The intolerable burden of malaria has for too long plagued humanity and the prospect of eradicating malaria is an optimistic, but reachable, target in the 21 st century. However, extensive knowledge is needed about the spatial structure of mosquito populations in order to develop effective interventions against malaria transmission. We hypothesized that the microbiota associated with a mosquito reflects acquisition of bacteria in different environments. By analyzing the whole-body bacterial flora of An. gambiae mosquitoes from Burkina Faso by 16 S amplicon sequencing, we found that the different environments gave each mosquito a specific bacterial profile. In addition, the bacterial profiles provided precise and predicting information on the spatial dynamics of the mosquito population as a whole and showed that the mosquitoes formed clear local populations within a meta-population network. We believe that using microbiotas as proxies for population structures will greatly aid improving the performance of vector interventions around the world.As a part of a holistic approach to controlling the disease, the eradication of malaria requires a strong intervention against the vectors transmitting malaria. Today, impregnated bednets are efficient barriers of night-time malaria transmission. However, the anthropophilic malaria mosquitoes have shifted their feeding patterns to circumvent the bednet barriers, whilst an increasing pesticide resistance in the mosquitoes also reduces the effectivity of bednets and indoor residual spraying 1 . Under these prevailing conditions it is anticipated that new strategies must be explored which eliminate the parasites in the mosquitoes themselves. Control strategies involving genetic modification of mosquitoes (transgenesis) and genetic modification of their gut bacteria (paratransgenesis) both build on the premises that once released, the modified organism is maintained and spreads by itself through the vector population 2,3 . In paratransgenesis, the delivery of modified bacteria can be either through larval breeding sites or artificial sugar sources, but fundamental knowledge of where malaria mosquitoes acquire their bacteria is lacking. For transgenesis, the prerequisite for a successful intervention is good knowledge of malaria mosquito life history including dispersal distances, formation of local-and metapopulations and rates of exchange between populations.Based on the hypothesis that the microbiota associated with a mosquito can be seen as a shadow cast by life-history events, where different environments leave their mark by contributing to the flora associated with the insect, we collected An. gambiae adult mosquitoes from three villages in Burkina Faso (Fig. 1). We analyzed the whole-body bacterial flora of the mosquitoes using 16 S amplicon sequencing yielding an average of 17,300 (SD 13,161.83) sequences per mosquito (see Methods and Supplementary Fig. S1). After selecting the mosquitoes with at least 7500 sequences based on rarefaction-curves analysis ( Supplementary Fi...

The midgut microbiota of disease vectors plays a critical role in the successful transmission of human pathogens. The environment influences the microbiota composition; however, the relative mosquitospecies contribution has not been rigorously disentangled from the environmental contribution to the microbiota structure. Also, the extent to which the microbiota of the adult sugar food source and larval water can predict that of the adult midgut and vice versa is not fully understood. To address these relationships, larvae and adults of Anopheles gambiae and Aedes albopictus were either reared separately or in a co-rearing system, whereby aquatic and adult stages of both species shared the larval water and sugar food source, respectively. Despite being reared under identical conditions, clear intraand interspecies differences in midgut microbiota-composition were observed across seven cohorts, collected at different time points over a period of eight months. Fitting a linear model separately for each OTU in the mosquito midgut showed that two OTUs significantly differed between the midguts of the two mosquito species. We also show an effect for the sugar food source and larval water on the adult midgut microbiota. Our findings suggest that the mosquito midgut microbiota is highly dynamic and controlled by multiple factors.

Water-storage containers are common in households where access to water is scarce and often act as breeding sites for vector mosquitoes. Bacteria in these containers may be important for attracting or repelling ovipositing mosquitoes. We hypothesized that bacterial community composition in water-storage containers would represent either inhibitory or suitable environmental conditions for mosquito larvae. To investigate this, we characterized the bacterial community composition in water-storage containers and correlated these communities to Aedes and Anopheles larval densities. Water samples were collected over two years from 13 containers in an Indian village and analyzed by high throughput 16S rRNA gene amplicon sequencing. Comparisons of bacterial community composition between water with and without mosquito larvae showed that Xanthomonadaceae, Comamonadaceae and Burkholderiaceae were more common (P < 0.05) in absence of larvae, while Lachnospiraceae, Synechococcaceae, Alcaligenaceae and Cryomorphaceae were more common (P < 0.05) in presence of larvae. Indicator analysis identified operational taxonomic units designated as CL500-29 marine group (Acidimicrobiaceae) and FukuN101 (Microbacteriaceae) for absence and presence of larvae, respectively. These results contribute to the understanding of which bacteria, directly or indirectly, can be linked to absence or presence of mosquitoes around households and set the basis for potential measures to be taken against these vector mosquitoes.