Geometric morphometric wing analysis represents a robust tool to identify female mosquitoes (Diptera: Culicidae) in Germany
Data files
Oct 19, 2020 version files 3.69 GB
-
fig_S1_pictures_examples_wing.docx
-
tab_S1_specimens_table_wing.xlsx
-
Wing_pictures_Sauer_et_al..rar
Abstract
Accurate species identification is the prerequisite to assess the relevance of mosquito specimens, but is often hindered by missing or damaged morphological features. The present study analyses the applicability of wing geometric morphometrics as a low-cost and practical alternative to identify native mosquitoes in Germany. Wing pictures were collected for 502 female mosquitoes of five genera and 19 species from 80 sampling sites. The reliable species identification based on interspecific wing geometry of 18 landmarks per specimen was tested. Leave-one-out cross validation revealed an overall accuracy of 99% for the genus and 90% for the species identification. Misidentifications were mainly due to three pairings of Aedes species: Aedes annulipes vs. Aedes cantans, Aedes cinereus vs. Aedes rossicus and Aedes communis vs. Aedes punctor. Cytochrome oxidase subunit I (COI) gene region was sequenced to validate the morphological and morphometric identification. Similar to the results of the morphometric analysis, the same problematic three Aedes-pairs clustered, but most other species could be well separated. Overall, our study underpins that morphometric wing analysis is a robust tool for reliable mosquito identification, which reach the accuracy of COI barcoding.
Methods
Mosquito sampling
The right wings of 502 females of 19 mosquito species were analysed (Table 1). Mosquito specimens were collected at 80 study sites in Germany 2017 and 2018 (Figure 1). Adults were sampled with two different methods: CO2-baited Biogents-traps (Biogents, Regensburg, Germany) or hand-made aspirators modified from Vazquez-Prokopec, et al.27. Immature stages were sampled with a standard dipper (Bioquip, CA, USA) in breeding sites and subsequently reared to adults in the laboratory. All specimens were identified by morphology17,28 and stored at -18°C in a freezer until further analysis. Specimens of each species were selected from at least three different sampling sites (Table 1). The coordinates of the sampling location, sampling date and sampling method for each specimen are given in the supplementary material (Table S1).
Measuring wing shape
The right wing of each specimen was removed and mounted under a cover slip (15 × 15 mm) with Euparal (Carl Roth, Karlsruhe, Germany). Pictures of each wing were taken under 20× magnification with a stereomicroscope (Leica M205 C, Leica Microsystems, Wetzlar, Germany). Fiji29 as bioscience package of ImageJ30 was used to digitize 18 landmarks (Figure S1). The landmark selection was in accordance with other studies analysing mosquito wing morphometry21-23,31. The wing pictures were divided among two observers (authors LE and FGS) and digitalised in random order to minimize a memory biased landmark collection between the mosquito specimens of the same species. One month later, the measurement was repeated for three specimens per species by four observers (authors LE, LJ, RL and FGS) to assess the degree of observer error and repeatability in landmark collection32.
Genetic identification
DNA isolation was performed from the whole mosquito body, except of the right wing. Individual specimens were placed into 2 ml tubes and about 10 pieces of 2.0 mm zirconia beads (BioSpec Products, Bartlesville, USA) as well as 1 ml of cell culture medium (high-glucose Dulbecco’s modified Eagle’s medium; Sigma-Aldrich, St. Louis, MO, USA) were added. The homogenization was performed with a Tissuelyser II (Qiagen, Hilden, Germany) for 2 min at 30 oscillations/s and 200 μl of the homogenate were used for DNA extraction, which was performed with KingFisher™ Flex Magnetic Particle Processor using MagMAX™ CORE Nucleic Acid Purification Kit (both Thermo Fisher Scientific, Waltham, MA, USA). Polymerase chain reaction (PCR) amplification of cytochrome oxidase subunit I (COI) gene region was conducted with the protocol published by Fang, et al.33 using the primers by Folmer, et al.34. Sanger sequencing was applied for all positive amplicons (LGC Genomics, Berlin, Germany). Furthermore, morphologically identified Cx. pipiens s.l. and An. maculipennis s.l. specimens were typed to species level (Cx. pipiens pipiens form pipiens resp. Anopheles messeae) using two molecular assays35,36.
Usage notes
Herein, we provide all mosquito wing pictures used in this study, including metadata on the sampling location, date of sampling and sampling method for each specimen.