Richard C. Wilkerson

Phylogeny of Anophelinae (Diptera: Culicidae) based on nuclear ribosomal and mitochondrial DNA sequences

Abstract Phylogenetic relationships among thirty‐two species of mosquitoes in subfamily Anophelinae are inferred from portions of the mitochondrial genes COI and COII, the nuclear 18S small subunit rRNA gene and the expansion D2 region of the nuclear large subunit 28S rRNA gene. Sequences were obtained from the genera Anopheles, Bironella and Chagasia. Representatives of all six subgenera of Anopheles were included: Anopheles, Cellia, Kerteszia, Lophopodomyia, Nyssorhynchus and Stethomyia. Using parsimony and maximum likelihood methods, various combinations of these DNA sequence data were analysed separately: 18S, 28S, combined 18S and 28S, combined COI and COII, and combined 18S, 28S, COI and COII (‘total evidence’). The combined rDNA data contain strong phylogenetic signal, moderately to strongly supporting most clades in MP and ML analyses; however, the mtDNA data (analysed as either nucleotide or amino acid sequences) contain little phylogenetic signal, except for relationships of very recently derived groups of species and, at the deepest level, for the monophyly of Anophelinae. The paraphyly of Anopheles relative to Bironella is confirmed by most analyses and statistical tests. Support for the monophyly of subgenera Anopheles, Cellia, Kerteszia and Nyssorhynchus is indicated by most analyses. Subgenus Lophopodomyia is reconstructed as the sister to Bironella, nested within a clade also containing Nyssorhynchus and Kerteszia. The most basal relationships within genus Anopheles are not well resolved by any of the data partitions, although the results of statistical analyses of the rDNA data (S‐H‐tests, likelihood ratio tests for monophyly and Bayesian MCMC analyses) suggest that the clade consisting of Bironella, Lophopodomyia, Nyssorhynchus and Kerteszia is the sister to the clade containing Cellia and Anopheles.

Toward understanding Anophelinae (Diptera, Culicidae) phylogeny: insights from nuclear single-copy genes and the weight of evidence.

A phylogeny of the mosquito subfamily Anophelinae was inferred from fragments of two protein-coding nuclear genes, G6pd (462 bp) and white (801 bp), and from a combined data set (2,136 bp) that included a portion of the mitochondrial gene ND5 and the D2 region of the ribosomal 28S gene. Sixteen species from all three anopheline genera and six Anopheles subgenera were sampled, along with six species of other mosquitoes used as an outgroup. Each of four genes analyzed individually recovered the same well-supported clades; topological incongruence was limited to unsupported or poorly supported nodes. As assessed by the incongruence length difference test, most of the conflicting signal was contributed by third codon positions. Strong structural constraints, as observed in white and G6pd, apparently had little impact on phylogenetic inference. Compared with the other genes, white provided a superior source of phylogenetic information. However, white appears to have experienced accelerated rates of evolution in few lineages, the affinities of which are therefore suspect. In combined analyses, most of the inferred relationship were well-supported and in agreement with previous studies: monophyly of Anophelinae, basal position of Chagasia, monophyly of Anopheles subgenera, and subgenera Nyssorhynchus + Kerteszia as sister taxa. The results suggested also monophyletic origin of subgenera Cellia + Anopheles, and the white gene analysis supported genus Bironella as a sister taxon to Anopheles. The present data and other available evidence suggest a South American origin of Anophelinae, probably in the Mesozoic; a rapid diversification of Bironella and basal subgeneric lineages of Anopheles, potentially associated with the breakup of Gondwanaland; and a relatively recent and rapid dispersion of subgenus Anopheles.

Phylogeny of Anophelinae (Diptera Culicidae) Based on Morphological Characters

Abstract Phylogenetic relationships of mosquitoes in the subfamily Anophelinae are presented based on a cladistic analysis of 163 morphological characters from females, males, fourth-instar larvae, and pupae of 64 species. Species examined include one Chagasia Cruz, three Bironella Theobald, and 60 species representing all six subgenera of Anopheles Meigen. Uranotaenia lowii Theobald and Aedeomyia squamipennis (Lynch Arribalzaga) are used as outgroups. This analysis indicates that Anophelinae is monophyletic, that Chagasia is the earliest-diverged lineage within Anophelinae, and that the genus Anopheles, as currently defined, is paraphyletic because it excludes Bironella. Four nonhomoplastic synapomorphies support the monophyly of the clade composed of the genera Bironella and Anopheles. Three major lineages are recognized within this clade. The basal lineage (lineage 1) contains An. (Nyssorhynchus), An. (Kerteszia), and An. implexus (Theobald). This analysis supports the monophyly of the sister groups Nyssorhynchus Blanchard and Kerteszia Theobald, but finds the Albimanus, Argyritarsis, and Myzorhynchella sections of the subgenus Nyssorhynchus to be paraphyletic. The second lineage (lineage 2) contains species of the subgenus Cellia Theobald, which is monophyletic. Except for the Cellia series, however, all other series of the subgenus Cellia are paraphyletic. The third lineage (lineage 3) contains species of the subgenera Anopheles Meigen, Stethomyia Theobald, Lophopodomyia Antunes, and of the genus Bironella. Bironella and Stethomyia are monophyletic sister groups. The Lophoscelomyia and Arribalzagia series are monophyletic. The Arribalzagia and Cycloleppteron series are sister groups nested within the Myzorhynchus series. We conclude that there is no support for the generic recognition of Bironella, nor the subgeneric rank of Lophopodomyia and Stethomyia.

Six new species of the Anopheles leucosphyrus group, reinterpretation of An. elegans and vector implications.

Abstract.  Among Oriental anopheline mosquitoes (Diptera: Culicidae), several major vectors of forest malaria belong to the group of Anopheles (Cellia) leucosphyrus Dönitz. We have morphologically examined representative material (> 8000 specimens from seven countries) for taxonomic revision of the Leucosphyrus Group. Six new species are here described from adult, pupal and larval stages (with illustrations of immature stages) and formally named as follows: An. latens n. sp. (= An. leucosphyrus species A of Baimai et al., 1988b), An. cracens n. sp., An. scanloni n. sp., An. baimaii n. sp. (formerly An. dirus species B, C, D, respectively), An. mirans n. sp. and An. recens n. sp. Additionally, An. elegans (James) is redescribed and placed in the complex of An. dirus Peyton & Harrison (comprising An. baimaii, An. cracens, An. dirus, An. elegans, An. nemophilous Peyton & Ramalingam, An. scanloni and An. takasagoensis Morishita) of the Leucosphyrus Subgroup, together with An. baisasi Colless and the An. leucosphyrus complex (comprising An. balabacensis Baisas, An. introlatus Baisas, An. latens and An. leucosphyrus). Hence, the former Elegans Subgroup is renamed the Hackeri Subgroup (comprising An. hackeri Edwards, An. pujutensis Colless, An. recens and An. sulawesi Waktoedi).

Making Mosquito Taxonomy Useful: A Stable Classification of Tribe Aedini that Balances Utility with Current Knowledge of Evolutionary Relationships.

The tribe Aedini (Family Culicidae) contains approximately one-quarter of the known species of mosquitoes, including vectors of deadly or debilitating disease agents. This tribe contains the genus Aedes, which is one of the three most familiar genera of mosquitoes. During the past decade, Aedini has been the focus of a series of extensive morphology-based phylogenetic studies published by Reinert, Harbach, and Kitching (RH&K). Those authors created 74 new, elevated or resurrected genera from what had been the single genus Aedes, almost tripling the number of genera in the entire family Culicidae. The proposed classification is based on subjective assessments of the “number and nature of the characters that support the branches” subtending particular monophyletic groups in the results of cladistic analyses of a large set of morphological characters of representative species. To gauge the stability of RH&K’s generic groupings we reanalyzed their data with unweighted parsimony jackknife and maximum-parsimony analyses, with and without ordering 14 of the characters as in RH&K. We found that their phylogeny was largely weakly supported and their taxonomic rankings failed priority and other useful taxon-naming criteria. Consequently, we propose simplified aedine generic designations that 1) restore a classification system that is useful for the operational community; 2) enhance the ability of taxonomists to accurately place new species into genera; 3) maintain the progress toward a natural classification based on monophyletic groups of species; and 4) correct the current classification system that is subject to instability as new species are described and existing species more thoroughly defined. We do not challenge the phylogenetic hypotheses generated by the above-mentioned series of morphological studies. However, we reduce the ranks of the genera and subgenera of RH&K to subgenera or informal species groups, respectively, to preserve stability as new data become available.

Aedes (Finlaya) japonicus (Diptera: Culicidae), a Newly Recognized Mosquito in the United States: Analyses of Genetic Variation in the United States and Putative Source Populations

Abstract Introduction of potential disease vectors into a new geographic area poses health risks to local human, livestock, and wildlife populations. It is therefore important to gain understanding of the dynamics of these invasions, in particular its sources, modes of spread after the introduction, and vectorial potential. We studied the population genetics of Aedes (Finlaya) japonicus japonicus (Theobald), an Asian mosquito that was recognized for the first time in the United States in 1998. We examined patterns of genetic diversity using random amplified polymorphic DNA and sequences of ND4 of mtDNA by comparing samples from populations spanning the range of this mosquito in Japan (six samples) and the United States (nine samples) as well as specimens intercepted in New Zealand in 1999. We found geographically differentiated populations in Japan, indicating limited gene flow even on small spatial scales. In the United States, we found evidence of significant genetic differentiation between samples from New York, Connecticut, and New Jersey and those from mid-Pennsylvania and Maryland. We were unable to pinpoint the source location(s) in Japan, although some of the U.S. samples are genetically close to samples from south Honshu and western Kyushu. Further studies should include samples from Korean populations. Distinct genetic signatures in U.S. populations undergoing expansion suggest the possibility of local increases in genetic diversity if and where they meet.

Random amplified polymorphic DNA (RAPD) markers readily distinguish cryptic mosquito species (Diptera: Culicidae: Anopheles)

The usefulness of random amplified polymorphic DNA (RAPD) was examined as a potential tool to differentiate cryptic mosquito species. It proved to be a quick, effective means of finding genetic markers to separate two laboratory populations of morphologically indistinguishable African malaria vectors, Anopheles gambiae and An. arabiensis. In an initial screening of fiftyseven RAPD primers, 377 bands were produced, 295 of which differed between the two species. Based on criteria of interpretability, simplicity and reproducibility, thirteen primers were chosen for further screening using DNA from thirty individuals of each species. Seven primers produced diagnostic bands, five of which are described here. Some problematic characteristics of RAPD banding patterns are discussed and approaches to overcome these are suggested.

Ribosomal DNA ITS2 sequences differentiate six species in the Anopheles crucians complex (Diptera: Culicidae).

Abstract Anopheles crucians Wiedemann (sensu lato) was investigated for the presence of cryptic species using rDNA ITS2 sequences. This complex of species presently contains the named species An. crucians, An. bradleyi King, and An. georgianus King. Adult female mosquitoes were collected at 28 sites in Alabama, Florida, Georgia, North Carolina, Mississippi, and Louisiana, resulting in 245 progeny broods. Species were identified using preliminary morphological characters, and the internal transcribed spacer two (ITS2) was amplified from all broods. The result was five distinct sizes of amplification product, and based on morphological characters, one of the size classes was suspected to consist of two species. All six putative species were then sequenced: five directly, and the sixth, because of extreme intragenomic (each individual with many variants) size variability, cloned. The ITS2 sequences were markedly distinct for all six species. Species designations and ITS2 sequence lengths (base pairs in parentheses) were A (461), B (1,000+), C (204), D (293), E (195), and An. bradleyi (208). Species B showed both large intraspecific and intragenomic sequence variability and is distinguished by having the longest ITS2 found so far in an Anopheles. Based on these data, we found that all species could be identified with polymerase chain reaction (PCR) using a mixture of four primers in a single reaction. Members of this complex were often found in sympatry, with the adults of five species collected at a single site in central Florida.

Revision of the Leucosphyrus group of Anopheles (Cellia) (Diptera, Culicidae)

Revision of the Leucosphyrus Group of Anopheles (Cellia) (Diptera, Culicidae). This is a comprehensive revision of 20 species of the Leucosphyrus Group of the Neomyzomyia Series of Anopheles (Cellia). Morphological description of the adults, male and female, male genitalia, pupa and fourth-instar larva are provided for each taxon in addition to bionomics, distribution data and systematic discussion for each species, including diagnostic characters. Identification keys for females and fourth-instar larvae are provided. When possible medical importance of each species is included. Illustrations of of the adults, fourth-instar larvae and pupae are provided. Distribution maps for each species are mainly based on the material examined; however, when possible published data were also used. Tables on adult character variations, fourth-instar larval and pupal setal branching are included as appendices. A neotype for An. takasagoensis Morishita and An. sulawesi Koesoemawinangoen, and a lectotype for An. balabacensis Baisas, are designated. The authorship of An. sulawesi previously cited as Waktoedi is corrected to Koesoemawinangoen.

Insight into Global Mosquito Biogeography from Country Species Records

Abstract To advance our limited knowledge of global mosquito biogeography, we analyzed country occurrence records from the Systematic Catalog of the Culicidae (http://www.mosquitocatalog. org/main.asp), and we present world maps of species richness and endemism. A latitudinal biodiversity gradient was observed, with species richness increasing toward the equator. A linear log-log species (y)–area (x) relationship (SAR) was found that we used to compare observed and expected species densities for each country. Brazil, Indonesia, Malaysia, and Thailand had the highest numbers of species, and Brazil also had the highest taxonomic output and number of type locations. Brazil, Australia, the Philippines, and Indonesia had the highest numbers of endemic species, but excluding small island countries, Panama, French Guiana, Malaysia, and Costa Rica had the highest densities of total species and endemic species. Globally, 50% of mosquito species are endemic. Island countries had higher total number of species and higher number of endemic species than mainland countries of similar size, but the slope of the SAR was similar for island and mainland countries. Islands also had higher numbers of publications and type locations, possibly due to greater sampling effort and/or species endemism on islands. The taxonomic output was lowest for some countries in Africa and the Middle East. A consideration of country estimates of past sampling effort and species richness and endemism is proposed to guide mosquito biodiversity surveys. For species groups, we show that the number of species of Anopheles subgenus Anopheles varies with those of subgenus Cellia in a consistent manner between countries depending on the region. This pattern is discussed in relation to hypotheses about the historical biogeography and ecology of this medically important genus. Spatial analysis of country species records offers new insight into global patterns of mosquito biodiversity and survey history.