Investigation of the seasonal microbiome of Anopheles coluzzii in Mali
- Published
- Accepted
- Subject Areas
- Entomology, Microbiology
- Keywords
- Anopheles, Mosquito, Seasonality, Microbiome
- Licence
- This is an open access article, free of all copyright, made available under the Creative Commons Public Domain Dedication. This work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
- Cite this article
- 2017. Investigation of the seasonal microbiome of Anopheles coluzzii in Mali. PeerJ Preprints 5:e3391v1 https://doi.org/10.7287/peerj.preprints.3391v1
Abstract
The poorly understood mechanisms of the seasonal maintenance of Anopheles spp. mosquitoes through the dry season in Africa remain a critical gap in our knowledge of Plasmodium disease transmission. While it is thought that adult mosquitoes remain in a dormant state throughout this seven-month dry season, the nature of this state remains unknown and has largely not been recapitulated in laboratory settings. To elucidate possible life history traits allowing for this phenotype, the spatiotemporal change in the microbiome of mosquitoes in the dry and wet seasons in Mali was analyzed by sequencing the 16S ribosome bacterial region in whole bodies of adult mosquitoes collected from two locations with varying water availability. These locations were a village near the Niger River with year-round water sources (N’Gabakoro, “Riparian”), and an area closer to the Sahara with highly seasonal breeding sites (Thierola Area, “Sahelian”). The 16S bacterial data consisted of 2057 unique sequence variants in 426 genera across 184 families. With these, we found several compositional differences that were seasonally and spatially linked. Counter to our initial hypothesis, there was a more pronounced seasonal difference in the bacterial microbiome in the Riparian than Sahelian area. These major seasonal shifts were in Ralstonia, Sphingorhabdus, and Duganella spp. bacteria that are usually soil and water-associated, indicating that these changes may be from bacteria acquired in the larval environment, rather than during adulthood. In the Sahelian dry season mosquitoes, there was a unique intracellular bacteria, Anaplasma, which likely was acquired through non-human blood feeding. Coupled with this finding, cytochrome B analysis showed a greater heterogeneity in host choice of An. coluzzii independent of season in the Thierola area compared to N’Gabakoro (77.5% vs. 94.6% human-origin blood meal, respectively), which may indicate a relaxation of anthropophily in some locations. This study highlights the diversity present in the bacterial composition of individual mosquitoes, characterizes the spatial and seasonal differences in this composition, and indicates some possible qualitative biomarkers in areas of intense seasonal change.
Author Comment
This is a submission to PeerJ for review.
Supplemental Information
Seasonal fungal biome of mosquitoes determined via ITS region sequencing
Comparison of Shannon diversity between samples groups
Changes between season/location (A), season (B), and field vs. lab (C) are shown. Significance determined via Kruskal-Wallis chi-square with Dunn’s multiple comparison adjustment where applicable.
Comparison of mean number of sequence variants (A) and genera (B) for each season/location
Significance determined via Kruskal-Wallis chi-square with Dunn’s multiple comparison adjustment where applicable.
High presence of sample singletons at the Genus (A) and sequence variant (B) level
Genus-level DESeq2 analysis of samples between seasons and locations
A negative log2 fold change represents a genus-agglomerated taxon that loses read prevalence from the Wet to Dry Season, and a positive log2 fold change represents a genus that increases during this transition. p-adj is the Benjamini-Hochberg adjusted p-value accounting for a false discovery rate of 0.1. Each panel compares transition from Wet (left) to Dry (right, panels A and B) or from Field (left) to Laboratory (right, panel C).
Random forest supervised learning to discriminate season and location for each sample group
Analysis is performed on all samples (pooled and individuals) for each location. The top ten variables (genera) important to the created model is shown in Panel B. Significantly different center points in the ordination via PERMANOVA are present in Panel C. Prevalence of the genera for each sample group is presented in Panel D.
Sequences variants present in 19 most abundant genera
Phylogenetic tree representing each 16S sequence variant (A) and OTU-level (B, 97% identity) for the 19 most prevalent taxa found in the Anopheles coluzzii samples.
Supplementary information on the 19 most abundant genera in the study
Presence in arthropods is from Minard et al. unless otherwise noted. Non-insect environment and presence in arthropods is not comprehensive.
No correlation of known “kitome” genera with read count in our study
Kitome genera are based on summary table from Salter et al. (Salter et al., 2014)