand years ago. Despite extensive admixture, all African cattle possess taurine mitochondrial haplotypes, even populations with significant zebu biparental and male uniparental nuclear ancestry. This has been interpreted as the result of human-mediated dispersal ultimately stemming from zebu bulls imported from South Asia during the last three millennia. Here, we assess whether ancestry at mitochondrially targeted nuclear genes in African admixed cattle is impacted by mitonuclear functional interactions. Using high-density SNP data, we find evidence for mitonuclear coevolution across hybrid African cattle populations with a significant increase of taurine ancestry at mitochondrially targeted nuclear genes. Our results, therefore, support the hypothesis of incompatibility between the taurine mitochondrial genome and the zebu nuclear genome.

Keywords: Biological sciences; Ecology; Evolutionary biology; Genetics; Genomics.

Graphical abstract

Figure 1

Geographical patterns of African Bos taurus and Asian Bos indicus admixture in hybrid African cattle populations Map of Africa showing sampled cattle populations and an interpolated synthetic map illustrating spatial distribution of African Bos taurus and Asian Bos indicus admixture. Also shown are two European B. taurus and four Asian B. indicus comparator breeds. Admixture data were generated from the first principal component (PC1) of a principal component analysis (PCA) of microsatellite genetic variation across African cattle populations (Hanotte et al., 2002). Modified from McHugo et al. (2019) under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0). Individual cattle art images modified from Felius (1995) with permission of the author.

Figure 2

Autosomal genomic diversity and admixture in African, Asian, and European cattle (A) Results of the principal component analysis (PCA) for 605 animals from 18 different cattle breeds genotyped for 562,635 SNPs. The PCA plot shows the coordinates for each animal based on the first two principal components. Principal component 1 (PC1) differentiates the Bos taurus and Bos indicus evolutionary lineages, whereas PC2 separates the African and European taurine groups. A histogram plot of the relative variance contributions for the first 10 PCs is also shown with PC1 and PC2 accounting for 58.4 and 17.9% of the total variation for PC1–10, respectively. (B) Unsupervised genetic structure plot for Asian zebu, East and West African admixed cattle, and West African and European taurine breeds. Results for an inferred number of ancestry clusters of K = 3 is shown, which corresponds to Asian Bos indicus (red), European Bos taurus (green), and African B. taurus (blue) ancestral components, respectively.

Figure 3

Haplotype diversity and molecular evolution of the cattle mitochondrial genome (A) Network of 491 cattle mtDNA haplotypes generated using 39 ancestry-informative mtDNA SNPs. This mtDNA haplotype network demonstrates that all surveyed African cattle (47 taurine, 156 East African admixed, and 136 West African admixed) possess Bos taurus mitochondrial genomes. (B) Evidence for positive selection of protein-coding genes in the cattle mitochondrial genome. The significant p values (<0.05) shown in the gene callouts were obtained using the branch-site test of positive selection at the OXPHOS protein genes. B. taurus and Bos indicus values are shown with blue and red cattle icons, respectively. The mitochondrially encoded 12 and 16S RNA genes (RNR1 and RNR2) are also shown in green (some figure components created with a BioRender.com license).

Figure 4

African taurine local ancestry deviations for three different functional gene sets (A) Violin plots of African taurine local ancestry deviations for the HMG, LMG, and NMG subsets with positive deviations indicating retention of African taurine gene haplotypes. Black data points indicate the median values and horizontal lines represent the 95% confidence interval. (B) Boxplot of African taurine local ancestry deviations for the HMG subset in the East African and the West African admixed groups. White lines indicate the median values and yellow and purple boxes indicate the interquartile ranges.