Random Amplification of Polymorphic DNA as a Tool forTaxonomic Studies of Triatomine Bugs (Hemiptera: Reduviidae )

ANA L. GARCIA,l HERNAN J. CARRASCO,2 C. J. SCHOFIELD,2 J. RUSSELL STOTHARD,2

IAIN A. FRAME,2 SEBASTIAO A. S. VALENTE,3 AND MICHAEL A. MILES2

J. Med Entorno]. 35(1).38-45 (1998)ABSTRACT Eleven of 27 decameric primers were found to be suitable for random amplifi-cation of polymorphic DNA (RAPD) from triatomine bugs on the basis that they produceddiscrete profiles and distinguished among Panstrongylus megistus (Burmeister), Rhodnius pro-lixus Stal, and Triatoma infestans (Klug) .The legs, or single leg segments, of individual bugs wereused as the source of DNA so that the taxonomic value of the bug was conserved. Within thescope of the specimens studied, RAPD profiles allowed assignment to species even when bugswerekept dry for up to 12 mo.Profiles for individuaIs within a species were not identical. RAPDprofiles, with the specimens tested, distinguished among species of 3 pairs considered to bemorphologically similar and closely related, namely, Rhodnius ecuadorensis Lent & León andRhodnius pictipes Stal; Rhodnius nasutus Stal, and Rhodnius neglectus Lent; Rhodnius prolixus Staland Rhodnius robustus Larrousse. RAPD data conformed with the perceived affinities amongthese species. RAPD polymorphisms were seen with T. infestans from 3 different localities, butnone of the polymorphisms was confined to 1 source. RAPD provided a molecular basis toreassess taxonomic I:elationships within the Triatomine subfamily. The accurate distinction oftriatomine species and of intraspecific bug populations may contribute to elimination of vector-borne Chagas disease from the Americas.

KEY WORDS Panstrongylus, Rhodnius, Triatoma, triatomines, random amplification of poly-morphic DNA, molecular taxonomy

CHAGAS DISEASE (AMERICAN trypànosomiasis) is oneQf the most important public health problems inLatin America. The kinetoplastid protozoan parasiteresponsible, Trypanosoma cruzi, is thought to infectup to 18 million people, and another 100 million livein endemic areas where they are at risk of acquiringinfection (WHO 1991) .Arthropod vectors of T.cruzi are blood-sucking triatomine bugs(Hemiptera: Reduviidae) .Transmission of the par-asite occurs while bugs are feeding if T. cruzi in-fected bug feces contaminates the mucous mem-branes or abraded skin of the mammalian host.Transmission can also occur through contaminatedblood transfusion, organ transplant, congenitally,and by oral coptamination. The most frequent causeof infection is triatomine attack in infested houses.The initial acute phase of T. cruzi infection is seldomrecognized and may be asymptomatic. There is nosatisfactory treatment for chronic infections, and novaccine, in part because pathogenesis appears toinvolve autoimmunity. Therefore, control ofChagasdisease relies principally on elimination of domestic

1 Facultad de Medicina, Universidad Mayor de San Simon, Casilla

3119, Cochabamba, Bolívia.2 Department of Infectíous and Tropical Diseases, London

School of Hygiene and Tropical Medicine, Keppel Street, LondonWC1E 7HT, U.K.

3 Serviço de Parasitologia, Programa de Doença de Chagas, In-

stítuto Evandro Chagas, Belém, Pará, Brazil.

triatomine populations by spraying with residualinsecticides, supported by health education andhousing improvement, in conjunction with screen-ing and treatment of suspect transfusion blood. Ifsustained by adequate resources, these publichealth measures are effective. They have been usedto eliminate domestic Triatoma infestans (Klug)from Sao Paulo state, Brazil (Wanderley 1993), and,with residual pyrethroid insecticides, are the basisfor the international Southern Cone Programme toeliminate T. infestans in Argentina, Bolivia, Brazil,Chile, Paraguay, and Uruguay (Dias 1993).

Residual infestation and reinfestation of sprayedhouses obstruct control campaigns, especially ifspraying does not cover the entire endemic focus.Recolonization can arise from untreated domestichabitats or by immigration ofbugs from silvatic foci(Schofield 1985, Harry et al. 1992), and peridomes-tic or silvatic species may replace a domestic vectorthat has been removed by insecticides. Thus, theaccurate identification of triatomine species, someof which are morphologically similar but differ inhabitat, behavior, and potential for domiciliation,and the distinction of intraspecific bug populations,has a significant role to play in the elimination ofChagas disease from the Americas.

Triatomine species are traditionally identified us-ing morphological criteria, to which the most com-prehensive guide is the monograph by Lent and

0022-2585/98/0038-0045$02.00/0 @ 1998 Entomological Society of America

GARCIA ET AL.: MOLECULAR TAXONOMY OF TRIATOMINAE 39January 1998

Table I. Species and origins of triatomine bngs (colonies orcollections) nsed as sonrces of ONA for RAPO analyses

IEC, BrazilIEC, BrazilIEC, BrazilIEC, Brazi\IEC, BrazilIEC, BrazilLSHTM, U.KIEC, BrazilColombiaoIEC, Brazi\Cumetrop, Bo\iviaLSHTM, UK.IEC, Brazi\

IEC, Instituto Evandro Chagas, Belém, Pará State, Brazil;LSHTM, London School of Hygiene and Tropical Medicine, Kep-pel Street, London WCIE 7HT, U.K.

a Red-eyed mutant; courtesy of Chris Curtis and the Instituto

Nacional de Salud (Bogota, Colombia).

Materiais and Methods

Triatomine Bugs. The domestic triatomines P.megistus, R. prolixus, and T. infestans were selectedto represent distinct and divergent species that areepidemiologically important as vectors of T. cruzi.Within the genus Rhodnius we examined 6 species(Table 1) and included 3 pairs of species that aresaid to be morphologically similar and closely re-lated, namely: R. ecuadoriensis Lent & León and R.pictipes Stal; R. nasutus Stal and R. neglectus Lent; R.prolixus and R. robustus Larrousse. T. infestans carnefrom 3 geographically separate collection sites atTarija, Cochabamba, and Sucre in Bolivia. The ma-jority ofbugs were from established laboratory col-onies (Table 1) , except that R. pictipes and R. ro-bustus were freshly collected in 1995 from theirsilvatic habitats in Pará state, Brazil, using methodsdescribed by Miles et al. ( 1981) ; T. infestans wereprogeny of recent field collections in Bolivia. In thecase of R. pictipes and R. robustus, new field collec-tions were made because pure bred laboratory col-onies were not available. Species identification wasdetermined by morphology as described by Lentand Wygodzinsky (1979). Some bug specimenswere kept dry for up to 12 mo so that RAPD profilescould be compared with those of living insects.

DNA Samples. Each bug was examined individu-ally. Legs were selected as the tissue of choice forthe source of DNA because they were free of host(blood meal) contaminants and gut symbionts, andthey could be removed without destroying the tax-onomic value of the specimen. For some individuaisthe 61egs were pooled before DNA extraction; withothers, samples were obtained from a single leg or1 leg segment. Legs were ground to a fine powderwith tissue grinders (Kontes, Vineland, NJ) in a1.5-ml microcentrifuge tube that contained liquidnitrogen. After grinding, the liquid nitrogen wasallowed to evaporate and the powder resuspendedin 1 ml oflysis buffer (50 mM NaCI, 50 mM EDTA,

Wygodzinsky (1979) .On the basis of morphology,123 species of the subfamily Triatominae have beendescribed, grouped into 5 tribes and 14 genera (Lentand Wygodzinsky 1979, Schofield 1994). There are110 species restricted to the Americas, and of the 13Old World species, 8 are part of the T. rubrofasciatagroup and 5 species form the Indian genus Linsh-costeus ( Gorla et al. 1997) .Most triatomine specieslive in silvatic habitats, such as palm trees, hollowtrees, burrows, among rocks, or in caves, feedingfrom mammals, birds, or reptiles ( only mammals aresusceptible to T. cruzi infection) .The majority ofNew World triatomine species have been recordedas infected with T. cruzi but the 5 most importantdomestic vectors falI into 3 genera and the 2 tribesTriatomini [ Panstrongylus megistus ( Burmeister ) , T.infestans, T. brasiliensis Neiva, T. dimidiata (La-treille) ] and Rhodniini ( Rhodnius prolixus Stâl) .

Some bug species, such as R. prolixus and R. ro-bustus Larrousse, are difficult to separate taxonom-ically with confidence (Schofield 1994) and othershave controversial taxonomic status. Few studieshave attempted any cladistic analysis within a majorgenus, or between the 5 tribes of triatominae, or theabundant N ew World species and those few speciesrestricted to the- Old World ( Gorla et al. 1997) .Theevolutionary histories of triatomine taxa are there-fore uncertain. Isoenzyme electrophoresis (Doreaet al. 1982, Dujardin et al. 1987, Harry et al. 1992,Garcia et al. 1995a, Pereira et al. 1996) and cytoge-netic studies (Perez et al. 1992) have been intro-duced to assist morphological characterization.Isoenzyme ptofiles have been applied to examinethe population genetics of silvatic and domesticpopulations of T. infestans (Dujardin et al. 1987,Garcia et al. 1995b) and R. prolixus (Harry et al.1992, Lopez and Moreno 1995) and to detect geneflow between populations, but a small number ofpolymorphic loci are amenable to isoenzyme anal-ysis and living or freshly frozen samples are neces-sary. Cytogeneticmethods are time consuming andaffected by both the life cycle stage and the sex ofthe triatomine specimen.

Molecular biological methods can provide highlydiscriminatory markers to clarify triatomine taxon-omy and population genetics. One relativelystraightforward approach is genetic fingerprintingby random amplification of polymorphic DNA(RAPD) in polymerase chain reactions (PCR) withsingle short oligonucleotide primers of arbitrary se-quence. RAPD has become widely adopted for ge-nome analysis and population genetics of variousinsect species (Black 1993, Haymer 1994, Wilkersonet al. 1995) especially those species that are prob-lematic with classical taxonomic approaches (Ad-amson et al. 1991, Ballinger-Crabtree et al. 1992,Kambhapati et al. 1992, Favia et al. 1994). Here wehave examined the potential of RAPD protocols fortaxonomic and epidemiologic studies of triatomine

bug populations.

Vol. 35, no.JOURNAL OF MEDICAL ENTOMOLOGY40

Table 2. Eleven (of 27) decameric oligonncleotide primersthat gave RAPD profiles with template DNA from triatomine bngsand distingcnished P. megislus, P. prolíxus, and T. infeslu1's

5 6 7 8 9 10 11 12M 2 3 4

5'-3' SequencePrimer designation

10 11 12 M2 3 4 6 7M

Fig. 1. RAPD profiles with DNA templates isolatedfrom fresh or dry ( 12 mo) specimens, with primers A2(top) and A5 (bottom) .Tracks: (1 and 3) R. prolixus fresh,(2 and 4) R prolixus dry, (5) R. nasutus fresh, (6) R.nasutus dry, (7) R. neglectus fresh, (8) R. neglectus dry, (9)T. infestans fresh, (10) T. infestans dry, (11) P. megistusfresh, (12) P. megistus dry. Track M contains 1-kb laddermarkers (Gibco).

phoresis in polyacrylamide ( 6%, in 1 X TBE buffer )and silver staining according to the conditions de-scribed by Vidigal et al. 1994.

Results

Discrete RAPD profiles were obtained with tis-sues from Iive triatomine bugs and from dry bugsstored for up to 12 mo at room temperature (Fig. 1) .Profiles for alI individuaIs within a species were notprecisely identicaI, but profiles for dried specimenswere compatible with those for Iive insects. Speci-mens were, with the Iimited scope of the samplesstudied, readily assignable to species on the basis ofRAPD profiles (Fig. 1, tracks 5- 8) .Conditions werecontrolled such that the concentrations of reactantsand cycling conditions were unaltered, and a singlecommerciaI source of Taq polymerase was used.Suitable templates could be obtained without Gene-clean (not shown). Clearly, identifiable profileswere given from alI Iife cycle stages from eggs( whole organism as samples) , nymphs (Iegs or sin-gIe Ieg segment) , and adults (Iegs or single Iegsegment) although some profile diversity was againseen between individuaIs (Fig. 2, primer A2).

1% SDS, 50 mM Tris, pH 8.0) containing proteinaseK (Sigma;St. Louis, MO) at 2 mg mIl. The suspen-sion was incubated overnight ( 37°C) , gently ex-tracted sequentially with equal volumes of phenol,phenol-chloroform, and chloroform, centrifuged(7,000 X g, 10 min, 4°C) and DNA purified from thesupernatant using the Geneclean method ( Gene-clean II, Bio 101, La Jolla, CA) adapted from themanufacturer's instructions. The resultant DNA wasresuspended in 100 JLl TE buffer (10 mM Tris, 1 mMEDT A, pH 7.2 ) ; 5 JLl was electrophotesed in agarose(0.8%, in TAE buffer) and stained with ethidiumbromide to check for integrity and purity. Concen-tration and purity were further determined with a10-JLl sampleby spectrophotometry (at 260 nm and280 nm, assuming 1 absorbance unit at 260 nm cor-responds with 50 JLg mIl dsDNA). Samples were

\stored at +4°C.

Random Amplification ofPolymorphic DNA. TheRAPD protocol (Carrasco et al. 1996) was modifiedfrom Williams et al. ( 1990) .Reaction mixtures of 20JLl contained 0.2 mM dATP, dCTP, dGTP, and dTTP(Pharmacia LKB Biotechnology, Uppsala, Sweden) ,20 pmol oligonucleotide primer, 1.0 U Taq DNApolymerase (Stratagene, Cambridge, U.K.), =5 ngtemplate DNA, 0.001 gelatin (in 1.5 mM MgCI2, 50mM KCL, 10 mM Tris HCl, pH 8.8) overlaid with 40Jil ofmineral oil (3416, Sigma). PCR primers (Brit-ish Biotechnology Products, Oxon, U .K. ) givingRAPD profiles that distinguish between triatominegenera are listed in Table 2. Amplifications wereperformed on a programmable thermal cycler (Hy-baid,Middlesex, U.K.) with the following cycle con-ditions: 94°C for 5 min, 30°C for 2 min, 72°C for 1 min(2 cycles); 94°C for 30 s, 40°C for 2 min, 72°C for 1min ( 32 cycles) ; 95°C for 30 s, 40°C for 2 min, 72°Cfor 5 min ( 1 cycle) .N egative control PCR reactionsincluded alI reaction components except templateDNA, made up to an equal vblume with sterilewater.

Random amplification of polymorphic DNA prod-uctswere electrophoresed in agarose (2%, in TAE)stained with ethidium bromide and recorded on agel documentation system with a UV light transil-luminator. In addition, the RAPD products from R.prolixus and R. robustus were compared by electro-

January 1998 GARCIA ET AL.: MOLECULAR TAXONOMY OF TRIATOMINAE 41

M 1 2M 1 2 3 4 5 6 7 5 6 7 8 9

M 1 2 3 4 5 6 7 8 9

M 2 3 4 51 6 7

Fig. 3. RAPD profiles for representatives of the 3 gen-era, Rhodnius, Triatoma, and Panstrongylus, were distinctwith primers A2 (top) and L5 (bottom). Tracks: (1-5) R.prolixus, (6 and 7) T infestans, (8 and 9) P. megistus. TrackM contains l-kb ladder markers.

Fig. 2. RAPD profiles were comparable from differentlife cycle stages: R prolixus, with primers A2 ( top ) and L5(bottom). Tracks: (1) R prolixus egg, (2-6) R. prolixuslst-5th stage nymphs, (7) R prolixus adult. Track M con-tains l-kb ladder markers.

Of 27 decameric primers tested, 11 gave bothdiscrete RAPD profiles and distinguished among P.megistus, R. prolixus, and T. infestans. Examples ofresults with 4 of the 11 primers are shown in Figs.1- 6. Examples of RAPD profiles for the 3 generawith the 2 primers A2 and L5 are shown in Fig. 3; fewprofile components were shared across the genera.

We determined whether RAPD profiles could dis-tinguish between species within the same genus,including species that were thought to be closely

related, such as R. ecuadorensis and R. pictipes, R.nasutus, and R. neglectus, or for which the taxonomicstatus is controversial ( R. prolixus and R. robustus )and also to see whether RAPD data gave any indi-cation of affinities between species. The sharing ofprofile components across such species, with prim-ers that were highly discriminatory for the 3 genera,did suggest a relationship between the species pairs( Fig. 4) .Yet in each case, profile components sep-arated species pairs and might prove to be species-specific characters. Thus, 3 bands distinguished be-tween R. pictipes and R. ecuadorensis, 2 between R.nasutus and R. neglectus, and 2 between R. prolixusand R. robustus, with primer A2.

Random amplification of polymorphic DNA pro-files of individual adult T. infestans from 3 differentlocalities in Bolivia were compared using 4 primers.Polymorphisms were seen with primers H3 (Fig. 5)and L5 (not shown), but none of these polymor-phisms was confined to a single locality. T. infestansfrom 2 further Bolivian localities showed the samerange of polymorphisms ( not shown ) .Colonies of R.prolixus from Brazil, Colombia, and the UnitedKingdom, but of unknown early history, were alsocompared and no colony-specific characteristicswere seen with 6 primers tested.

The resolution of RAPD products by nondena-turing polyacrylamide gel electrophoresis and silver

42 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 35, no. 1

M 21 3 4 5 6 7 83 4 5 6 7 8 9 10 11 12M 2

4 5 6 7 8 9 10M 1 2 3

Fig. 5. RAPD profiles of T. infestans from 3 differentlocalities in Bolivia with primer H3. Tracks: (1-3) Tarija,(4-6) Cochabamba, (7 and 8) Sucre. Track M contains1-kb ladder marker.

tinguished R. prolíxus and R. robustus, allowing, forthe samples available to us, correct assignment tospecies; variation was seen between individuaIswithin the same species.

10M 2 3 4 5 6 7 8 91

Discussion

We have obtained consistent RAPD profiles fromliving specimens and dried specimens stored for upto 12 mo, and from alllife cycle stages from egg toadult by using rigorously standardized PCR mix-tures.

The use oflegs or leg segments as source ofDNAtemplate avoids the complication of contaminationwith host blood meal templates or with DNA fromparasites, such as T. cruzi, or from symbionts in thetriatomine gut. As expected, occasional differenceswere seen in the RAPD profiles for individuaIswithin the same species. Usually these differenceswere in intensity of common components but insome cases in the presence or absence of discreteintense staining bands (Figs. 2 and 4, R. prolixus; Fig.5, T. infestans ) .Obviously, much greater diversitymight be found if specimens representing the entiregeographical range of a given species are examined.The profiles for representatives of the 3 generaPanstrongylus, Rhodnius, and Triatoma were verydifferent with few components shared (Fig. 3).

A few species of triatomine bugs are so alikemorphologically that their independent taxonomicstatus has been questioned. We selected 3 speciespairs, R. pictipes, R. ecuadorensis; R. neglectus, R.nasutus; R. robustus, R. prolixus, to give us an indi-cation of the capacity ofRAPD profiles to assess andresolve relationships between closely related spe-cies of the same genus. R. neglectus is distinguishedfrom R. nasutus by various aspects of the color, thewid~r basis of the median process of the pygophore

Fig. 4. RAPD profiles suggest affinities between re-lated species; top panel: Rhodnius species with primer A2.Tracks: (1 and 2) R.prolixus, (3 and 4) R robustus, (5 and6) R. neglectus, (7 and 8) R. nasutus, (9 and 10) R. pictipes,(11 and 12) R. ecuadorensis; center panel, primers H3;lower panel, primer A2. Tracks: (1-4) R. prolixus, (5 and6) R. neglectus, (7 and 8) R. nasutus, (9 and 10) R. ecua-dorensis. Track M contains 1-kb ladder markers.

staining was checked with R. prolixus and R. robus-tus and primers H3, Al, L5. A greater number ofproducts was visualized than by ethidium bromidestaining of agarose gels. Multiple components dis-

.

January 1998 GARCIA ET ALo: MOLECULAR TAXONOMY OF TRIATOMINAE 43

These results imply that RAPD profiles provide amolecular basis to reassess taxonomic relationshipsin the triatomine subfamily.

A further important aspect of vector biology withimportant implications for control of Chagas diseaseis the ability of populations of the same bug speciesto move between silvatic and domestic habitats.Other authors have used RAPD for the identifica-tion of cryptic species (Hadrys and Siva-Jothy 1994,Wilkerson et al 1995) , for genome analysis of themalaria vector Anopheles gambiae (Favia et al.1994) , to identify related species of the mosquitogenus Aedes and populations of Aedes albopictus(Kambhapati et al. 1992) and to identify the sandflyspecies Lutzomyia spinicrassa Morales, Osorno-Mesa, Osorno & Hoyos, and Lutzomyia youngi Fe-liciangeli & Murillo (Adamson et al. 1993) .Notsurprisingly, RAPD analysis has shown high levels ofpolymorphism where little allozyme diversity hasbeen detected (Black et al. 1992). T. infestans isconsidered to be entirely domestic throughout mostof its distribution and this has simplified the South-em Cone Programme for elimination of T. infestans.In Cochabamba, Bolivia, however, T. infestans isfound in both silvatic and domestic habitats. Isoen-zyme characters (Dujardin et al. 1987) did not de-tect any signs of speciation between the silvatic anddomestic T. infestans populations, although morpho-metric analysis suggested separation between them(Dujardin et al. 1996a) .Nevertheless, isoenzymeanalysis has helped to distinguish between residualinfestation and reinfestation of sprayed houses (Du-jardin et al. 1996b). We used 3 geographically sep-arate Bolivian populations to show that some RAPDprimers can detect polymorphisms among popula-tions of T. infestans (Fig. 5) and thus, in principIe,might provide a means of following bug migrationand establishing source of reinfestation. Carlier etal. (1996) report that RAPD with T. infestans revealshigher intraspecific variability than isoenzymes.

Combinations of biochemical, cytogenetic, andmorphological criteria have been used to addressthe limitations of classical taxonomic methods. Itseems likely that molecular methods of genomeanalysis, of which RAPD is but one approach, willprovide a valuable means of reassessing triatominephylogeny, including relationships between the Oldand New World species and with other Reduviidaewithout destroying important morphological char-acteristics. It is likely that automated fragment anal-ysis of fluorescently labeled PCR products ofknownidentity will provide the basis for informative qual-itative comparisons. The possibility of using driedspecimens, for example from museum collections,suggests that such a study could, in part, be retro-spective. As well as the intrinsic academic interest,a clear view of triatomine taxonomy and vectorpotential will assist the worthy and achievable goalof eliminating vector-borne T. cruzi transmission inthe Americas.

~~

Fig. 6. RAPD profiles of R. prolixus and R. robustusresolved by polyacrylamide gel electrophoresis and silver

staining.

of the male genitalia in the latter species, and byisoenzyme profiles (J. P. Dujardin, ORSTOM/IBBA,La Paz, Bolivia, personal communication) .R. robus-tus shares most morphological and color characterswith R. prolixus and its taxonomic status as a distinctspeci~s has been doubted (Harry et al. 1992). Intheir monograph, Lent and Wygodzinsky (1979)regard R. robustus and R. prolixus as valid taxa basedon differences in color of the hind tibia of 4th and5th stage nymphs (light in R. robustus) and smalldifferences in the shape of the modified basal platestruts of the aedeagus of the male genitalia. Harry etal. (1992) found no isoenzyme differences betweenwhut they cQnsidered to be silvatic R. robustus anddomestic R. prolixus from Trujillo, Venezuela. Thetaxonomic status of these 2 species is of considerableepidemiological importance because R. robustus iswidely distributed in silvatic habitats where no do-mestic R. prolixus have been found. In the Amazonbasin, R. robustus is abundant in palm trees (Mileset al. 1981, 1983) and yet truly domiciliated triato-mine bug species have not been found anywhere inthe Amazon region.

Although we have not yet attempted any analysisbased on pairwise presence/ absence comparisons,the RAPD profiles for species within the genusRhodnius broadly accorded with perceived relation-ships from traditional morphological comparisons.Thus, in Fig. 4, R. prolixus and R. robustus, R. ne-glectus and R. nasutus share a number of features incommon. R. pictipes and R. ecuadorensis, whilst sim-ilar to each other, appeared to diverge from theother 4 species ofRhodnius (Figs. 4 and 6). Never-theless, in each case, with the samples examinedhere, RAPD cr-iteria could assign individual speci-mens to species. Clearly the ability to define genusand species-specific criteria must depend onbroader comparisons within and across species us-ing samples representing wider areas of distribution,for example, to exclude differences caused by ge-netic drift between populations of the same species.

44 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 35, no. 1

Acknowledgments

We thank H. Bermudez (Bolivia), c. Curtis (U.K.), andthe Instituto Nacional de Salud (Bogota, Columbia) fortheir encouragement and cooperation and the WorldHealth Organization~ European Commission, Consejo Na-cional de Investigaciones Cientificas y Tecnologicas(CONICIT, Venezuela), Fundaçao Nacional de Saude(Brazil), and the Wellcome Trust (U.K.) for financial

support.

ticidal control interventions in central Bolivia. ActaTrop. 61: 263-266.

Favia, G" G, Dimopoulos, and C, Louis, 1994, Analysis ofthe Anopheles gambiae genome using RAPD markers.Insect. MoI. Biol. 3. 149-157.

Garcia, B, A" D, M, Canale, and A. Blanco, 1995a, Geneticstructure of four species of Triatoma (Hemiptera:Reduviidae) from Argentina. J. Med. Entomol. 32: 134 -137.

Garcia, B, A" J, M, Barata, and A, Blanco. 1995b, Enzymepolymorphism among Triatoma infestans (Hemiptera:Reduviidae) colonies. J. Med. Entomol. 32: 126-133.

Gorla, D, E" J, P, Dujardin, and C. J, Schofield. 1997,Biosystematics of Old World Triatominae. Acta Trop.63: 127-140.

Hadrys, H., and M, T, Siva.Jothy, 1994, Unravelling thecomponents that underlie insect reproductive traitsusing simple molecular approach. EXS 69: 75-90.

Harry, M" I, Galindez, and M, L, Cariou, 1992, Isozymevariability and differentiation between Rhodnius pro-lixus, R. robustus and R. pictipes, vectors of Chagasdisease in Venezuela. Med. Vet. Entomol. 6: 37-43.

Haymer, D, S, 1994, Arbitrary (RAPD) primer sequencesused in insect studies. Insect Moi. Biol. 3: 191-194.

Kambhampati, S" W , C, Black IV, and K. S, Rai. 1992,Random amplified polymorphic DNA of mosquito spe-cies and populations (Diptera: Culicidae): techniques,statistical analysis, and applications. J. Med. Entomol.29: 939 -945.

Lent, H" and P, Wygodzinsky, 1979, Revision of the Tri-atominae (Hemiptera, Reduviidae) and their signifi-cance as vectors ofChagas' disease. BulI. Am. Mus. Nat.Hist. 163: 123-520.

Lopez, G" and J, Moreno, 1995, Genetic variability anddifferentiation between populations of Rhodnius pro-lixus and R. pallescens, vectors of Chagas' disease inColombia. Mem. Inst. Oswaldo Cruz 90: 353-357.

Miles, M, A" A, A, de Souza, and Povoa, M, 1981, Chagas'disease in the Amazon Basin: III. Ecotopes of ten tri-atomine bug species (Hemiptera: Reduviidae) fromthe vicinity of Belem, Para State, Brazil. J. Med. Ento-mol. 18: 266-278.

Miles, M, A" J ' R. Arias, and A, A, de Souza. 1983, Chagas'disease in the Amazon Basin: V. Periurban palms ashabitats of Rhodnius robustus and Rhodniuspictipes-triatomine vectors of Chagas' disease. Mem.Inst. Oswaldo Cruz 78: 391-398.

Pereira,J"J, P, Dujardin, R, Salvatella, and M. Tibayrenc,1996, Enzymatic variability and phylogenetic related-ness among Triatoma infestans, T. platensis, T. delponteiand T. rubrovaria Heredity 77: 47-54.

Perez, R" y, Panzera, S, Scafiezzo, M, C, Mazzella, F,Panzera, J, P, Dujardin, and E. Scvortzoof, 1992, Cy-togenetics as a tool for Triatomine species distinction(Hemiptera-Reduviidae). Mem. Inst. Oswaldo Cruz87: 353-361.

Schofield, C, J, 1985. Control of Chagas disease vectors.Br. Med. BulI. 41: 187-194.

1994. Triatominae- biology and control. Eurocommu-nica Publications, Bognor Regis, U.K.

Vidigal, T,H,D,A" E, D, Neto, O,D,S, Carvalho, and A,J,G,Simpson, 1994, Biomphalaria glabrata. Extensive ge-netic variation in Brazilian isolates revealed by randomamplified polymorphic DNA analysis. Exp. Parasitol.79: 187-194.

Wanderley, D, M, 1993, Control of Triatoma infestans inthe State of Sao Paulo. Rev. Soc. Bras. Med. Trop. 26:17-25 (suppl. 3).

References Cited

Adamson, R. E., M. L. Chance, R. D. Ward, D. Feliciangeli,and R.D.C. Maingon. 1991. Molecular approaches ap-plied to the analysis of sympatric sandHy populations inendemic areas of western Venezuela. Parassitologia 33:45-53 (suppI. 1).

Adamson, R. E., R. D. Ward, M. D. Feliciangeli, and R.Maingon. 1993. The application of random amplifiedpolymorphic DNA for sandHy species identification.Med. Vet. EntomoI. 7: 203-207.

Ballinger.Crabtree, M. E., W. C. Black IV, and B. B. Miller .1992. Use of genetic polymorphism detected by therandom polymerase chain reaction (RAPD-PCR) fordlfferentiation and identification of Aedes aegypti sub-species and populations. Am. J. Trop. Med. Hyg. 47:893-901.

Black, W. C., N. M. Duteau, G. J. Puterka, J. R. Nechols,and J. M. Pettorini. 1992. Use of the random ampli-fied polymorphic DNA-polymerase chain-reaction(RAPD-PCR) to detect DNA polymorphisms inaphids (Homoptera, Aphidae). BulI. Entomol. Res.82: 151-159. ,

Black, w. C., IV. 1993. PCR with arbitrary primers: ap-proach with care. Insect MoI. BioI. 2: 1-6.

Carlier, L., M. Munoz, and J. P. Dujardin. 1996. A RAPDProtocol for Triatominae, pp. 81-83. In C. J. Schofield,J. P. Dujardin, and J. Jurberg [ eds. ] , Proceedings of theInternational Workshop on Population Genetics andControl of Triatominae, Santo Domingo de los Colo-rados, Ecuador. INbRE, M~xico City.

Carrasco, H. J., I. A. Frame, S. A. Valente, and M. A. Miles.1996. Genetic exchange as a possible source ofgenomic diversity in sylvatic populations of Trypano-soma cruzi, Am. J. Trop. Med. Hyg. 54: 418-424.

Días, J. C. 1993. PAHO/WHO-Task force on the initia-tive of the South Cone for the elimination of vectortransm,ission and interruption of transfusion transmis-sion of Trypanosoma cruzi. Epidemiologiç surveillanceof Triatoma infestans, Rev. Soc. Bras. Med. Trop. 26:39-44 (suppI. 3).

Dorea, R. C., M. M. Povoa, M. A. Miles, A. A. de Souza, andJ. M. Barata. 1982. Enzyme electrophoresis for thestudy of Triatomidae with speciaI reference to Pan-strongylus megistus subpopulations. Rev. Bras, Biol. 42:521-526.

Dujardin, J. P., M. Tibayrenc, E. Venegas, L. Maldonado,P.Desjeux,andF.J.Ayala. 1987. Isoenzymeevidenceof lack of speciation between wild and domestic Tri-atoma infestans (Heteroptera: Reduviidae) in Bolivia.J. Med. Entomol. 24: 40-45.

Dujardin, J. P ., H. Bermudez, and C. J. Schofield 1996a.The use of morphometrics in entomologicaI surveiI-1ance of silvatic foci of Triatoma infestans in Bolivia.Acta Trop. (in press)c

Dujardin, J. P., L. Cardozo, and C. J. Schofield. 1996b.Genetic anaiysis of Triatoma infestans following insec-

GARCIA ET AL.: MOLECULAR TAXONOMY OF TRIATOMINAE 45

.

January 1998

Williams, J. C., A. R. Kubelik, K. J. Livak, J. A. Rafalski,andS. v. Tingey. 1990. DNApolymorphismsamplifiedby arbitrary primers are useful as genetic markers.Nucleic Acid Res. 18: 6531-6535.

[WHO] World Health Organization. 1991. Control ofChagas disease. W.H.O. Tech. Rep. 8l16.

Wilkerson, R. C., T. J. Parsons, T. A. Klein, T. V. Gaffigan,E. Bergo, andJ. Consolim. 1995. Diagnosis by randomamplified polymorphic DNA polymerase chain reac-tion of four cryptic species related to Anopheles (Nys-sorhynchus) albitarsis (Diptera: Culicidae) from Par-aguay, Argentina, and Brazil. J. Med. Entomol. 32: 697-704.

Received for publication 24 October 1996; accepted 19June 1997.