Journal of the American Mosquito Control Association, l9(2):ll5-120,20O3Copyright @ 2003 by the American Mosquito Control Association, Inc.

POLYMERASE CHAIN REACTION ASSAY IDENTIFIESCULEX NIGRIPALPUS: PART OF AN ASSAY FOR MOLECULAR

IDENTIFICATION OF THE COMMON CULEX (CULEN MOSQUITOESOF THE EASTERN UNITED STATES

STEPHEN ASPEN, MARY B. CRABTREE AND HARRY M. SAVAGE

Division of Vector-Borne Infectious Disease, Centers for Disease Control and Prevention, Fort Collins, CO 80522

ABSTRACT, Nucleotide sequence information on intemal transcribed spacer (ITS) 1 and ITS 2 regions ofthe nuclear ribosomal DNA multigene family was used to develop a polymerase chain reaction assay thatidentifies Culer nigripeilpas Theobald. The assay uses species-specific forward and reverse primers for C.x.nigripalpus and can be used along with previously described primers to distinguish among 4 common taxa ofCulex (Culex) of the eastern USA with a single thermal cycler program. The assay distinguishes among the 4taxa Cx. nigripalpus, Cx. restuans Theobald, Cx. salinarius Coquillett, and members of the Cx. pipiens Linnaeuscomplex. This assay may be used to verify the morphological identification of individual specimens of Culexor to confirm the species composition of mosquito pools.

KEY WORDS Culex (Culex), Culex nigripalpzs, internal transcribed spacers, ribosomal DNA, molecularspecies identification

INTRODUCTION

In North America, mosquitoes of the subgenusCulex (Culex) are the primary enzootic vectors ofWest Nile virus (Nasci et al. 2001b) and the pri-mary enzootic and epidemic vectors of St. Louisencephalitis virus (Mitchell et al. 1980). Adult fe-males of these mosquitoes are frequently collectedin gravid traps (Reiter 1983), Cor-baited Centersfor Disease Control light traps, or with aspiratorsas part of arbovirus surveillance programs (Nasciet al. 2001a), blood host studies (Apperson et al.2OO2), or other mosquito investigations. Morpho-logical characters are used to identify specimensand sort field-collected mosquitoes into speciespools. Mosquito pools can be tested for virus aspart of surveillance programs and to assess the vec-tor status of particular species. However, adult fe-males of some Culex common in the eastern USA(Cx. pipiens Linnaeus, Cx. quinquefosciatus Say,Cx. nigripalpas Theobald, Cx. restuans Theobald,and Cx. salinarius Coquillett) are difficult to distin-guish morphologically if specimens are worn ordamaged during the trapping process. The correctidentification of mosquito species is essential in de-termining the roles of vector species and for thedevelopment of effective arbovirus control and pre-vention strategies.

Polymerase chain reaction (PCR)-based assaysthat use species-specific primers based on differ-ences in ribosomal DNA (rDNA) nucleotide se-quence are useful in distinguishing among closelyrelated anopheline (Paskewitz and Collins 1990,Porter and Collins 1991, Scott et al. 1993, Townsonand Onapa 1994, Cornel et al. 1996) and culicine(Crabtree et al. 1995, Toma et al. 2000) mosquitotaxa. Ribosomal DNA makes an ideal template forspecies-diagnostic PCR for at least 2 important rea-sons. First, the high copy number of rDNA arrays

within the adult mosquito genome (in Aedes aegyp-ri see Park and Fallon [l990l) means that a smallfraction of extracted DNA supplies a sufficient tem-plate for PCR. Second, rDNA contains highly con-served coding regions, ideal as annealing sites forsequencing primers, separated by less conservedspacers, which are the source of interspecific se-quence variation.

Crabtree et al. (1995) describe a PCR assay forthe identiflcation of Cx. restuans, Cx. salinarius,and members of the C-r. pipiens complex. This as-say makes use of interspecific nucleotide sequencedifferences in the noncoding internal transcribedspacers (ITS) I and ITS 2 of rDNA. Different PCRprimers anneal to taxon-specific ITS sequences,producing a uniquely sized amplicon in the pres-ence of the appropriate template DNA. Subtractivehybridization was used to design primers thatwould further distinguish among the sibling taxaCx. pipiens and Cx. quinquefasciatus (Crabtree etal. 1991): however, only Cx. pipiens-specific prim-ers could be designed. Recently, a PCR assay thatidentifles Cx. pipiens, Cx. quinquefasciatus, andtheir F, hybrids based on nucleotide sequence dif-ferences in the acetylcholinesterase gene Ace.2 hasbeen developed (Aspen and Savage, personal com-munication).

The appearance of West Nile virus (WN) in Flor-ida and the isolation of WN from Cx. nigripalpusduring the sunnner of 2001 (CDC 2OO2) imposedthe need to add Cx. nigripalpus to the list of taxathat can be distinguished by the assay of Crabtreeet al. (1995). Within the eastern USA, Cx. nigri-palpus is found in the southernmost latitudes,where it may be collected along with Cx. restuans,Cx. quinquefasciatus, and the morphologically srm-ilar species Cx. salinarius. For this study, the ITSregions of Cx. nigripalpus were sequenced andcompared with the ITS sequences from 7 other taxa

1 1 5

1 1 6 JounNar- oF THE AMERTcIN Mosquno CoNrnol AssocrATroN VoL. 19 , No.2

Taxa

lations of Culex used in this study.

Geographic origin Source

Culex pipiens complex

Cx. pipiens

Cx. quinquejbsciatus

Cx. nigripalpus

Cx. restuans

Cx. salinarius

Cx. tarsalis

Alleghany County, NCLos Angeles County, CAFt. Collins, CO

Queens, NYMonroe, LAJefferson County, FLJefferson County, FLVero Beach, FLGainesville, FLBurlington, VTAlleghany County, NCMonroe, LAJefferson County, FLLos Angeles County, CA

B. HarrisonM. MaddenH. SavageC. AppersonH. SavageM. GodseyM. GodseyR. RutledgeJ. ReinertH. SavageB. HarrisonH. SavageM. GodseyM. Madden

of Culex (Culex). Based upon the results of thiscomparison, species-specific PCR primers were de-signed for the identification of Cx. nigripalpus.

MATERIALS AND METHODS

Mosquitoes: The geographic origin and sourceof each mosquito population used for this study aregiven in Table 1.

Mosquito genomic DNA prepararlon.' Mosquitogenomic DNA was extracted by using the QiagenDNeasy Tissue Kit (Qiagen Inc., St. Clarita, CA)with the following modifications. Individual orpooled mosquitoes were ground in BA-l medium(lX M199-H, 0.05 M Tris, pH 7.6, l%o bovine se-rum albumin, 0.35 g/liter of sodium bicarbonate,100 units/ml of penicillin, 100 pg/ml of strepto-mycin, 1 pg/ml of fungizone). The DNA was elutedtwice in 200 pl of warm (40"C) nuclease-free water(Amresco, Solon, OH) for a total eluate volume of400 pl. The DNA concentration was determined byusing the ultraviolet absorbance at 260-nm wave-length on a spectrophotometer.

Cloning and sequencing.' A region of Cx. nigri-palpus rDNA (including part of the 18S gene, theITS 1, the 5.8S gene, the ITS 2, and part of the28S gene) was amplified as described by Crabtreeet al. (1995). The PCR product was purified by us-ing the Qiaquick PCR Purification Kit (Qiagen) andligated into the plasmid pGEM-T Easy (Stratagene,La Jolla. CA) before transformins the Escherichia

coli strain XL-l Blue (Stratagene) by heat shock.Recombinant clones were selected on agar plateswith tetracycline and ampicillin. Clones were se-quenced with 3 different primers in each direction(the 2 PCR primers, 2 plasmid primers, and 2 prim-ers complementary to the 5.8S gene) in an ABIPrism 377 sequencer (Applied Biosystems, FosterCity, CA), by using the BigDye Terminator version3 reaction mix (Applied Biosystems).

Species-specffic PCR primer design: The rDNAnucleotide sequences of Cx. nigripalpus from theJefferson County, FL, and Vero Beach, FL, speci-mens were aligned with rDNA sequences from 49clones from 7 other taxa of Culexby using PILEUPin the Wisconsin Package, ver. 7O.2 (Devereux etal. 1984). Other taxa included in the alignmentwere Cx. pipiens, Cx. quinquefasciatus, Cx. pi-piens-Cx, quinquefasciarus hybrids, Cx. restuans,Cx, salinarius, Cx. tarsalis Coq. and Cx. erythro-thorax Dyar (Genbank acquisition numbersu22114-U22144, U330r8, U33022-U33024,U33030-U33A3D. A pair of PCR primers comple-mentary to Cx. nigripalpus-specific sequences inITS I (primer N90l) and ITS 2 (primer NR1080)was designed by using the PrimerSelect module ofLasergene version 5.03 (DNAStar, Inc., Madison,WI). The nucleotide sequences of these primers areincluded in Table 2.

Species-diagnostic PCR: The species-diagnosticPCR described by Crabtree et al. (1995) was mod-ified as follows. Separate reaction mixtures, con-

Table 2. Polymerase chain reaction primers, complementary to Culex rDNA, used for molecular identification.

Sequence (5' to 3') SpecificityAmplicon

Sense Pairs with size (bpr)Primer

CPI 6 GCGGGTACCATGCTTAAATTTAGGGGGTAPQIO CCTATGTCCGCGTATACTAR6 CCAAACACCGGTACCCAAS2O TGAGAATACATACCACTGCTN9OI ATACCCATGCGAAAGCATACNRIO8O GTACCGCGACCACACGACTT

Consensus ReverseCulex pipiens complex ForwardCx. restuans ForwardCx. salinarius ForwardCx. nigripalpus ForwardCx. nigripalpus Reverse

See belowcPl6cPl6CPI6NRl080N90l

698506175404404

rbp, base pairs

JUNE 2003 PCR IopNrmcATroN oF CuzEX MoseurroEs l l 9

698-bp amplicon resulting from a reaction contain-ing PQIO and CPl6 is diagnostic of members ofthe Cx. pipiens complex. When DNA from a mos-quito pool containing all 5 taxa was tested (Fig. 2),each of the diagnostic amplicons was observed.When DNA from each of the Cx. nigripalpus pop-ulations was tested with each of the non-Cx. nigri-palpus-specific primers, no amplicon was observed(not shown), indicating that no cross-reactions oc-cur that could lead to false-positive identification.

DISCUSSION

Accurate identification of field-collected mosqui-toes is an important step in arbovirus surveillance,vector incrimination, blood host preference inves-tigations and other mosquito studies. However,many species of Culex (Culex) are difficult ro dis-tinguish from one another by using morphologicalcharacters. Difficulty with the identification processcan lead to some mosquito pools being mislabeledor some pools containing multiple species and la-befed as Culex spp. The species composition ofmorphologically identified mosquito pools or indi-viduals can be verified by using a PCR assay withspecies-speci f i c primers.

We have designed a PCR primer pair that accu-rately identifies Cx. nigripalpus. ^fhese primers dis-tinguish Cx. nigripalpzs from all specimens of Cu-lex tested, including the morphologically similarspecies Cx. salinarius, and the morphologically dis-tinct species Cx. tarsalis, a species which sharesmuch ITS nucleotide sequence homology with C"r.nigripalpus. Species identification can be deter-mined by the PCR assay even after specimens havebeen damaged or ground. A single specimen of Cx.nigripalpus in a pool of 40 (containing 39 otherspecimens of Culex) mosquitoes can be identified.This sensitivity allows arbovirus investigators tosort Culex mosquitoes into pools as large as 40specimens that can be ground and tested for virusbefore molecular identification of the mosquito spe-cies is undertaken.

The primers for Cx. nigripalpus, in conjunctionwith a modified protocol based on the work ofCrabtree et al. (1995) allow the identification of Cx.nigripalpus, Cx. restuans, Cx. salinariu^r, and mem-bers of the Cx. pipiens complex with a single ther-mal cycler program. This provides a means of ver-ifying the morphological identification of Culexcommonly collected in the eastern USA, and thepresence of Culex species in virus-positive mos-quito pools. The modified protocol described hereseparates each primer pair into a different reactiontube and avoids a loss of sensitivity caused byprimer competition that can occur in a mixed prim-er or cocktail assay. In the future, it would be ben-eficial to develop a quantitative pCR assay thatcould determine the number of specimens of eachspecies in a pool.

This assay does not distinguish among the mem-

bers of the Cx. pipiens complex: Cx. pipiens, Cx.quinquefasciatus, and their hybrids. Nucleotide se-quence comparison among members of the Cx. pi-piens complex reveals insufficient fixed differencesin the ITS I and ITS 2 regions to allow PCR prim-ers to be designed that will distinguish among thesemosquitoes. Recently, a PCR assay that identifiesCx. pipiens, Cx. quinquefasciatus, and their F, hy-brids based on nucleotide sequence differences inthe acetylcholinesterase gene Ace.2 has been de-veloped (Aspen and Savage, personal communica-tion). Just as this research was spawned by the dis-covery of WN in Florida, the continuing spread ofWN into different ecosystems may impose the needfor primers that identify additional species. Poly-merase chain reaction primers specific to othermosquito species likely can be designed by usingtechniques described here.

ACKNOWLEDGMENTS

We thank C. Apperson, North Carolina StateUniversity, Raleigh, NC; M. Godsey, Centers forDisease Control and Prevention, Ft. Collins, CO;B. Harrison, North Carolina Department of Envi-ronment and Natural Resources, Winston-Salem,NC; M. Madden, Greater Los Angeles County Vec-tor Control District, Sante Fe Springs, CA; J. Rein-ert, Center for Medical, Agricultural and VeterinaryEntomology, Gainesville, FL; and R. Rutledge,Florida Medical Entomology Lab, Vero Beach, FLfor providing mosquito specimens.

REFERENCES CITED

Apperson CS, Harrison BA, Unnasch TR, Hassan HK,Irby WS, Savage HM, Aspen SE, Watson DW, RuedaLM, Engber BR, Nasci RS. 2002. Host-feeding habitsof Culex and other mosquitoes (Diptera: Culicidae) inthe borough of Queens in New York City, with char-acters and techniques fbr identification of Culex mos-quitoes. J Med Entomol 39:777-785.

CDC [Centers for Disease Control]. 2002. West Nile virusactivity-United Srates, 2OOl. Morb Mortal Wkly Repsr(23):497-5O1.

Cornel AJ, Porter CH, Collins FH. 1996. Polymerasechain reaction species diagnostic assay for Anophelesquadrimaculatu,s cryptic species (Diptera: Culicidae)based on ribosomal DNA ITS2 sequences. J Med En-tomol 33:lO9-116.

Crabtree MB, Savage HM, Miller BR. 1995. Developmentof a species-diagnostic polymerase chain reaction assayfor the identification of Culex vectors of St. Louis en-cephalitis virus based on interspecies sequence variationin ribosomal DNA spacers. Am J Trop Med Hyg 53:,I O5-1 09.

Crabtree MB, Savage HM, Miller BR. 1997. Developmentof a polymerase chain reaction assay for differentiationbetween Culex pipiens pipiens and, Cx. pipiens quin-quefosciatus (Diptera: Culicidae) in North Americabased on genomic differences identified by subtractivehybridization. J Med Entomol 34:532-53i.

Devereux J, Haeberli P, Smithies O. 1984. A comorehen-

120 JounNlr- op rne AvemcaN Moseurro Cournol AssocrerroN VoL. 19. No. 2

sive set of sequence analysis programs for the VAX.Nucleic Acids Res 12:387-395.

Mitchell CJ, Francy DB, Monath TP 1980. Arthropodvectors. In: Monath TB ed. Sl. Louis encephaljtjs Wash-ington, DC: American Public Health Association. p313-379.

Nasci RS, Savage HM, White DJ, Miller JR, Cropp BC,Godsey MS, Kerst AJ, Bennett P, Gottfried K, LanciottiRS. 2001a. West Nile virus in overwintering Calexmosquitoes, New York City, 2000. Emerg Infect Dis 7:742144.

Nasci RS, White DJ, Stirling H, Oliver JA, Daniels TJ,Falco RC, Campbell S, Crans WJ, Savage HM, Lan-ciotti RS, Moore CG, Godsey MS, Gottfried KL, Mitch-ell CJ. 200lb. West Nile virus isolates fiom mosquitoesin New York and New Jersey, 1999. Emerg Infect Dis7:626-63O.

Park YJ, Fallon AM. 1990. Mosquito ribosomal RNAgenes: characterization of gene structure and evidencefor changes in copy number during development. InsectBiochem 20: l - l l .

Paskewitz SM, Collins FH. 1990. Use of the polymerasechain reaction to identify mosquito species of the

Anopheles gambiae complex. Med Vet Entomol 4:367-3 t 3 -

Porter CH, Collins FH. 1991. Species-diagnostic differ-ences in a ribosomal DNA internal transcribed soacerfrom the sibling species Anopheles freeborni andAnopheles hermsi (Diptera: Culicidae). Am J Trop MedHys 45:271-279.

Reiter P 1983. A portable, battery-powered trap for col-lecting gravid Czlex mosquitoes. Mosq News 43:496-498.

Scott JA, Brogdon WG, Collins FH. 1993. Identificationof single specimens of the Anopheles gambiae complexby the polymerase chain reaction. Am J Trop Med Hyg49:52O-529.

Toma I Miyagi I, Crabtree MB, Miller BR. 2000. Iden-tification of Culex vishnui sttbgroup (Diptera: Culici-dae) mosquitoes from the Ryukyu Archipelago, Japan:development of a species-diagnostic polymerase chainreaction assay based on sequence variation in ribosomalDNA spacers. J Med Entomol 37:554-558.

Townson H, Onapa AW. 1994. Identification by IDNA-PCR of Anopheles bwambae, a geothermal spring spe-cies of the An. gambiae complex. Insect Mol Biol 3:279-282.