Exploring the Diversity of Microbial Communities Associated with Two Anopheles Species During Dry Season in an Indigenous Community from the Colombian Amazon

Abstract

Malaria disease affects millions of people annually, making the Amazon Basin a major hotspot in the Americas. While traditional control strategies rely on physical and chemical methods, the Anopheles microbiome offers a promising avenue for biological control, as certain bacteria can inhibit parasite development and alter vector immune and reproductive systems, disrupting the transmission cycle. For this reason, this study aimed to explore the bacterial communities in An. darlingi and An. triannulatus s.l., including breeding sites, immature stages, and adults from San Pedro de los Lagos (Leticia, Amazonas) through next-generation sequencing of the 16S rRNA gene. The results revealed a higher bacterial genus richness in the L1-L2 larvae of An. triannulatus s.l. Aeromonas and Enterobacter were prevalent in most samples, with abundances of 52.51% in L3-L4 larvae and 48.88% in pupae of An. triannulatus s.l., respectively. In breeding site water, Verrucomicrobiota bacteria were the most dominant (52.39%). We also identified Delftia (15.46%) in An. triannulatus s.l. pupae and Asaia (98.22%) in An. triannulatus, linked to Plasmodium inhibition, and Elizabethkingia, in low abundances, along with Klebsiella and Serratia, known for paratransgenesis potential. Considering the high bacterial diversity observed across the different mosquito life stages, identifying bacterial composition is the first step towards developing new strategies for malaria control. However, the specific roles of these bacteria in anophelines and the malaria transmission cycle remain to be elucidated.

Keywords: Amazon region; Anopheles darlingi; Anopheles triannulatus s.l.; endosymbionts; microbiota.

Figure 1

Sample collection area. (A) Satellite view of the Anopheles collection site in the Amazon basin, Colombia, near the border with the Peruvian and Brazilian Amazon rainforest. (B) Indigenous community of S.P.L. and its surrounding area within the Yahuarcaca watershed.

Figure 2

Active search for potential Anopheles breeding sites in the Yahuarcaca watershed at S.P.L. locality and installation of light traps. (A) Palm tree bracts that store rainwater. (B) Seed husk, described by the community as a potential breeding site. (C) Abandoned canoe in a fishing pond, flooded by rainwater, without fish inside the pond. (D) Collection using CDC light traps before dawn, ensuring 10 h of activity during the night. (E) Fishing pond. (F) Puddles adjacent to the Yahuarcaca stream.

Figure 3

Neighbor-joining phylogenetic tree of the COX1 sequences (710 bp) from anopheline species, reconstructed with the K2P model, and a bootstrap value of 1000. Colors mark the different life stages of the specimens collected in S.P.L., Leticia, Amazonas. Yellow: Larvae. Blue: Pupae. Green: Adults. Ae. albopictus was used as an outgroup. Bootstrap values for each cluster are as follows: value of 99 for An. triannulatus s.l.; value of 100 for An. darlingi; value of 99 for An. dunhami; value of 95 for An. squamifemur.

Figure 4

Stacked bar chart of the relative abundance profiling of phyla (A) and genera (B) found in An. triannulatus s.l., its breeding site, and An. darlingi. Samples of mosquitoes are grouped by life stage. All samples were collected in the fishing pond, except for the An. darlingi pool of adults that were collected from the peridomestic area of S.P.L.-Leticia, Amazonas. The Core Community sample was used as a control. Core Community: Control sample; BSW: Breeding site water; At. Larvae: An. triannulatus s.l. larvae samples; At. Pupae: An. triannulatus s.l. pupae samples; At. Adult: An. triannulatus s.l. adult sample; Ad. Adult: An. darlingi adult sample.

Figure 5

Box-and-whiskers plot representing α-diversity and differences in community structures of immature and adult stages of An. triannulatus s.l., adults of An. darlingi from the peridomestic area of S.P.L., and An. triannulatus s.l. breeding site in terms of (A) observed richness (p-value 0.017, T-test statistic: 3.5182); (B) Shannon index (p-value: 0.28, T-test statistic: 1.9405); (C) Simpson index (p-value 0.42, T-test statistic: 1.2518); (D) Chao1 index (p-value 0.017, T-test statistic: 3.5182); (E) PCoA (F-value: 2.3931; R-squared: 0.25477; p-value: 0.03); and (F) NMDS (F-value: 2.3931; R-squared: 0.25477; p-value: 0.028; stress: 0.018854).

Figure 6

Correlation network using Pearson’s correlation coefficient (r) and log-transformed abundances of microbial groups in the different anopheline samples and their breeding site. The green circle represents the cluster associated with bacteria from the water sample. The yellow circle represents the cluster of bacteria from An. darlingi adults. Clusters with no circles represent bacteria of immature and adult stages of An. triannulatus s.l.

Figure 7

Log-transformed abundances of Delftia (A), Thorsellia (B), Serratia (C), Elizabethkingia (D), and Bacillus (E). Water: sample from the breeding site. I1: larvae of An. triannulatus s.l. in the first and second instar. I3: larvae of An. triannulatus s.l. in the third and fourth instar. P: pupae of An. triannulatus s.l. At: adults of An. triannulatus s.l. Ad: adults of An. darlingi.