Complete Anopheles funestus mitogenomes reveal an ancient history of mitochondrial lineages and their distribution in southern and central Africa

Abstract

Anopheles funestus s.s. is a primary vector of malaria in sub-Saharan Africa. Despite its important role in human Plasmodium transmission, evolutionary history, genetic diversity, and population structure of An. funestus in southern and central Africa remains understudied. We deep sequenced, assembled, and annotated the complete mitochondrial genome of An. funestus s.s. for the first time, providing a foundation for further genetic research of this important malaria vector species. We further analyzed the complete mitochondrial genomes of 43 An. funestus s.s. from three sites in Zambia, Democratic Republic of the Congo, and Tanzania. From these 43 mitogenomes we identified 41 unique haplotypes that comprised 567 polymorphic sites. Bayesian phylogenetic reconstruction confirmed the co-existence of two highly divergent An. funestus maternal lineages, herein defined as lineages I and II, in Zambia and Tanzania. The estimated coalescence time of these two mitochondrial lineages is ~500,000 years ago (95% HPD 426,000-594,000 years ago) with subsequent independent diversification. Haplotype network and phylogenetic analysis revealed two major clusters within lineage I, and genetic relatedness of samples with deep branching in lineage II. At this time, data suggest that the lineages are partially sympatric. This study illustrates that accurate retrieval of full mitogenomes of Anopheles vectors enables fine-resolution studies of intraspecies genetic relationships, population differentiation, and demographic history. Further investigations on whether An. funestus mitochondrial lineages represent biologically meaningful populations and their potential implications for malaria vector control are warranted.

Figure 1

Map and phylogenestic relationships of 43 An. funestus mitogenomes. (A) Map indicating the collection sites for 43 An. funestus samples, created using ArcGIS v10.5.1 (www.esri.org). (B) Bayesian maximum clade credibility phylogeny of complete mitogenomes from the 43 An. funestus samples of the best fitting model (GTR + G + I, Bayesian skyline plot, and a relaxed molecular clock) inferred using BEAST2. Samples are colored by geographic origin: blue indicates Zambia (N = 28), orange indicates Tanzania (N = 10), green indicates DRC (N = 5). Divergence dates (median estimates and 95% HPD) are given in parenthesis for major nodes. Posterior probabilities > 0.5 are indicated at each node. The timescale is indicated below the tree and is in years before present.

Figure 2

Haplotype network of 43 An. funestus mitogenomes. In this TCS network, each node indicates a haplotype, with nodes colored according to origin. The number of mutational steps between nodes are indicated in parentheses beside the line connecting one node to another. One group of samples (all lineage II) did not connect to the main cluster within 95 mutational steps (over a 95% confidence limit for connectivity): these are shown in the box in the lower right. There are two distinct groups within the main cluster (lineage I): one more highly clustered (cluster A), and another with fewer, more-distant nodes (cluster B). Cluster A and B in lineage I are separated by ≥42 mutations. The size of each node indicates the number of samples sharing a specific haplotype.