(Faran 1979) and Puerto Cabello, Cojedes state, Venezuela (Gabalddn et al. ..... to geographical barriers such as the Orinoco or Amazon River deltas (Conn.
Mitochondrial DNA Variation Within and Between Two Species of Neotropical Anopheline Mosquitoes (Diptera: Culicidae) J. E. Conn, S. E. Mitchell, and A. F. Cockburn
From the Florida Medical Entomology Laboratory, University of Florida, Vero Beach, Florida (Conn) and the USDA/ARS, Center for Medical, Agricultural, and Veterinary Entomology, Gainesville, Florida. Address reprint requests to Dr. Conn, Department of Biology, University of Vermont, Marsh Life Sciences Building, Burlington, VT 05405-0086. S. E. Mitchell Is currently at the USDA/ARS Plant Genetic Resources Conservation Unit, University of Georgia Agricultural Experimental Station, Griffin, Georgia. We thank A. Anselmf (Direccidn de Malarlologfa y Saneamlento Ambiental, Venezuela), H. Perez (Institute Venezolano de Investlgaclones Clentlficas), R. Vargas (Minlsterio de Prevision Social y Salud Publlca, Bolivia), H. Bermudez and R. Rodriguez (Unlversldad Mayor de San Simon, Cochabamba, Bolivia), J. Cespedes, J. Cuba, A. Melgar, C. T6ffen, S.VIIlarroel, and F. Yabeta Pedraza (Unidad Sanitaria, Bolivia), and J. Alarcdn, E. Castro, G. Shlguango, and J. Yepez (Servlclo Nacional de la Erradlcacl6n de Malaria, Ecuador) for technical and logistic support with field collections. Morphological Identifications were confirmed by E. L. Peyton and R. Wllkerson (Walter Reed Army Institute, Smithsonian Institution) and L. Hribar and L P. Lounlbos (Florida Medical Entomology Laboratory). R. Wllkerson generously provided the BrazilIan samples. Special thanks to B. Bowen, A. Z. Brower, J. A. Danoff-Burg, D. D. Judd, V. L. Kulasekera. and B.W. Wiegmann for Insightful comments and discussions of the data set. Also thanks to R. DeSalle and an anonymous reviewer for constructive comments on the manuscript. J. A. Seawright (CMAVE/USDA/ARS, Gainesville, Florida) was very supportive of this research In all stages. The technical assistance of M. Montenegro and D. Rowold Is greatly appreciated. This research was supported by a National Institutes of Health grant (AI-31034) and by a Natural Sciences and Engineering Research Council (Canada) postdoctoral research fellowship to J.E.C. This article Is University of Florida IFAS journal series no. R-04771. Journal of Heredity 1997^838-107; 0022-1503/97/J5.00
98
Neotropical anopheline mosquitoes in the subgenus Nyssortiynchus are responsible for approximately 15 million cases of malaria annually in Latin America (Marques 1986; OPS 1988). Adult females, which transmit malaria parasites, exhibit considerable color variation in wing spots and leg bands which have historically been used as key taxonomic characters (Faran 1980; Harbach 1994; Wilkerson and Peyton 1990). The taxonomic status of species in this subgenus has been reevaluated with polytene chromosomes (Conn et al. 1993b; Moncada P6rez and Conn 1992), mitochondrial and nuclear sequences (Danoff-Burg J and Conn J, unpublished data; Fritz et al. 1994; Perera 1993), scanning electron micrographs of eggs (Linley and Lounibos 1993; Linley et al. 1993), RFLPs (Conn et al. 1993a; Freitas-Sibajev et al. 1995), and RAPDs (Wilkerson et al. 1995). Nevertheless, the status of many species remains unresolved. Species complexes (comprised of cryptic species) appear to be fairly common (Rosa-Freitas et al. 1990; Wilkerson et al. 1995), but species boundaries have not been rigorously tested to determine whether putative taxa are independent evolutionary units. Members of species complexes vary in vector capacity, as well as in the nature of gene flow and genetic variation, both of which are critical to explaining disease epidemiology (Coluzzi 1970; Tabachnick and Black 1995).
As a genetic marker, mtDNA can reveal historical and phylogeographic patterns as well as rates of gene Dow, and it can also be used to define maternal gene genealogies within species (Avise 1994; Avise et al. 1987; Wilson et al. 1985). The significant advantages of mtDNA for these studies are its maternal mode of inheritance, lack of recombination, and rapid evolution (Avise and Lansman 1983; Brown 1983). In part because of its rapid evolution, mtDNA is particularly appropriate for studies of recently evolved groups (Brown 1985; but see Powell et al. 1986) such as anopheline mosquitoes (Coluzzi et al. 1985). In comparison with a set of nuclear genes, where each gene, if chosen from a different chromosome, can provide an independent gene tree, mtDNA can provide only one independent estimate of a species tree since mitochondrial genes are inherited as a single linkage group (Moore 1995). However, using a coalescence theory argument, Moore (1995) concludes that a mitochondrial gene tree is more likely to be congruent with the species tree than is a nuclear gene tree. Because we were Interested in species limits in two closely related anopheline mosquitoes, Anopheles rangeli and A. trinkae, we chose mtDNA as our analytical tool. The distribution of A. rangeli includes most of northwestern South America (Colombia, Guyana, Venezuela, eastern Ecuador and Peru, Bolivia, and northwest-
Downloaded from jhered.oxfordjournals.org by guest on July 15, 2011
We analyzed variation in mitochondrial DNA (mtDNA) of two neotropical mosquitoes, Anopheles rangell (n = 181) and A. trinkae (n = 45), with very different distribution patterns in Latin America, to assess species boundaries for these putative sister taxa and to examine population genetic structure. Phylogenetic analyses revealed (1) support for the monophyletic origin of each species; (2) diagnostic restriction site differences between the species; (3) geographic partitioning of haplotypes by country in A. rangell from Bolivia, Ecuador, and Venezuela compared with considerable overlap in haplotypes of A. trinkae from Bolivia and Ecuador; and (4) similar levels of mean haplotype and nucleotlde diversity In both species, but lower levels of mean nucleotlde divergence in A. trinkae compared with A. rangell. We hypothesize that higher maternal gene flow and lower divergence in A. trinkae are most likely due either to a distinctive matrllineal history or to a smaller effective population size, which may have been Influenced by a smaller, essentially linear geographic range along the eastern flank of the Andes. In the cladistlc analysis of A. rangell, the Bolivian haplotypes appear to be more derived than those from Ecuador or Venezuela, yet there Is no evidence to support the hypothesis of a recent range expansion from Ecuador into Bolivia.
KNOWN DISTRIBUTION OF A. TRINKAE
VENEZUELA - RANGEU (Puerto CabeJIo)
TRINKAE (Puyo)
ECUADOR
DISTRIBUTION OF A. RANGEU COLLECTION SFTE (RANGEU) COLLECTION SITE (RANGEU + TRINKAE) TYPE LOCALITY
Figure 1. Collection sites and distribution of A ranged and A tnnkae. Collection site abbreviations are as In Table 1. The open circles In IB, SM, CO, and SY denote sites where both species were found In sympatry. The squares In the upper left Insert represent the extent of the known distribution (by collection site) of A tnnkae. Collections available for the present analysis were from Ecuador and Bolivia only.
ern Brazil) (Faran 1980; Peyton et al. 1983) while that of A trinkae is considerably more restricted (i.e., eastern Andean slopes from central Colombia to central Bolivia; Faran 1979, 1980; Wilkerson R and Lounibos LP, personal communication; Figure 1). Both species have been implicated as vectors of malaria parasites (Plas-
modium vivax and P. malariae) in Junin, eastern Peru, by salivary gland dissection (Hayes et al. 1987), and A rangeli has been found infected with P. vivax in the Caquetia-Putumayo region of southern Colombia using the EUSA technique (Suarez M, personal communication). A. rangeli is also a suspected malaria vector in Ecua-
Table 1. LUt of A. rangeli and A. trinkae collection sites, locations, dates, and sample size per site Population A rangeli PG PV SM IB SR SG CO JM SY LA PY VG EN A trinkae SM IB CO SY
Location
Country
Coordinates
Collection date(s)
Sample size
Puerto Grether, Santa Cruz Puerto VUlarroel, Cochabamba San Mateo, V. Tunarl, Cochabamba Ibuelo, V. Tunarl, Cochabamba San Ram6n, Rlberalta, Benl Sen. Gulomard, Acre Coca, Napo Juan Montalgo, Napo Sardlna Yacu, Napo Lago Agrlo, Sucumblos Paca Yacu, Sucumblos Vegulta, Barinas Camlno Real, El Nula, Apure
Bolivia Bolivia Bolivia Bolivia Bolivia Brazil Ecuador Ecuador Ecuador Ecuador Ecuador Venezuela Venezuela
17-10'S, 64°21'W 16°47'S, 64'45'W 16°56'S, 65°23'W 16°S3'S, 65°24'W 11°2'S,66°4'W 10-14'S, 67"35'W 0°22'S, 76°S4'W 0°33'S, 77°5'W 0°10'S, 77°S'W 0°5'N, 76°S3'W 0°5'N, 76°33'W 8*S2'N, 70°00'W 7*21'N, 71°52'W
15.VHI.91-, 25.X1.91 30.XI.91, 1JCII.91 28, 29JC1.91 29.XJ.91 4JUI.91 16, 18.K.92 5VM.92 7-9.Vm.92 6.VT11.92 4.Vm.92 ll.Vin.92 12.VI1I.93 13.Vin.93
42 18 2 4 8 3 36 54 3 3 2 3 2
San Mateo, V. Tunarl, Cochabamba Ibuelo, V. Tunarl, Cochabamba Coca, Napo Sardlna Yacu, Napo
Bolivia Bolivia Ecuador Ecuador
16*56'S, 6 5 * 2 3 ^ 16°53'S, 65°24'W
