The sustainability of malaria control in Africa is threatened by the rise of insecticide resistance in Anopheles mosquitoes that transmit the disease1. To gain a deeper understanding of how mosquito populations are evolving, we sequenced the genomes of 765 specimens of Anopheles gambiae and Anopheles coluzzii sampled from 15 locations across Africa, identifying over 50 million single nucleotide polymorphisms within the accessible genome. These data revealed complex population structure and patterns of gene flow, with evidence of ancient expansions, recent bottlenecks, and local variation in effective population size. Strong signals of recent selection were observed in insecticide resistance genes, with multiple sweeps spreading over large geographical distances and between species. The design of novel tools for mosquito control using gene drive will need to take account of high levels of genetic diversity in natural mosquito populations.

The population structure of Anopheles gambiae in Africa was studied using 11 microsatellite loci in 16 samples from 10 countries. All loci are located outside polymorphic inversions. Heterogeneity among loci was detected and two putative outlier loci were removed from analyses aimed at capturing genome-wide patterns. Two main divisions of the gene pool were separated by high differentiation (F(ST) > 0.1). The northwestern (NW) division included populations from Senegal, Ghana, Nigeria, Cameroon, Gabon, Democratic Republic of Congo (DRC), and western Kenya. The southeastern (SE) division included populations from eastern Kenya, Tanzania, Malawi, and Zambia. Inhospitable environments for A. gambiae along the Rift Valley partly separate these divisions. Reduced genetic diversity in the SE division and results of an analysis based on private alleles support the hypothesis that a recent bottleneck, followed by colonization from the NW populations shaped this structure. In the NW division, populations possessing the M rDNA genotype appeared to form a monophyletic clade. Although genetic distance increased with geographic distance, discontinuities were suggested between certain sets of populations. The absence of heterozygotes between sympatric M and S populations in the DRC and the high differentiation in locus 678 (F(ST)>0.28) contrasted with low differentiation in all other loci (-0.02

Plague, a zoonosis caused by Yersinia pestis, is still found in Africa, Asia, and the Americas. Madagascar reports almost one third of the cases worldwide. Y. pestis can be encountered in three very different types of foci: urban, rural, and sylvatic. Flea vector and wild rodent host population dynamics are tightly correlated with modulation of climatic conditions, an association that could be crucial for both the maintenance of foci and human plague epidemics. The black rat Rattus rattus, the main host of Y. pestis in Madagascar, is found to exhibit high resistance to plague in endemic areas, opposing the concept of high mortality rates among rats exposed to the infection. Also, endemic fleas could play an essential role in maintenance of the foci. This review discusses recent advances in the understanding of the role of these factors as well as human behavior in the persistence of plague in Madagascar.

Anopheles funestus is a primary vector of malaria in Africa south of the Sahara. We assessed its rangewide population genetic structure based on samples from 11 countries, using 10 physically mapped microsatellite loci, two per autosome arm and the X (N = 548), and 834 bp of the mitochondrial ND5 gene (N = 470). On the basis of microsatellite allele frequencies, we found three subdivisions: eastern (coastal Tanzania, Malawi, Mozambique and Madagascar), western (Burkina Faso, Mali, Nigeria and western Kenya), and central (Gabon, coastal Angola). A. funestus from the southwest of Uganda had affinities to all three subdivisions. Mitochondrial DNA (mtDNA) corroborated this structure, although mtDNA gene trees showed less resolution. The eastern subdivision had significantly lower diversity, similar to the pattern found in the codistributed malaria vector Anopheles gambiae. This suggests that both species have responded to common geographic and/or climatic constraints. The western division showed signatures of population expansion encompassing Kenya west of the Rift Valley through Burkina Faso and Mali. This pattern also bears similarity to A. gambiae, and may reflect a common response to expanding human populations following the development of agriculture. Due to the presumed recent population expansion, the correlation between genetic and geographic distance was weak. Mitochondrial DNA revealed further cryptic subdivision in A. funestus, not detected in the nuclear genome. Mozambique and Madagascar samples contained two mtDNA lineages, designated clade I and clade II, that were separated by two fixed differences and an average of 2% divergence, which implies that they have evolved independently for approximately 1 million years. Clade I was found in all 11 locations, whereas clade II was sampled only on Madagascar and Mozambique. We suggest that the latter clade may represent mtDNA capture by A. funestus, resulting from historical gene flow either among previously isolated and divergent populations or with a related species.

There were epidemic-epizootics of Rift Valley Fever (RVF) affecting humans and cattle in Madagascar in the district of Anjozorobe in 2008. Little is known about the role of Malagasy mosquitoes in the circulation of RVF virus. Therefore, we investigated the species diversity, dynamics and biology of potential RVF virus vectors in the rainforest, rainforest edge (village of Anorana), and savanna biotope (village of Antanifotsy) of this district between November 2008 and July 2010. We captured 56,605 adults of 35 different species. Anopheles squamosus (Theobald), Anopheles coustani (Laveran), Culex antennatus (Becker), Culex pipiens (L.), and Culex univittatus (Theobald) were the most abundant during the rainy season with Cx. pipiens the most abundant species in the rainforest (47%), and An. squamosus the most abundant species in the rainforest edge and in the savanna biotope (56%, 60%, respectively). Only Cx. univittatus was abundant in the dry season. The parous rate was > 60% throughout the rainy season for An. squamosus and it was > 50% from the middle to the end of the rainy season for Cx. pipiens. Two additional species have been found only at larval stage. Cattle were the most attractive bait for all species, followed by sheep and poultry. Human was the least attractive for all species. Most of the 163 bloodmeals tested were taken from cattle. Three were from poultry, one was from dog and one was a mixed bloodmeal taken from sheep and cattle. These results on vectorial capacity parameters may allow considering the involvement of mosquito transmission of the virus in the district of Anjozorobe during the recent epidemic-epizootic.

Point mutations in the voltage-gated sodium channel gene involved in knockdown resistance to DDT and pyrethroid insecticides have been described in several insect species. In the malaria vector Anopheles gambiae Giles sensu stricto (Diptera: Culicidae) two mutations have been identified. The first, consisting of a leucine-phenylalanine substitution at amino acid position 1014, is widespread in West Africa. The second, a leucine-serine substitution at the same position, has to date only been detected in western Kenya. Analysis of the kdr polymorphism in a sample of 106 An. gambiae s.s. of the rDNA S-form/Type I collected in Libreville (Gabon) surprisingly revealed the presence of both East and West African kdr mutations with frequencies of 63% and 37%, respectively. No wild-type alleles were detected and there was an excess of heterozygous genotypes (P = 0.04). In addition, an inconsistency was found during the kdr genotyping procedures by polymerase chain reaction, which could have lead to an underestimation of resistance alleles. The implications of these findings are discussed.

Biological invasions generally induce profound effects on the structure of resident communities. In Mayotte, where Aedes aegypti and Ae. lilii were already present, the recent introduction of Ae. albopictus raises public health concerns because it may affect the risk of arbovirus transmission. Entomological surveys were carried out in six locations on the island following a transect defined by a gradient from urban to rural habitats during a dry and a wet season. A total of 438 larval habitats containing Aedes spp. immature stages were surveyed. We evaluated the characteristics of larval habitats and analyzed the field distribution of Aedes spp. throughout Mayotte using generalized linear models. Artificial containers used for water storage were significantly more productive for Ae. albopictus immature stages than for Ae. aegypti ones. Most of natural larval habitats collected were colonized by Ae. aegypti, and it was also the most common Aedes species encountered in rural habitats. Conversely, Ae. albopictus greatly predominated in urban and suburban habitats and during the dry season. Ae. lilii was uncommon and occurred preferentially in leaf axillae and dead leaves on the ground. Ae. albopictus has rapidly colonized the inhabited areas of Mayotte. A displacement of Ae. aegypti populations by Ae. albopictus populations in urban areas might be happening. The increasing urbanization seems to greatly favor the presence of the invasive species. Thus, arbovirus surveillance programs should focus, as a priority, on areas where this vector is present because of a higher risk of emergence of an epidemic source of arboviruses.

Following veterinary alerts of Rift Valley fever (RVF) in the districts of Fianarantsoa I and II in November 2008 and in the district of Ambalavao in April 2009, entomological and virological investigations were carried out to identify the mosquito species that could act as RVF virus (RVFV) vectors in the region. A total of 12,785 adult mosquitoes belonging to 5 genera and 21 species were collected. After identification, mosquitoes were pooled by species, sex, and female status (fed or unfed) and then stored at -80°C. Of 319 pools of unfed monospecific female mosquito tested by real-time RT-polymerase chain reaction, RVFV was detected in 1 pool of Anopheles coustani, 5 pools of An. squamosus, and 2 pools of Culex antennatus mosquitoes. The virus was isolated in mosquito cell lines from two of the five Real Time-RT-polymerase chain reaction (real time-RT-PCR) positive pools of An. squamosus mosquitoes. From the eight RVFV strains detected, partial S, M, and L genome segments sequences were obtained. The phylogenetic analysis of these sequences showed that the strains circulating in mosquitoes were genetically close to those that circulated in livestock and humans during RVF outbreaks in 2008 and 2009. This study, therefore, provides strong evidence that An. squamosus, An. coustani, and Cx. antennatus could play a role as vectors of the RVFV during the disease outbreaks in 2008-2009. Bioecological, genetic, and RVF transmission studies on these three mosquito species are needed to address this question and thus improve prevention and control of future RVF outbreaks in Madagascar, where these species are present.