Microevolution of Aedes aegypti

Abstract

Scientific research into the epidemiology of dengue frequently focuses on the microevolution and dispersion of the mosquito Aedes aegypti. One of the world's largest urban agglomerations infested by Ae. aegypti is the Brazilian megalopolis of Sao Paulo, where >26,900 cases of dengue were reported until June 2015. Unfortunately, the dynamics of the genetic variability of Ae. aegypti in the Sao Paulo area have not been well studied. To reduce this knowledge gap, we assessed the morphogenetic variability of a population of Ae. aegypti from a densely urbanised neighbourhood of Sao Paulo. We tested if allelic patterns could vary over a short term and if wing shape could be a predictor of the genetic variation. Over a period of 14 months, we examined the variation of genetic (microsatellites loci) and morphological (wing geometry) markers in Ae. aegypti. Polymorphisms were detected, as revealed by the variability of 20 microsatellite loci (115 alleles combined; overall Fst = 0.0358) and 18 wing landmarks (quantitative estimator Qst = 0.4732). These levels of polymorphism are higher than typically expected to an exotic species. Allelic frequencies of the loci changed over time and temporal variation in the wing shape was even more pronounced, permitting high reclassification levels of chronological samples. In spite of the fact that both markers underwent temporal variation, no correlation was detected between their dynamics. We concluded that microevolution was detected despite the short observational period, but the intensities of change of the markers were discrepant. Wing shape failed from predicting allelic temporal variation. Possibly, natural selection (Qst>Fst) or variance of expressivity of wing phenotype are involved in this discrepancy. Other possibly influential factors on microevolution of Ae. aegypti are worth searching. Additionally, the implications of the rapid evolution and high polymorphism of this mosquito vector on the efficacy of control methods have yet to be investigated.

Fig 1. Sampling locations of Aedes aegypti.

Left: outline map of Sao Paulo City showing in detail the political boundaries of neighbourhood “Subprefeitura Butanta”. Right: Magnified outline of “Subprefeitura Butanta” depicting the exact location of each egg trap.

Fig 2. (A) Wing of Aedes aegypti (female) depicting the 18 landmarks chosen. (B) Geometric diagram linking all landmarks.

Fig 3. UPGMA dendrogram using Nei’s genetic distance between chronological samples.

Fig 4. Factorial correspondence analysis of allelic temporal variation.

Each polygon represents the multilocus genetic variation of each chronological sample. Between brackets, the relative contribution of each factor (accounted for 61.19% of the total variation).

Fig 5. Graphical presentation of the Bayesian model-based clustering analysis.

Each individual is represented by a vertical bar. (A) k = 7, the best value of K for all mosquitoes of the chronological samples. (B) k = 5, arbitrarily chosen to match the number of chronological samples. (C) k = 3, the best value of K for all mosquitoes from trap T6.

Fig 6. Wing shape temporal variations represented by thin-plate splines.

Only the 18 landmarks, the wing outline and deformation vectors are shown. In each chronological sample depicted variations are relative to the previous sample. Deformations were magnified 10X to facilitate visualization.

Fig 7. Morphological space of canonical variables (CVs) 1 and 2 yielded by discriminant analysis of wing principal components.

Each polygon represents a chronological sample. Relative discriminant power of each variable is between brackets (accounted for 77.5% of the total variation).

Fig 8. UPGMA phenograms using the Mahalanobis distance between chronological samples.