Morphometrical diagnosis of the malaria vectors Anopheles cruzii, An. homunculus and An. bellator

Abstract

Background: Anopheles (Kerteszia) cruzii is a primary vector of Plasmodium parasites in Brazil's Atlantic Forest. Adult females of An. cruzii and An. homunculus, which is a secondary malaria vector, are morphologically similar and difficult to distinguish when using external morphological characteristics only. These two species may occur syntopically with An. bellator, which is also a potential vector of Plasmodium species and is morphologically similar to An. cruzii and An. homunculus. Identification of these species based on female specimens is often jeopardised by polymorphisms, overlapping morphological characteristics and damage caused to specimens during collection. Wing geometric morphometrics has been used to distinguish several insect species; however, this economical and powerful tool has not been applied to Kerteszia species. Our objective was to assess wing geometry to distinguish An. cruzii, An. homunculus and An. bellator.

Methods: Specimens were collected in an area in the Serra do Mar hotspot biodiversity corridor of the Atlantic Forest biome (Cananeia municipality, State of Sao Paulo, Brazil). The right wings of females of An. cruzii (n= 40), An. homunculus (n= 50) and An. bellator (n= 27) were photographed. For each individual, 18 wing landmarks were subjected to standard geometric morphometrics. Discriminant analysis of Procrustean coordinates was performed to quantify wing shape variation.

Results: Individuals clustered into three distinct groups according to species with a slight overlap between representatives of An. cruzii and An. homunculus. The Mahalanobis distance between An. cruzii and An. homunculus was consistently lower (3.50) than that between An. cruzii and An. bellator (4.58) or An. homunculus and An. bellator (4.32). Pairwise cross-validated reclassification showed that geometric morphometrics is an effective analytical method to distinguish between An. bellator, An. cruzii and An. homunculus with a reliability rate varying between 78-88%. Shape analysis revealed that the wings of An. homunculus are narrower than those of An. cruzii and that An. bellator is different from both of the congeneric species.

Conclusion: It is possible to distinguish among the vectors An. cruzii, An. homunculus and An. bellator based on female wing characteristics.

Figure 1

Morphological differentiation of 5th posterior tarsomere visualized in adults of An. cruzii (a) and An. bellator (b).

Figure 2

(a) Wing of female Anopheles bellator colored with acid fuchsin showing the 18 landmarks selected for geometric morphometrics. (b) Diagram of immaginary links between landmarks used to depict wing consensus.

Figure 3

Descriptive statistics of centroid sizes (in mm).

Figure 4

Morphological space of canonical variates resulting from comparison among the three species.

Figure 5

Pairwise comparison of thin plate spline between species. Landmarks displacement correspond to extreme of differentiation in each species. (a) An. cruzii (gray) and An. bellator (black); (b) An. bellator (black) and An. homunculus (gray) and (c) An. homunculus (gray) and An. cruzii (black).

Figure 6

Graphical presentation of the distance between landmarks used for ratios x/y and z/w.

Figure 7

Descriptive statistics of ratio x/y (in mm).

Figure 8

Descriptive statistics of ratio z/w (in mm).