Geometric morphometrics of nine field isolates of Aedes aegypti with different resistance levels to lambda-cyhalothrin and relative fitness of one artificially selected for resistance

Abstract

Aedes aegypti, a mosquito closely associated with humans, is the principal vector of dengue virus which currently infects about 400 million people worldwide. Because there is no way to prevent infection, public health policies focus on vector control; but insecticide-resistance threatens them. However, most insecticide-resistant mosquito populations exhibit fitness costs in absence of insecticides, although these costs vary. Research on components of fitness that vary with insecticide-resistance can help to develop policies for effective integrated management and control. We investigated the relationships in wing size, wing shape, and natural resistance levels to lambda-cyhalothrin of nine field isolates. Also we chose one of these isolates to select in lab for resistance to the insecticide. The main life-traits parameters were assessed to investigate the possible fitness cost and its association with wing size and shape. We found that wing shape, more than wing size, was strongly correlated with resistance levels to lambda-cyhalothrin in field isolates, but founder effects of culture in the laboratory seem to change wing shape (and also wing size) more easily than artificial selection for resistance to that insecticide. Moreover, significant fitness costs were observed in response to insecticide-resistance as proved by the diminished fecundity and survival of females in the selected line and the reversion to susceptibility in 20 generations of the non-selected line. As a practical consequence, we think, mosquito control programs could benefit from this knowledge in implementing efficient strategies to prevent the evolution of resistance. In particular, the knowledge of reversion to susceptibility is important because it can help in planning better strategies of insecticide use to keep useful the few insecticide-molecules currently available.

Figure 1. Arrangement of landmarks digitized on right and left wings of 758 Aedes aegypti females.

The landmarks were digitized in a clockwise sequence from the number one.

Figure 2. Dendrogram showing the relationships in wing size of isolates from Cúcuta and Quibdó.

Between parentheses is the RR₅₀ to lambda-cyhalothrin. Cophenetic correlation coefficient: 0.892522. isolates from Cúcuta are colored in red, and those from Quibdó in green ROCK: the Rockefeller strain.

Figure 3. Wing size variation among Aedes aegypti isolates from field.

Violin plots enclose box plots. Each box is divided by the median (black circles), which top and bottom correspond to 25th and 75th quartiles, respectively. Isolates from Cúcuta are colored in red, and those from Quibdó in green. Between parentheses is the RR₅₀ to lambda-cyhalothrin. CS: centroid-size.

Figure 4. Dendrogram showing the relationships in wing shape of isolates from Cúcuta and Quibdó.

Cophenetic correlation coefficient: 0.8935737. Isolates from Cúcuta are colored in red, and those from Quibdó in green. Between parentheses is the RR₅₀ to lambda-cyhalothrin. ROCK: the Rockefeller strain.

Figure 5. Wing size variation of both lambda-cyhalothrin selected and non-selected lines.

Violin plots enclose box plots. Each box is divided by the median (black circles), which top and bottom correspond to 25th and 75th quartiles, respectively. CS: centroid-size; P: parental Comuneros isolate; F9-NS and F20-NS: ninth and twenty generations non-selected for lambda-cyhalothrin resistance; F9-S and F20-S: ninth and twenty generations selected for lambda-cyhalothrin resistance. Sample-sizes were 51 parental females, 42 F9-NS, 49 F9-S, 50 F20-NS, and 56 F20-S.

Figure 6. Scatterplot of the scores along the first two principal components (relative warps), the convex hull for each group and the average scores for each group.

Convex hulls enclose individuals from parental and its descendants at ninth and twenty generations which were ordered on the two first relative warps. Squares represent the centroids (average scores for each group). To easy viewing of the group's centroids, the individual positions are not shown. Deformation grids for the extreme values of the first and the second relative warps on the mean configuration, are also shown. P: parental Comuneros isolate; F9-NS and F20-NS: ninth and twenty generations non-selected for lambda-cyhalothrin resistance; F9-S and F20-S: ninth and twenty generations selected for lambda-cyhalothrin resistance. Sample-sizes were 51 parental females, 42 F9-NS, 49 F9-S, 50 F20-NS, and 56 F20-S.

Figure 7. Survival analysis using Kaplan-Meier estimates.

Curves represent daily survival of the Rockefeller strain (ROCK) females, and the ninth (F9-S) and tenth (F10-S) generations of the lambda-cyhalothrin selected line from COM isolate, and the ninth (F9-NS) and tenth (F10-NS) generations of the non-selected line from the same isolate. Sample-sizes were 100 females for both F10-S and F10-NS, 124 for F9-NS, 150 for F9-S, and 225 for the ROCK strain.

Figure 8. Box-plots showing the fecundity variation of both lambda-cyhalothrin selected and non-selected lines.

Each box is divided by the median, which top and bottom correspond to 25th and 75th quartiles, respectively. ROCK: the Rockefeller strain, F9-S and F10-S: the selected line at ninth and tenth generations, and F9-NS and F10-NS: the non-selected line at ninth and tenth generations. Sample-sizes were 100 females for both F10-S and F10-NS, 124 for F9-NS, 150 for F9-S, and 225 for the ROCK strain.