Diptera: Psychodidae - Oxford Journals

1 downloads 0 Views 125KB Size Report
assumptions of panmixia within samples collected at each sampling point, the observed genotypic fre- quency distributions were tested against Hardy–.
ARTICLE

Gene Flow Between Natural and Domestic Populations of Lutzomyia longipalpis (Diptera: Psychodidae) in a Restricted Focus of American Visceral Leishmaniasis in Venezuela L. M. MA´RQUEZ,1 M. LAMPO, M. RINALDI,

AND

P. LAU2

Centro de Ecologõ´a, Instituto Venezolano de Investigaciones Cientõ´Þcas, Apartado 21827, Caracas-1020, Venezuela

J. Med. Entomol. 38(1): 12Ð16 (2001)

ABSTRACT The epidemiology of the visceral leishmaniasis in the Americas is associated with both a natural and a domestic cycle. The existence of reproductively isolated populations of Lutzomyia longipalpis (Lutz & Neiva), and the scarcity of records of this species from natural habitats in areas where it has been associated with domestic habitats indicated that natural populations could be genetically distinct from domestic ones. Therefore, we compared the genetic structure and estimated the gene ßow between L. longipalpis from domestic and peridomestic habitat and from an adjacent undisturbed natural environment along a 1.2-km transect. The analyses were performed on electrophoretic data from eight isozyme loci. The absence of Þxed differences in the diagnostic loci Ak and Hk indicated that all specimens belonged to one of the two cryptic species identiÞed in Venezuela. The average number of alleles per locus ranged from 2.0 to 2.9 and the average heterozygosity ranged from 7.8 to 13.4%. No differences were detected in the genetic structure of this species from domestic or peridomestic habitats and those trapped as far as 1.2 km from human dwellings. Nm, estimated from WrightÕs Fst, indicated that at least 208 individuals per generation migrated between the peridomestic habitat and a 1.2-km distant point to maintain the observed similarities in allelic frequencies. This high rate of gene ßow indicated that this species has high migration rates between domestic and natural environments, and has the potential to transport for Leishmania from natural to domestic environments. KEY WORDS Lutzomyia longipalpis, sand ßies, visceral leishmaniasis, gene ßow, isozyme electrophoresis, cryptic species

LAINSON (1983) SUGGESTED the existence of two cycles associated with American visceral leishmaniasis. In the enzootic cycle, the parasite Leishmania donovani chagasi Cunha & Chagas is maintained by reservoirs such as rodents, opossums, and foxes, and is transmitted among them by Lutzomyia longipalpis Lutz & Neiva. The domestic cycle involves the transmission of the parasite to dogs and humans by domestic populations of L. longipalpis. He also indicated the following three mechanisms by which the feral cycle may initiate the domestic cycle: (1) dogs and humans acquire infection when entering enzootic areas, (2) infected foxes may transit across domestic areas in search of food and infect peridomestic populations of L. longipalpis, and (3) infected sand ßies could migrate from natural to peridomestic habitats. However, it was not until a decade ago that the presence of high densities of L. longipalpis in a forest near two small rural foci in Brazil indicated the existence of large natural populations of these sand ßies (Lainson et al. 1990). More than 1,000 adults of L. longipalpis were collected during eight 1 Current address: Department of Biochemistry and Molecular Biology, James Cook University of North Queensland, Townsville, Qld. 4811, Australia. 2 Current address: Centro de Agroecologõ´a Tropical. Universidad Simo´ n Rodrõ´guez. Caracas, Venezuela.

trap nights in the forest, which was 500 m from the nearest house. These Þndings indicate that the enzootic cycle may be maintained by populations of L. longipalpis, and that migration of sand ßies from these natural areas to peridomestic habitats could play a critical role in the initiation of new foci (Lainson et al. 1990). The existence of reproductively isolated L. longipalpis populations has been explored for some time. Mangabeira (1969) suggested the existence of two morphologically distinct forms of L. longipalpis that differed in their degree of association to domestic habitats and in their preference for human hosts. Although the 2-spot phenotype always was associated with houses and characterized as highly anthropophilic, the 1-spot phenotype rarely entered houses or bit humans. The Þndings of Ward et al. (1983) indicated the existence of additional forms, and discussed the possibility that L. longipalpis represented more than a single taxon. However, the discovery of intermediate forms (Ward et al. 1988), and the fact that no isoenzymatic differences were detected between the 1- and 2-spot phenotypes relegated spot morphology to an intraspeciÞc polymorphism (Mukhopadhyay et al. 1998). The presence of 1- and 2-spot phenotypes in samples from urban and peri-urban zones (Mukho-

0022-2585/01/0012Ð0016$02.00/0 䉷 2001 Entomological Society of America

January 2001

M´ARQUEZ ET AL.: GENE FLOW IN L. longipalpis

padhyay et al. 1998) indicated that variation in this character was not related to the degree of domesticity. However, the presence of reproductively isolated sympatric populations in a small town in Venezuela indicated that cryptic species may be sympatric within domestic habitats (Lampo et al. 1999). Whether L. longipalpis associated with enzootic cycles differs genetically from ßies present in domestic areas remains to be determined because the genetic data available for L. longipalpis is based on specimens caught exclusively within domestic or peridomestic habitats. Possibly, a behavioral shift has permitted sand ßies to move to domestic areas, and little gene ßow exists between these differentiated populations and natural ones. We used data from eight isozyme loci to test whether differences occur in the genetic structure of natural and domestic populations of L. longipalpis that conform to genetically distinct populations. Implications for vector control are discussed. Materials and Methods Study Site. The study site was located in the agricultural valley of Curarigua (Lara State), Venezuela (69⬚ 55⬘ W, 9⬚ 59⬘ N; 600 m asl). The climate is semiarid with two rainy periods, one in May and another in October, with a mean annual rainfall of 636.6 mm and a mean monthly temperature of 25.4⬚C (data from the Venezuelan Ministry of Natural and Renewable Resources). The community of La Rinconada, a rural settlement and endemic focus of visceral leishmaniasis with ⬇80 houses and ⬇600 inhabitants (Mazzarri et al. 1997), was designated as the domestic habitat. For the natural environment, we chose a xerophytic shrub forest adjacent to La Rinconada that had no signs of human disturbance. Foxes, bats, and small rodents were observed in this area. Phlebotomine sand ßies were collected between May and September 1997 using Center for Disease Control (CDC) light traps operated between 1800 and 0600 hours. In the domestic habitat, traps were placed inside and outside the houses to include intra- and peridomestic habitats. In the natural habitat, traps were placed along a transect at 500, 1,000, and 1,200 m from the peridomestic habitat. The number of trapnights within each habitat varied to ensure that at least 200 adults of L. longipalpis were collected at each site. To compare samples from a more distant location, 106 individuals were collected in January and September 1998 within and outside houses from El Paso, a rural community 8 km from La Rinconada. Live adults were cold-anesthetized at approximately Ð2⬚C for about 10 min. For males, species diagnosis was carried out based on the presence of dorsal curved setae inserted directly on the paramere (Young and Duncan 1994). Males were stored in liquid nitrogen. For females, the distal abdominal tergites and heads were separated and stored in ethanol 70%, whereas the rest of the bodies, mainly thorax and the proximal tergites, were stored in liquid nitrogen. The distal abdominal tergites and heads of females were

13

cleared and mounted in Berlese solution, and species diagnosis was based on the morphology of the spermathecae and the cibaria (Young and Duncan 1994). Electrophoretic Methods. Standard vertical polyacrylamide gel electrophoresis protocols were used (Black IV and Munstermann 1996). Each sample was homogenized in 20 ␮l of grinding solution (10% sucrose, 0.1% Triton X-100, 0.2 M Tris-citrate buffer, and 0.002% [wt:vol] bromophenol blue) and distributed among eight gels. Tris-citrate and Tris-borate buffer systems were used to maximize electrophoretic enzyme separation. Eight enzyme assays revealed distinguishable phenotypes: glucose-6-phosphate isomerase (Gpi, E.C.5.3.1.9), hexokinase (Hk, E.C.2.7.7.1), and malate dehydrogenase, decarboxylating (Me, E.C.1.1.1.40) for Tris-borate buffer gels (TBE); and malate dehydrogenase (Mdh, E.C.1.1.1.37), isocitrate dehydrogenase (Idh, E.C.1.1.1.42), glycerol-3-phosphate dehydrogenase (Gpd, E.C.1.1.1.8), adenylate kinase (Ak, E.C.2.7.4.3), and arginine kinase (Ark, E.C.2.7.3.3) for Tris-citrate buffer gels (TC). The banding phenotypes of these eight loci were examined. Numerical values for the allelic phenotypes (electromorphs) were based on the relative migration of bands from the gel origin. One Aedes aegypti (L.) from the Rockefeller strain was included as a reference standard on some gels. For each locus, the Ae. aegypti electromorph was designated as the “1.00” electromorph. Following convention, loci were denoted as polymorphic if the most common allele had a frequency ⬍95% in at least one population. Analyses Methods. The proportion of polymorphic loci, the average heterozygosity, and the mean number of alleles per locus were quantiÞed. To test the assumptions of panmixia within samples collected at each sampling point, the observed genotypic frequency distributions were tested against HardyÐ Weinberg equilibrium frequencies using BIOSYS (Swofford and Selander 1981). Allele frequencies between sampling points were compared using PearsonÕs chi-square contingency test algorithm in BIOSYS (Swofford and Selander 1981). The structure of genetic variation among sampling points and dates was examined with F-statistics developed by Wright (1965) and modiÞed by Nei (1977). Means and standard deviations over all loci for FIS and FST were calculated using a jacknife method (Weir and Cockerham 1984) to assess the degree of inbreeding within and the degree of differentiation among sampling points. The number of migrants per generation, Nm, was estimated from values of FST as an indication of gene ßow between sampling points. Finally, a dendrogram based on the modiÞed RogersÕ distance (Wright 1978) was constructed to determine the genetic relationship between samples from the domestic and natural habitats using the unweighted pair group method algorithm implemented in BIOSYS (Swofford and Selander 1981).

14

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 38, no. 1

Table 1. Allelic frequencies for eight isozyme loci from five populations of L. longipalpis from Venezuela Locus (Rf) Ak 1.08 0.82 n Ark 0.89 n Gpd 1.51 1.34 1.17 n Gpi 0.48 0.38 0.26 n Hk 0.83 0.77 n Idh-2 0.77 0.64 0.44 n Mdh 0.69 0.58 0.43 n Me 1.00 0.93 0.88 n

Domestic (LR)

500m (LR)

1,000m (LR)

1,200m (LR)

Domestic (EP)

0.981 0.019 188

0.972 0.028 123

0.967 0.033 90

0.969 0.031 128

1.000 0 98

1.000 198

1.000 156

1.000 156

1.000 152

1.000 106

0.011 0.989 0 185

0.006 0.981 0.013 157

0.003 0.968 0.029 155

0.018 0.968 0.014 140

0 1.000 0 78

0.081 0.839 0.081 180

0.063 0.875 0.063 160

0.081 0.816 0.103 179

0.072 0.834 0.093 145

0.053 0.856 0.090 94

0.006 0.994 180

0 1.000 145

0 1.000 154

0.004 0.996 134

0.014 0.986 105

0.043 0.993 0.023 150

0.036 0.923 0.040 124

0.051 0.911 0.037 107

0.030 0.945 0.025 100

0.018 0.970 0.012 84

0.008 0.864 0.128 199

0.013 0.843 0.144 156

0.014 0.891 0.095 179

0.019 0.825 0.157 134

0.005 0.880 0.115 104

0.013 0.922 0 198

0.013 0.903 0.109 149

0.004 0.882 0 127

0.030 0.883 0.042 149

0 0.952 0 104

LR, La Rinconada; EP, El Paso; Rf ratio of the distance migrated from the gel origin to the distance migrated for the same enzyme by the reference standard.

Results Overall, 193 specimens were analyzed from the domestic habitat and 491 from three distances from houses: 160 from 500 m, 179 from 1,000 m, and 152 from 1,200 m. Previous analyses showed that for most enzymes the most common allele had a frequency of ⬎90%. Hence, we detected differences in allele frequencies as small as 10% (5% conÞdence level, and power of 80%) between sampling points (Baverstock and Moritz 1990). Table 1 summarizes gene frequency data for each sampling point. Except for Ark, all loci were variable. Gpi and Mdh showed the largest mean heterozygosities (⬎20%), followed by Idh and Me (10 Ð20%), Gpd and Ak (1Ð10%), and Hk (⬍1%). At La Rinconada the proportion of polymorphic loci was 50% for all sampling points and the mean heterozygosity over all loci varied between 10.7 and 13.4% within sampling points. Mean heterozygosity at El Paso was 7.8%, and only 25% of all loci were polymorphic. This difference may be attributed to the smaller sample size, as most variation was explained by the presence of low frequency alleles in some samples. The presence of alleles Ak1.08 and

Fig. 1. Unweighted pair-group method with arithmetic average dendrogram based on RogerÕs modiÞed genetic distance (Wright 1978) for L. longipalpis from natural and domestic habitats in Venezuela. LR, LA, Rinconada; EP, El Paso.

Hk0.77 and the absence of Ak1.21 and Hk0.69 in all samples indicated that all electromorphs corresponded to only one of the two subspecies of cryptic L. longipalpis present in this area (Lampo et al. 1999). Samples from natural and domestic habitats did not depart from panmixia, nor were they genetically different from each other at La Rinconada. None of the loci at any of these locations deviated signiÞcantly from HardyÐWeinberg expectations (P ⬎ 0.05). The mean FIS value within La Rinconada was not statistically different from zero, indicating no evidence of heterozygote deÞciencies within sampling points at this location. A slight reduction in the heterozygosity was evidenced by a small, but positive mean FST value (0.0012 ⫾ 0.00082) when sampling points within La Rinconada were pooled together. The mean migration rate, Nm, between these sampling points was estimated to be 208 individuals per generation. However, we found no signiÞcant heterogeneity in the allele frequencies for any of the loci among all sampling points at La Rinconada (␹2 ⫽ 47.43, df ⫽ 36, P ⫽ 0.101). In contrast, signiÞcant differences were detected in the frequency of rare alleles for Ak (␹2 ⫽ 5.31, df ⫽ 1, P ⫽ 0.021), Hk (␹2 ⫽ 6.04, df ⫽ 1, P ⫽ 0.019), and Me (␹2 ⫽ 6.76, df ⫽ 2, P ⫽ 0.0340) between La Rinconada and El Paso. Alleles Ak0.82 and Me1.00 present in all samples from La Rinconada were absent from El Paso. However, the frequency of Hk0.83 was signiÞcantly lower at La Rinconada than at El Paso. Small but signiÞcant genetic variation between La Rinconada and El Paso also was evidenced. The mean FST value doubled (0.0024 ⫾ 0.00293) when the sample from El Paso was pooled with those from La Rinconada. The mean migration rate, Nm, between these two locations was estimated to be 104 individuals per generation. RogersÕ genetic distances among samples within La Rinconada were very low (0.016 Ð 0.028) and did not separate sampling points according to habitat type (natural versus domestic) (Fig. 1). Within La Rinconada, samples grouped according to their vicinity, except for the samples at 1,000 and 1,200 m, which switched positions. However, a regression analysis between genetic and geographic distances within La Rinconada did not show any signiÞcant relationship between these two variables (R2 ⫽ 0.296, F ⫽ 2.911;

January 2001

M´ARQUEZ ET AL.: GENE FLOW IN L. longipalpis

df ⫽ 1, 8; P ⫽ 0.125). However, genetic distances between samples from La Rinconada and those from El Paso were slightly higher (0.020 Ð 0.037). In the dendrogram, samples from La Rinconada formed a distinct group separated from those collected at El Paso (Fig. 1). Discussion Closely related species may differ in their degree of association with human habitats. Within the Anopheles farauti complex, for example, the antropophilic and nonantropophilic forms differ genetically (Foley et al. 1994). Genetic differences between the commensal Aedes aegypti aegypti and the feral Ae. aegypti formusus also have been detected, although sufÞcient gene ßow between both forms presumably has prevented populations from forming distinct species (Scott and McClelland 1975, Tabachnick et al. 1979, Moore 1979, Ballinger-Crabtree et al. 1992). Despite evidence indicating differences in the degree of association to domestic habitats in reproductively isolated populations of L. longipalpis (Ward et al. 1983), we found no evidence of genetic differentiation between sand ßies caught in natural and domestic or peridomestic habitats. Genetic similarities among samples collected at La Rinconada indicated the existence of a genetically homogeneous L. longipalpis population that extended from houses to at least 1.2 km into the natural habitat. Similar results were found for Culex pipiens and Aedes aegypti. Although a human commensal and a feral form of C. pipiens differing in some life history traits have been identiÞed, no evidence of genetic differentiation was found between both ecotypes (Chevillon et al. 1995). SufÞcient gene ßow between both types of habitats presumably has prevented populations from diverging either through genetic drift or speciation. The structure of the genetic variability of L. longipalpis from La Rinconada and El Paso is very similar to that of L. longipalpis from Central Colombia (Munstermann et al. 1998), despite the fact that they may represent different cryptic species (Lampo et al. 1999). Munstermann et al. (1998) suggested that although L. longipalpis populations tend to be homogeneous within a 20-km radius, some population substructure may be detected between populations separated by ⬇1 km. At La Rinconada, some substructuring also was evidenced between sampling points separated by 1.2 km. However, high migration rates also indicated minimal gene ßow restriction between populations separated by 8 km. In fact, migration rates between La Rinconada and El Paso seem higher (Nm ⫽ 104) than that between Colombian populations separated by 10 Ð15 km [Nm ⫽ 36; recalculated from Munstermann et al. (1998) and based on the eight loci used for Venezuelan populations]. Uniformity of the landscape between these two Venezuelan locations may have facilitated dispersal because suitable habitats probably occured between both localities. The high gene ßow recorded between domestic and natural habitats at La Rinconada indicated that this

15

cryptic L. longipalpis species may play an important role in the transport of Leishmania chagasi between enzootic and domestic cycles. Not only may new foci be initiated by infected sand ßies dispersing from natural to domestic areas (Lainson et al. 1990), but these foci also could be maintained by the high number of infected sand ßies that move between both habitats if the prevalence of infection in reservoir poplations is high. In such cases, control strategies that aim to reduce the sand ßy population in domestic areas should take into account recolonization from adjacent natural habitats.

Acknowledgments We are grateful to Yadira Rangel, M. Dora Feliciangeli, Moritz Benado, Alvaro Rodrõ´guez Leonard Munstermann, and anonymous reviewers for their comments and suggestions. We also thank Carolina Bastidas, Arturo Bravo, Hermes Pin˜ ango, and the personnel of the Center of Ecology of the Venezuelan Institute for ScientiÞc Research (IVIC) for their help during Þeld and laboratory work. This study was supported by the Consejo Nacional de Investigaciones Cientõ´Þcas y Tecnolo´ gicas (CONICIT) (No. 96 Ð 0001370) and the WHO-World Bank (No. 021Ð 007).

References Cited Ballinger-Crabtree, M. E., W. C. Black IV, and B. R. Miller. 1992. Use of genetic polymorphisms detected by the random-ampliÞed polymorphic DNA polymerase chain reaction (RADP-PRC) for differentiation and identiÞcation of Aedes aegypti subspecies and populations. Am. J. Trop. Med. Hyg. 47: 893Ð901. Baverstock, P. R., and C. Moritz. 1990. Project design, pp. 17Ð27. In D. M. Hillis, C. Moritz, and B. K. Mable [eds.], Molecular systematics. Sinauer, Sunderland, MA. Black IV, W. C., and L. E. Munstermann. 1996. Molecular systematics in vector biology, pp. 432Ð 470. In B. J. Beaty and W. C. Marquardt [eds.], Biology of disease vectors. University Press of Colorado, Niwot, CO. Chevillon, C., R. Eritja, N. Pasteur, and M. Raymond. 1995. Commensalism, adaptation and gene ßow: mosquitoes of the Culex pipiens complex in different habitats. Genet. Res 66: 147Ð157. Foley, D. H., S. R. Meek, and S. H. Bryan. 1994. The Anopheles punctulatus group of mosquitoes in the Solomon Islands and Vanuatu surveyed by allozyme electrophoresis. Med. Vet. Entomol. 8: 340 Ð350. Lainson, R. 1983. The American leishmaniases: some observations on their ecology and epidemiology. Trans. R. Soc. Trop. Med. Hyg. 77: 569 Ð595. Lainson, R., C. Dye, J. J. Shaw, D. W. Macdonald, O. Courtenay, A.A.A. Souza, and F. T. Silveira. 1990. Amazonian Visceral LeishmaniasisÑ distribution of the vector Lutzomyia longipalpis (Lutz & Neiva) in relation to the fox Cerdocyon thous (Linn.) and the efÞciency of this reservoir host as a source of infection. Mem. Inst. Oswaldo Cruz 85: 135Ð137. Lampo, M., D. Torgerson, L. M. Ma´ rquez, M. Rinaldi, C. Z. Garcı´a, and A. Arab. 1999. Occurrence of sibling species of Lutzomyia longipalpis (Diptera: Psychodidae) in Venezuela: Þrst evidence from reproductively isolated sympatric populations. Am. J. Trop. Med. Hyg. 61: 1004 Ð1009.

16

JOURNAL OF MEDICAL ENTOMOLOGY

Mangabeira, O. 1969. Sobre a sistema´tica e biologia dos Phlebotomus do Ceara´. Rev. Bras. Malariol. Doencas. Trop. 21: 3Ð26. Mazzarri, M. B., M. D. Feliciangeli, M. Maroli, A. Herna´ ndez, and A. Bravo. 1997. Susceptibility of Lutzomyia longipalpis (Diptera:Psychodidae) to selected insecticides in an endemic focus of visceral leishmaniasis in Venezuela. J. Am. Mosq. Control Assoc. 13: 335Ð341. Moore, D. F. 1979. Hybridization and mating behavior in Aedes aegypti. J. Med. Entomol. 16: 223Ð226. Mukhopadhyay, J., K. Ghosh, A.C.R. Azevedo, E. F. Rangel, and L. E. Munstermann. 1998. Genetic polymorphism of morphological and biochemical characters in a Natal, Brazil, population of Lutzomyia longipalpis (Diptera:Psychodidae). J. Am. Mosq. Control Assoc. 14: 277Ð282. Munstermann, L. E., A. C. Morrison, C. Ferro, R. Pardo, and M. Torres. 1998. Genetic structure of local populations of Lutzomyia longipalpis (Diptera: Psychodidae) in Central Colombia. J. Med. Entomol. 35: 82Ð 89. Nei, M. 1977. Statistics and analysis of gene diversity in subdivided populations. Ann. Hum. Genet. 41: 225Ð233. Scott, J. A., and G.A.H. McClelland. 1975. Electrophoresis differences between sympatric ecotypes. Nature (Lond.) 256: 405Ð 406. Swofford, D. L., and R. B. Selander. 1981. BIOSYS-1: a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. J. Hered. 72: 281Ð283. Tabachnick, W. J., L. E. Munstermann, and J. R. Powell. 1979. Genetic distinctness of sympatric forms of Aedes aegypti in East Africa. Evolution 33: 287Ð295.

Vol. 38, no. 1

Ward, R. D., A. L. Ribeiro, P. D. Ready, and A. Murtagh. 1983. Reproductive isolation between different forms of Lutzomyia longipalpis (Lutz & Neiva) (Diptera: Psychodidae), the vector of Leishmania donovani chagasi Cunha & Chagas and its signiÞcance to kala-azar distribution in South America. Mem. Inst. Oswaldo Cruz 78: 269 Ð280. Ward, R. D., A. Phillips, B. Burnet, and C. B. Marcondes. 1988. The Lutzomyia longipalpis complex: reproduction and distribution, pp. 257Ð269. In M. W. Service [eds.], Biosystematics of haematophagous insects. Clarendon, Oxford, UK. Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38: 1358 Ð1370. Wright, S. 1965. The interpretation of population structure by F-statistics with special regard to systems of mating. Genetica 19: 395Ð 420. Wright, S. 1978. Evolution and the genetics of populations, vol. 4. Variability within and among natural populations. University of Chicago Press, Chicago. Young, D. G., and M. A. Duncan. 1994. Guide to the identiÞcation and geographic distribution of Lutzomyia sand ßies in Mexico, the West Indies, Central and South America (Diptera: Psychodidae), vol. 2. Associate Publishers, Gainesville, FL. Received for publication 22 September 1999; accepted 24 March 2000.