| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Culex restuans Theobald, Cx. salinarius Coquillet, Cx. nigripalpus Theobald, Cx. tarsalis Coquillet, and the Cx. pipiens Linnaeus complex are among the most widely distributed representatives of the Culex subgenus Culex in North America. Mosquitoes belonging to this group are difficult to identify by morphologic characters alone, especially because adults lose discriminating scales and setae during trapping and with age. For instance, the only reliable morphologic character for separating within the Cx. pipiens complex that include Cx. p. pipiens Linnaeus, Cx. p. molestus Forskål, and Cx. p. quinquefasciatus Say is the DV/D ratio in male specimens.4 This ratio is an expression of the difference in the shape and relative positions of the dorsal (D) and ventral (V) arms of the male phallosome. This character is not useful for separating adult female specimens, which represent the only life stage that acts as vectors of communicable diseases. Because many of these species and subspecies are sympatric, field-collected mosquito pools used for arboviral surveillances are frequently contaminated with more than one representative of the subgenus Culex. Several molecular techniques have been developed to identify mosquitoes of the Culex group, including isoenzyme assay5–7 and polymerase chain reaction (PCR) amplification of the internal transcript spacer (ITS) region,8,9 acetylcholinesterase (Ace2),10–12 and microsatellites.13 Although these techniques have significantly advanced our understanding of the structure of the group, cross-reactions between closely related taxa have been observed using some of the species-specific assays12 (unpublished observations). These techniques are also not readily amenable for high-throughput identification of mosquitoes from large-size samples required in area-wide arbovirus surveillance programs.
In this study, we developed the first real-time TaqMan PCR assay for the identification of Culex (Culex) mosquitoes including Cx. p. pipiens, Cx. p. quinquefasciatus, Cx. restuans, Cx. salinarius, Cx. nigripalpus, and Cx. tarsalis, all important vectors of WNV and SLV. PCR primers and TaqMan fluorogenic probes specific to each mosquito species were designed based on the alignment of the acetylcholinesterase gene (Ace2) sequences. Both individual and mixed pooled mosquitoes were accurately identified. The sensitivity and specificity of the assay were validated using specimens that have been identified using a combination of morphologic and molecular data.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Laboratory colonies were maintained in 0.5-m3 cages at 28 ± 2°C, 75 ± 10% relative humidity, and 16:8-hour light:dark photo-period. Larvae were fed with a suspension of fish food (VitaPro Plus; M. Reeds Enterprises, Sutter Creek, CA). Adults had continuous access to a 10% honey solution and were fed on quail (Colina virginianus) 3 days after emergence. The animal handling was done according to the University of Illinois Institutional Animal Care and Use Committee guidelines.
|
). Both immature and adult developmental stages were examined. Larvae were collected in mosquito breeding sites using the standard dipping procedure. Egg rafts were obtained using grass infusion–baited oviposition traps. Adult mosquitoes were collected using gravid traps baited with grass infusions and CDC light traps baited with dry ice. On arrival in the laboratory, eggs were allowed to hatch, and larvae were identified in the fourth larval stage. Adult mosquitoes were frozen immediately at –70°C until processed for identification of species. Mosquito identification. The morphologic identification was performed first using standard identification keys and recently described characters14,15 by one of us (R.L.L.). We note that, of the species examined, Cx. tarsalis is easily differentiated from the other species using morphologic traits. Males of the Cx. pipiens complex were identified using measurements of the DV/D ratio of male phallosomes4 following the method of Barr.16 Individuals with a DV/D ratio < 0.2 were classified as Cx. p. pipiens, and those with DV/D > 0.4 were considered Cx. p. quinquefasciatus. DV/D values between 0.2 and 0.4 were considered hybrids. Before TaqMan identification, the last three segments of the male abdomen were used for DV/D measurements, and the remaining part of the mosquito was used for molecular identification according to Crabtree and others8 and Aspen and Savage.10
Nucleic acid isolation. Genomic DNA was isolated from single and pooled mosquitoes using the DNeasy kit (QIAGEN, Valencia, CA) following the manufacturer’s recommendations with slight modification. Before the addition of proteinase K solution (provided in the kit), samples were placed in tubes containing 40 µL of sterile Tris-EDTA (TE) buffer (pH = 8; IDT, Coralville, IA) and homogenized using a sterile pestle mounted in a battery-operated mortar (Fisher Scientific, Hampton, NH). After homogenization, 180 µL of ATL buffer and 20 µL of proteinase K (provided by the supplier) were added to the mosquito samples and mixed by vortexing. The remaining of the DNA isolation process was performed in accordance with instructions provided in the kit. DNA was eluted in a final volume of 200 µL and stored at – 20°C until used in PCR.
Amplification and sequencing of the Ace2 gene. Before designing the TaqMan real-time PCR primers and probes, we amplified and sequenced the Ace2 gene of the Culex mosquitoes used as controls in the study including Cx. p. pipiens, Cx. p. quinquefasciatus, Cx. restuans, Cx. salinarius, and Cx. tarsalis. For this purpose, a new universal primer pair CxAce2F (5'-TAC CRACGA ARA CCC GTT TGC-3') and CxAce2R (5'-TAG ATC CAG ACC AGC ATC GCG-3') was designed based on the alignment of partial Ace2 sequences of Cx. p. pipiens (AY196910), Cx. p. quinquefasciatus (AY196911), Cx. restuans (AY196912), and Cx. salinarius (AY196913). Primers and probes specific to Cx. nigripalpus were designed directly from sequences of the Ace2 from GenBank (AY196914). The universal 18S rRNA primers (Invitrogen, Carlsbad, CA) were used to test the DNA template quality.
PCR reactions were performed in a total volume of 25 µL, including 2.5 mmol/L MgCl2, 0.05 mmol/L dNTPs, 200 nmol/L each primer, 1 U Platinum Taq DNA polymerase (Invitrogen), 1x PCR buffer, and 2 µL DNA. Reactions were initiated at 95°C for 4 minutes, followed by 40 cycles of 95°C for 30 seconds, 60°C for 30 seconds, 72°C for 1 minute, and a final elongation step at 72°C for 10 minutes, with a final hold at 4°C. The PCR product was run on a TBE agarose gel and viewed under UV illumination after staining with ethidium bromide.
Direct sequencing of PCR products was performed on a Beckmann 8000 CEQ sequencing machine (Beckmann Coulter, Fullerton, CA) using the dye termination reaction. Sequences were edited using the SeqMan program (DNASTAR, Madison, WI) and aligned using ClustalW.17
Real-time PCR primer and probe design.
TaqMan probes and primers were designed based on alignment of the Ace2 sequences from GenBank and sequences obtained in our laboratory. The list of the primers and probes is found in Table 2. Primers and fluorogenic probes were synthesized by IDT.
|
. Validation of the test. The sensitivity of the TaqMan primers and probes was assessed using 10-fold serial dilutions of 10 ng/µL initial DNA concentration of Cx. p. pipiens, Cx. p. quinquefasciatus, Cx. restuans, Cx. salinarius, Cx. nigripalpus, and Cx. tarsalis. The initial amount of DNA was determined on a NanoDrop ND-1000 Spectrophotometer (Nanodrop Technologies, Wilmington, DE). Dilutions of the individual extracts contained up to 1 x 10–5 ng/µL of DNA as starting material. The reaction was run in a volume of 25 µL of master mix. Each concentration was replicated three times with each species. The detection limit was defined as the lowest concentration of DNA that displayed a detectable fluorescence above the cycle threshold (Ct). The data were analyzed by one-way analysis of variance (ANOVA) and linear regression of Ct and log-dose using Statview 5.0 (SAS Institute, Cary, NC).
Blinded test.
To further assess the specificity and the accuracy of the test, a set of mosquito samples was selected and assigned an identification code so that the sample identity was unknown when testing the primers (Table 3). The samples consisted of both single and mixed mosquitoes including Culex mosquitoes and distantly related species (i.e., Aedes aegypti and Anopheles gambiae, Ae. vexans, Ae. trivittatus). Both immature and adult life stages were included in the test. For the second blinded test, field-collected adults were used and identified by morphologic characteristics. The only Cx. restuans included in the sample had the characteristic scutal spots, whereas the Cx. pipiens included in the test lacked scutal spots and had typical abdominal banding for this species. In other tests, identification of adult Cx. pipiens and Cx. restuans was largely based on larval characters for both larval and adult specimens. The adults were from cages containing species-specific specimens identified in the larval stage from individual egg rafts.
|
. Great care was taken to avoid sample cross-contamination. Given that mosquitoes of the Cx. pipiens complex are morphologically closed to Cx. restuans, particularly during adult stage, we first identified specimens of the latter by real-time PCR before mixing them in pools with the former. Cx. pipiens complex mosquitoes were from the laboratory colony, and adult Cx. restuans were from egg rafts collected in the field and identified as larvae (this mosquito is difficult to colonize). For molecular identification of Cx. restuans used in the pools, the last two segments of the abdomen of individual mosquitoes was removed and placed in a tube that was labeled identically with a second tube containing the remainder of the mosquito body. DNA was extracted from the small segment of the mosquito and molecular identification by real-time PCR performed as described above. Only specimens identified unambiguously as Cx. restuans were used in test pools.
|
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), but not from Ae. aegypti and An. gambiae. The quality of extracted DNA was confirmed by amplification of a 488-bp fragment from all Culex mosquitoes using the 18S rRNA primers. PCR product was sequenced in both directions. Blastn search (http://www.ncbi.nlm.nih.gov/blast/) revealed that sequences obtained from amplification DNA from Cx. p. pipiens, Cx. p. quinquefasciatus, Cx. restuans, and Cx. salinarius (accession numbers AY196910, AY196911, AY196912, and AY196913, respectively) were identical to published sequences in GenBank. This is consistent with data obtained using morphologic and PCR identification. The Ace2 gene of Cx. tarsalis has not been previously published and was deposited in GenBank under accession number EF089567.
TaqMan real-time PCR.
The TaqMan probes and primers designed in this study accurately identified all the reference mosquitoes (Table 3). Amplification with the universal Culex primers CxAce2F/CxAce2R using species-specific probes generated irregular amplification plots. These primers were thus not used in real-time PCR. Using the Culex-specific primers and probes, no fluorescence was detected in wells where Ae. (Oc.) trivittatus, Ae. aegypti, and An. gambiae were used or in the negative water blank controls. As expected, all the Culex specimens were successfully detected using the 18S rRNA primers and probes.
Determination of the detection limit of the assay.
Tenfold serial dilutions of DNA from Cx. p. pipiens, Cx. p. quinquefasciatus, Cx. restuans, Cx. salinarius, Cx. nigripalpus, and Cx. tarsalis were made using an initial concentration of 10 ng/µL of DNA. Figure 1A and B displays the amplification and regression plots of the sensitivity test of each mosquito species. The slopes of all the log dose-Ct curves were ~3.0; that is, for every three Cts there was about a log or 10-fold change in concentration. Based on the regressions, the theoretical minimum amount of species-specific DNA that could be detected above the threshold by the 40th Ct would be 5.8 x 10–3 ng for Cx. pipiens, 5.3 x 10–4 ng for Cx. quinquefasciatus, 1.9 x 10–5 ng for Cx. restuans, 3.2 x 10–3 ng for Cx. salinarius, 2.6 x 10–6 for Cx. nigripalpus, and 8.7 x 10–4 ng for Cx. tarsalis.
|
|
). Ten of the 13 single mosquitoes were accurately identified. Among the three specimens that were not correctly identified, two were Cx. p. pipiens mosquitoes from Champaign-Urbana that amplified both Cx. p. pipiens–and Cx. p. quinquefasciatus–specific primers. Previous studies suggested that hybrids exist in our area,18 so we believe that our assay accurately identified both of these specimens as hybrids. The third specimen, a Cx. restuans identified as an adult only, did not yield any reaction with any of the primer set. The reason for this is not known. Eleven of the 12 mixed pools of mosquitoes were accurately identified. The only misidentified pool was a mixture of one Cx. p. pipiens and one Cx. restuans, both adults. The assay identified these two mosquitoes as one single species: Cx. pipiens. The identity of these mosquitoes was further tested using PCR according to Crabtree and others8 and Aspen and Savage.10 The results were consistent with the real-time TaqMan results. Thus, even with extreme care taken in providing only unquestionable specimens, 1 of 10 field-collected adult Cx. restuans was misidentified morphologically. This further highlights the need for an easy method of verifying the species. All the positive controls were properly amplified using the corresponding primers and probes. No amplification was generated from Aedes or Anopheles or from the water blank control.
Identification of mosquitoes in large mixed pools.
Unexpectedly, our first attempt to amplify large-size mosquito pools (N > 40), using the reaction conditions for single and small-size pools, failed, although all the controls were successfully amplified. Attempts to optimize the reaction conditions revealed that a critical factor for assaying large pools was the amount of nucleic acid used as template in the reaction. A 4-fold decrease of template concentration from an initial 240 to 60 ng/µL substantially improved the sensitivity and specificity of the assay. As shown in Table 4, the assay successfully and accurately detected Culex mosquitoes in all the mixed pools that have been tested. For example, one single Cx. p. pipiens was detected in a pool of 49 Cx. p. quinquefasciatus, and one Cx. p. quinquefasciatus was identified in a pool of 49 Cx. p. pipiens. The assay was also able to detect one single Cx. p. pipiens in a mixture of 74 and 99 Cx. p. quinquefasciatus, respectively. These results were confirmed by the detection of one Cx. p. pipiens in a pool of 49 Ae. aegypti. On the other hand, the amplification of a pool of 40 Ae. aegypti did not yield any product.
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The real-time TaqMan assay herein developed readily identified the major Culex species typically involved in WNV and SLEV transmission in North America. Cx. restuans, Cx. salinarius, Cx. nigripalpus, and Cx. tarsalis were unambiguously identified. Culex p. pipiens and Cx. p. quinquefasciatus were accurately identified in samples collected from Saint Augustine (Florida), Saint Tammany (Louisiana), Coachella Valley (California), and New Haven (Connecticut) by real-time PCR. Subspecies molecular identification of specimens collected from these areas by both conventional PCR and TaqMan real-time assay was congruent with measurements of the male DV/D ratios from specimens from the same cohort. Somewhat unexpected was a high proportion of hybrids in our blind studies associated with Cx. pipiens collected from Champaign Co., IL. About three of seven Cx. pipiens specimens were hybrids based on our assay method. Although in one case (blinded test vial 21), the specimen from the strain Cx. pipiens CP-CU was correctly identified in a pool of four Cx. pipiens and one Cx. restuans adults, in three cases, mosquitoes from the same areas amplified both Cx. p. pipiens and Cx. p. quinquefasciatus probes. Two of these samples were single mosquitoes, one larva (blinded test vial 9), and one adult (blinded test vial 13). The third sample was a pool of one Cx. p. pipiens and four Cx. restuans adults (blinded test vial 22). This suggests that the Cx. pipiens specimens were intermediate forms between Cx. p. pipiens and Cx. p. quinquefasciatus. It should be noted that the adult mosquitoes used in this particular case were obtained from a laboratory colony that was established using egg rafts collected from various locations in Champaign Co. Indeed, the occurrence of intermediate forms between Cx. p. quinquefasciatus and Cx. p. pipiens has been reported in Southern Illinois east of Saint Louis (38.5° N) some years ago.16 Recently, Lampman and others18 reported Cx. pipiens hybrids in ~4% of Cx. pipiens mosquitoes collected in Champaign Co. Because both Cx. p. pipiens and Cx. p. quinquefasciatus probes are specific, amplification with both probes seems unique to intermediate forms. This is in agreement with previously reported data, in which Cx. pipiens hybrids DNA was amplified using primers specific to both parents.12 To date, except the full microsatellite analysis,13 there is no specific identification technique for hybrids of Cx.p. pipiens and Cx. p. quinquefasciatus. Recently, a method for discriminating between hybrids from parental specimens was reported by Bahnck and Fonseca.20 However, the authors failed to reliably identify hybrids beyond the F1 generation, probably because of recombination within the diagnostic locus. This can obscure accurate identification of the mosquitoes, particularly in regions of strong introgression.7 Future studies should focus on the developing molecular markers that will allow reliable identification of hybrid forms in zone of sympatry and examine how extensive introgression affect on the genome of the hybrids. This is important because hybrid forms of the Cx. pipiens complex have been hypothesized to play the role of bridge vectors of WNV in North America.21 Examination of the male phallosomes from specimens collected from July to September in Champaign-Urbana suggests that morphologic variation increases in August and September, when introgression between Cx. pipiens complex members might be expected to occur (unpublished data).
The discrepancies between morphologic identification and molecular identification of pool 17 (N = 2) are noteworthy. Based on morphologic traits, these mosquitoes were identified as one Cx. pipiens and one Cx. restuans. However, conventional PCR and real-time Taqman PCR identified both mosquitoes as Cx. restuans. This observation underscores the challenge faced by mosquito researchers during identification of Cx. pipiens and Cx. restuans. Indeed, because of the inability to accurately identify the Cx. pipiens complex and Cx. restuans as adults, these two are usually pooled and labeled as Cx. pipiens/restuans.3 A review of the literature suggests that these two species exhibit considerable morphologic and biologic differences.18 In a recent work, Darbro and Harrington22 showed that as many as 23% of Cx. restuans mosquitoes caught as adults lack the characteristic scutal spot used to differentiate them from Cx. pipiens.14 This suggests that the usefulness of morphologic characters to differential Cx. pipiens and Cx. restuans is limited.
Our ability to detect one single Culex mosquito species in pools of 45, 49, 74, and even 99 other species shows that the assay used in this study is highly sensitive. However, the amplification of DNA from large size pools of mosquitoes required optimization of the procedure, mostly by reducing the total amount of nucleic acid used as template in the reaction. Other components of the reaction mixture such as buffer concentration, MgCl2, and primers concentration should also be adjusted. In a previous study, McAvin and others23 were able to detect and identify one single Ae. aegypti species in a pool of 99 non-Aedes mosquitoes by real-time TaqMan PCR. This suggests that vector identification by real-time PCR is a powerful and highly sensitive method that will greatly advance our understanding of the role played by arthropods in the transmission of arboviruses and other infectious agents. Given the differences in the vector ability of these mosquitoes, the contamination of pools used in arbovirus surveillance with representatives of the Culex subgenus Culex can obscure the prediction of transmission dynamics of the diseases that they transmit and complicate disease control strategies. Thus, the accurate identification of mosquito vectors can help to determine the best pest management strategy to adopt and decrease the effect of pesticide on non-target organisms.
Our study represents an important step toward the development of a rapid, sensitive, and high-throughput method for expounding the species-specific contribution of vectors to various phases of arbovirus transmission. We emphasize that this study is not quantitative but provides the basis for protocols that could lead to the development of a multiplexed, quantitative assay for the detection and identification of disease vector and the agent they carry. The assay as such can be readily used to detect and identify Culex mosquitoes both individually and in mixed pools.
Received October 31, 2006. Accepted for publication March 25, 2007.
Acknowledgments: The authors thank the team of the Champaign-Urbana Encephalitis Prevention Program for help with mosquito rearing and identification. The authors thank Dr. J. Anderson (The Connecticut Agricultural Experiment Station, New Haven, CT), Dr. R. Xue and Whitney Qualls (Anastasia Mosquito Control District, FL), and H. Lothrop (Coachella Valley Mosquito and Vector Control District, CA) and C. Palmisano (Saint Tammany Parish Mosquito Abatement District, LA) for providing the mosquito strains.
Financial support: This study was supported by a grant from the Illinois Waste Tire Fund and USDA/CREES Grant 2005-34523-15639 to R. J. Novak.
* Address correspondence to Yibayiri O. Sanogo, Illinois Natural History Survey, 1816 South Oak Street, Champaign, IL 61820. E-mail: sanogo{at}uiuc.edu 
Authors’ address: Yibayiri O. Sanogo, Chang-Hyun Kim, Richard Lampman, and Robert J. Novak, Illinois Natural History Survey, 1816 South Oak Street, Champaign, IL 61820.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. A. Hill, J. B. Davis, G. Hapgood, P. I. Whelan, G. A. Smith, S. A. Ritchie, R. D. Cooper, and A. F. van den Hurk Rapid Identification of Aedes albopictus, Aedes scutellaris, and Aedes aegypti Life Stages Using Real-time Polymerase Chain Reaction Assays Am J Trop Med Hyg, December 1, 2008; 79(6): 866 - 875. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
