Using DNA Sequences of the COI Barcode Region

GENETICS Molecular Identiﬁcation of Ceratitis capitata (Diptera: Tephritidae) Using DNA Sequences of the COI Barcode Region

1,2 3 4 3 N. B. BARR, M. S. ISLAM, M. DE MEYER, AND B. A. MCPHERON

Ann. Entomol. Soc. Am. 105(2): 339Ð350 (2012); DOI: http://dx.doi.org/10.1603/AN11100 ABSTRACT The utility of the cytochrome oxidase I gene barcode region for diagnosis of the Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 Mediterranean fruit ßy, Ceratitis capitata (Weidemann), is evaluated using African fruit ßy collections. The method fails to discern C. capitata from its close relative Ceratitis caetrata Munro, based on genetic distances, parsimony networks, or nucleotide diagnostic characters observed in the DNA barcode sequences. When treated as a single taxon, it is possible to discern the C. capitata ϩ C. caetrata lineage from other Ceratitis species. Levels of intraspeciÞc diversity vary within the genus Ceratitis and multiple copies of the mitochondrial gene are reported for Ceratitis cosyra (Walker). The DNA barcoding method based on genetic distance is compared with a molecular identiÞcation method using restriction fragment length polymorphism. The DNA barcode and restriction fragment-length polymorphism methods provide similar identiÞcation results, but the DNA sequence information is more suitable for quantitative analysis of the information.

KEY WORDS DNA barcode, restriction fragment-length polymorphism, diagnostic, Tephritidae

The fruit ßy genus Ceratitis MacLeay (Diptera: Te- Although numerous studies using molecular meth- phritidae) comprises 89 described species that are ods to identify C. capitata have been published, most native to Africa (De Meyer 2000a, De Meyer and include insufÞcient sampling of species to demon- Copeland 2005, Barr and McPheron 2006). The genus strate taxonomic speciÞcity in the techniques. For includes several highly polyphagous species that are of example, Douglas and Haymer (2001) and Kakouli- economic importance (White and Elson-Harris 1992, Duarte et al. (2001) each reported a technique to Yuval and Hendrichs 2000, Copeland et al. 2006) and discern C. capitata from C. rosa Karsch, and Armstrong recognized as invasive or potentially invasive (De and Ball (2005) a technique to discern C. capitata from Meyer et al. 2008). Of these species, Ceratitis capitata C. rosa Karsch and C. cosyra (Walker), but did not test (Wiedemann), commonly called the Mediterranean additional Ceratitis species. Some studies only com- fruit ßy, is regarded as the most serious international pare C. capitata to fruit ßies from other genera (Arm- pest because of its broad range of hosts, nearly world- strong et al. 1997, Huang et al. 2009). Despite the wide distribution, and impact on trade (White and limited taxon sampling, these tools can be of value in Elson-Harris 1992, Copeland et al. 2002, Vera et al. fruit ßy identiÞcation when nonmolecular informa- 2002, De Meyer et al. 2008, Barr 2009). tion can be integrated in the diagnosis process. Accurate identiÞcation of C. capitata is important To date, the most taxonomically comprehensive for implementing proper quarantine and pest man- molecular tool published for Ceratitis species identi- agement practices. Consequently, morphological Þcation is a Polymerase Chain ReactionÐRestriction tools have been developed to diagnose C. capitata and Fragment Length Polymorphism (polymerase chain other Ceratitis species using adult characters (De reaction [PCR]-restriction fragment-length polymor- Meyer 1996, 1998, 2000b; De Meyer and Freidberg phism [RFLP]) method by Barr et al. (2006). This tool 2005a). Unfortunately, at ports of entry, intercepted is based on relatively good, although not complete, larvae cannot be reliably identiÞed to the species-level sampling of the genus and includes collections from its by using morphology. White and Elson-Harris (1992) ancestral home range in Africa (De Meyer et al. 2004). developed keys for third-instar larvae, but these keys These African collections provide a better estimate of include only 11 species and are based on small sample genetic variation within C. capitata than do collections sizes for most of the species. derived from introduced populations (Barr 2009). In addition, the data set includes samples of Ceratitis 1 Center for Plant Health Science and Technology, Mission Labo- caetrata Munro and Ceratitis pinax Munro, two close ratory, USDA-APHIS, Moore Air Base, Edinburg, TX 78541. relatives of C. capitata that are useful for evaluating the 2 Corresponding author, e-mail: [email protected]. speciÞcity of the diagnostic tool (De Meyer 2005, Barr 3 Department of Entomology, Pennsylvania State University, Uni- and McPheron 2006, Barr and Wiegmann 2009). versity Park, PA 16802. 4 Royal Museum for Central Africa, Leuvensesteenweg 13, 3080 The PCR-RFLP method, developed using the 12S Tervuren, Belgium. rRNA, 16S rRNA, and NADH-dehydrogenase subunit 340 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2

6 mitochondrial DNA loci, was not able to distinguish part of a study of Kenyan fauna (De Meyer et al. 2002, C. capitata from C. caetrata but it could distinguish C. Copeland et al. 2006). With the exception of four C. pinax. However, differences in the internal transcribed capitata specimens (codes 620Ð623, Kenyan samples spacer region 1 (ITS-1) of C. capitata and C. caetrata 1382 by R. Copeland), DNA isolates were reported were able to discriminate the two species based on previously in the Barr et al. (2006) and Barr and DNA sequences and PCR amplicon length (Barr et al. McPheron (2006) studies. In addition to Ceratitis, spe- 2006). cies representing the genera Trirhithrum Bezzi, Cap- Many of the DNA samples used in the PCR-RFLP parimyia Bezzi, Carpophthoromyia Austen, and No- study also were used to estimate the phylogeny of the tomma Bezzi are included for comparison to the genus (Barr and McPheron 2006, Barr and Wiegmann restriction fragment-length polymorphism study. De 2009). Although DNA sequences were generated to Meyer and Freidberg (2005b) have shown that the estimate phylogenies and locate restriction enzyme Capparimyia melanspis Bezzi collection included in Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 recognition sites in the restriction fragment-length Barr et al. (2006) study includes two species: C. mela- polymorphism diagnostic tool, the ability to use the naspis and Capparimyia aenigma De Meyer and DNA sequences as diagnostic information has yet to Freidberg. These samples are treated as a mixture of be evaluated. Recent studies have demonstrated that species and reported as Capparimyia sp. in the current a DNA sequencing approach to identiÞcation of C. study. capitata geographic populations was superior to a Over 600 samples were available for analysis, but PCR-RFLP approach when using the same mitochon- only a subset of 249 samples was selected for DNA drial gene region (Lanzavecchia et al. 2008, Barr barcoding to minimize research costs. For each spe- 2009). cies included in the Barr et al. (2006) study, at least Þve There is growing interest in the use of DNA bar- samples were selected for barcode analysis (Table 1). codes to diagnose pest arthropod species (Armstrong Specimens were selected based on collection location 2010, Floyd et al. 2010). The technology is being in- and restriction fragment-length polymorphism geno- vestigated as a tool for identiÞcation of species that types to maximize geographic and genetic variation have economic, ecological, and health impacts (Ball within each species. In addition, several species rep- and Armstrong 2006, Scheffer et al. 2006, Yancy et al. resented by single specimens in the Barr and 2008, Lowenstein et al. 2009, deWaard et al. 2010, McPheron (2006) phylogenetic study also were in- Naro-Maciel et al. 2010). The Þrst published applica- cluded to increase taxonomic coverage in the data tion of DNA barcoding for tephritid fruit ßy identiÞ- base. Every specimen is numerically coded based on cation focused on the genus Bactrocera Macquart position in the database (Supp. Table S1). (Armstrong and Ball 2005). Currently, DNA barcod- All material was identiÞed to species based on adult ing of tephritid fruit ßies is an objective of interna- morphology (Supp. Table S1). Pinned vouchers (of tional barcode campaigns such as the Quarantine Bar- coding of Life (QBOL, www.qbol.org/UK/) and the entire adult body) representative of each Kenyan col- Tephritid Barcode Initiative (TBI, http://www. lection series are maintained at the Royal Museum for barcoding.si.edu/major_projects.html). Although the Central Africa (KMMA, Tervuren, Belgium), the number of public DNA barcode records is increasing Frost Entomological Museum at the Pennsylvania for Ceratitis species (e.g., 317 specimens with barcode State University (PSUC, University Park, PA), or both. records representing 45 taxa were reported on the For most DNA barcoded samples (Table 1), a tissue Barcode of Life Database, BOLD, on 3 May 2011), no voucher is maintained in ethanol at the APHIS- publications have evaluated the utility of these se- CPHST lab in Edinburg, TX. The tissue vouchers in- quences. clude wings and abdomens from each specimen (the We sequence the proposed barcode region (Hebert head, thoraces, and legs were used in the DNA isola- et al. 2003) of the cytochrome oxidase subunit 1 gene tion process). A description of DNA isolation proce- (COI) using the Ceratitis specimens included in the dures is reported in Barr et al. (2006). Barr et al. (2006) restriction fragment-length poly- PCR and DNA Sequencing. The COI gene was am- morphism study. These collections represent expertly pliÞed from all DNA isolates using the LCO-1490 (5Ј- identiÞed material collected in Africa. We use these GGTCAACAAATCATAAAGATATTGG-3Ј) and HCO- COI data to 1) test whether the COI barcode region 2198 (5Ј-TAAACTTCAGGGTGACCAAAAAATCA-3Ј) can reliably identify C. capitata, and 2) compare the primers reported by Folmer et al. (1994). Samples that diagnostic utility of the DNA sequence method to the failed to generate usable PCR product or DNA se- published PCR-RFLP methodology for diagnoses quences were re-analyzed using the LCO and HCO within the genus Ceratitis. primers. A portion of the DNA isolates that failed to generate good results using the LCO and HCO primer pair after two trials were then retested using the Materials and Methods primer pair TY-J-1460 (5Ј-TACAATTTATCGCCTA- Samples. All specimens were from collections made AACTTCAGCC-3Ј) and C1-N-2191 (5Ј-CCCGGTA- on the African mainland or Re´union. Collection in- AAATTAAAATATAAACTTC-3Ј) as reported by Si- formation per species is provided in Table 1. Collec- mon et al. (1994). PCRs were performed using the tion information per specimen is provided in Supp. Applied Biosystems Gene Amp PCR System 9700 (Ap- Table S1. The majority of samples were generated as plied Biosystems, Foster City, CA). March 2012 BARR ET AL.: DNA BARCODING THE MEDITERRANEAN FRUIT FLY 341

Table 1. General Barcode Information for Taxa included in study

Taxon NPOP NIND NSEQ NHAP NVOU Country (NIND) GenBank Accessions C. anonae 3 6 6 3 6 Kenya (6) JN705083ÐJN705088 C. argenteobrunnea 1 5 5 4 5 Kenya (5) JN705177ÐJN705181 C. bremii 1 1 1 1 1 Kenya (1) JN705240 C. caetrata 2 27 25 16 25 Kenya JN705043ÐJN70568 C. capitata 23 40 34 27 22 Kenya (25), Malawi (2), JN705009ÐJN705042 Reunion (2), Ghana (5) C. colae 4 8 8 1 8 Ghana (8) JN705137ÐJN705144 C. contramedia 1 5 5 1 5 Kenya (5) JN705167ÐJN705171 C. copelandi 2 5 4 3 4 Kenya (5) JN705111ÐJN705114 C. cornuta 1 5 5 1 5 Kenya (5) JN705235ÐJN705239 C. cosyra 4 11 10 7 10 Kenya (6), Mali (4) JN705157ÐJN705166 C. cristata 1 10 10 7 10 Kenya (10) JN705195ÐJN705204 Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 C. curvata 1 5 5 4 5 Kenya (5) JN705127ÐJN705131 C. ditissima 4 5 5 3 5 Kenya (3), Ghana (1), JN705074ÐJN705077 Nigeria (1) C. divaricata 1 6 6 3 6 Kenya (6) JN705151ÐJN705156 C. fasciventris 4 5 5 4 5 Kenya (5) JN705089ÐJN705093 C. ﬂexuosa 2 7 7 5 7 Kenya (7) JN705115ÐJN705121 C. gravinotata 1 1 1 1 1 Kenya (1) JN705231 C. marriotti 4 7 7 2 7 Kenya (7) JN705182ÐJN705188 C. millicentae 1 5 5 4 5 Kenya (5) JN705078ÐJN705082 C. oraria 1 6 6 6 6 Kenya (6) JN705145ÐJN705150 C. penicillata 1 1 1 1 1 Nigeria (1) JN705233 C. pennitibialis 1 1 1 1 1 Kenya (1) JN705232 C. perseus 4 5 5 3 5 Kenya (5) JN705101ÐJN705105 C. pinax 1 5 5 3 5 Kenya (5) JN705069ÐJN705073 C. podocarpi 2 5 5 2 5 Kenya (5) JN705122ÐJN705126 C. querita 1 5 5 4 5 Kenya (5) JN705132ÐJN705136 C. rosa 7 7 7 4 7 Kenya(5), Malawi (1), JN705094ÐJN705100 Reunion (1) C. rubivora 3 5 5 4 5 Kenya (5) JN705106ÐJN705110 C. simi 1 1 1 1 1 Kenya (1) JN705234 C. stictica 1 5 5 3 5 Kenya (5) JN705172ÐJN705176 C. stipula 1 1 1 1 1 Kenya (1) JN705230 C. venusta 1 6 6 1 6 Kenya (6) JN705189ÐJN705194 Ceratitis nsp. 1195 1 5 5 1 5 Kenya (5) JN705225ÐJN705229 Carpophthoromyia 1 5 5 5 5 Kenya (5) JN705215ÐJN705219 dimidiata Capparimyia sp. 1 5 5 2 5 Kenya (5) JN705210ÐJN705214 Notomma sp. (1000) 1 5 5 2 5 Kenya (5) JN705220ÐJN705224 T. coffeae 1 1 1 1 1 Kenya (1) JN705242 T. culcasiae 1 1 1 1 1 Kenya (1) JN705241 T. demeyeri 1 1 1 1 1 Kenya (1) JN705247 T. meladiscum 1 1 1 1 1 Kenya (1) JN705243 T. nigerrimum 2 5 5 3 5 Kenya (5) JN705205ÐJN705209 T. occipitale 1 1 1 1 1 Kenya (1) JN705246 T. senex 1 1 1 1 1 Kenya (1) JN705244 T. teres 1 1 1 1 1 Kenya (1) JN705245 Total 98 249 239 151 227

Sample sizes given for number of populations sampled (NPOP), individuals sampled (NIND), sequences generated (NSEQ), haplotypes identiÞed (NHAP), and tissue vouchers saved (NVOU).

PCR at the APHIS lab was performed in 50-␮l re- At PSU, PCR was performed in 25-␮l reaction vol- action volumes by using 1 ␮l of DNA template or water umes using 1 ␮l of DNA template or water as a neg- (as a negative control). Each reaction had a Þnal ative control. Each reaction had a Þnal concentration concentration of 1X buffer, 2.5 mM MgCl2, 0.2 mM of of 1X buffer, 1.5 mM MgCl2, 0.2 mM of each dNTP, 0.3 each dNTP, 0.2 ␮M of each primer, and 0.625U of ␮M of each primer, and 0.625 U of Qiagen Taq poly- Ex-Taq polymerase (Takara). Cycling conditions were merase. Cycling conditions were 3 min at 94ЊC fol- 3 min at 94ЊC followed by 39 cycles of 94ЊC (20 s)/50ЊC lowed by 39 cycles of 94ЊC (1 min)/50ЊC (1 min)/72ЊC (20 s)/72ЊC (30 s) and an extension of 5 min at 72ЊC. (1 min) and an extension of 10 min at 72ЊC. PCR PCR product was visualized on 1% agarose gels and product was visualized on 1.5% agarose gels and sam- samples were puriÞed using QIAquick puriÞcation col- ples were puriÞed using ExoSAP-IT (USB Corp.). umns (Qiagen, Valencia, CA). DNA sequences were DNA sequences were generated for each primer (i.e., generated for each primer (i.e., sense and anti-sense sense and anti-sense strands) using an ABI 3730XL strands) using an ABI 3730XL DNA analyzer at Davis DNA analyzer at the Huck InstituteÕs Nucleic Acid Sequencing (Davis, CA), a Beckman-Coulter CEQ8000 Facility at Penn State University (University Park, at the APHIS Mission Lab (Edinburg, TX), or both. PA). 342 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2

The TA cloning kit (Invitrogen, Carlsbad, CA) was set and the “nearest neighboring specimens” (i.e., used to clone amplicons of the COI barcode region. most similar sequences) between the two neighboring This cloning was done for samples that failed to gen- species. The average number of nucleotide substitu- erate reliable sequences using direct sequencing. tions per site between populations (Dxy) was calcu- Transformed colonies were grown and selected on lated between each pair of nearest neighboring species LB-agar plates with kanamycin and ampicillin. Plas- using DnaSP. The minimum and maximum intra-spe- mid DNA was puriÞed from transformed colonies us- ciÞc and inter-speciÞc values were graphed for each ing a miniprep kit (Qiagen) and sequenced using taxon. The separation between the maximum intra- universal M13 primers at the Penn State Genomics speciÞc and minimum inter-speciÞc values is the bar- Core Facility (University Park, PA). code gap (Meyer and Paulay 2005). DNA Editing and Analysis. All DNA sequence trace The diagnostic utility of the COI barcode was as- Þles were edited using Sequencher v.4.10 (Gene sessed for C. capitata by evaluating the barcode gap, Codes, Ann Arbor, MI). For each specimen, bidirec- inter-speciÞc connections in the haplotype network, Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 tional sequences were assembled into a contig and and possible character-based diagnostic sites using the used to corroborate base calls and identify conßicting nucleotide diagnostic (ND) process as described by signal. Sequences with evidence of polymorphism Wong et al. (2009). (dual peaks at a site in the trace Þles) in the mito- The genetic variation across the COI gene region chondrial COI marker were included in the analysis if for the entire data set was estimated in a sliding win- the frequency of polymorphism was low (Ͻ1%). Spec- dow of 100 bases with one base steps using DnaSP. The imens that failed to generate usable sequences in both effect of sequence length on the utility of a distance- directions were removed from the analysis. Consensus based identiÞcation was tested for C. capitata by per- sequences from Sequencher were imported into forming sliding window analyses. To measure the ef- MEGA4 (Kumar et al. 2008) and aligned by hand. fect of window size on the barcode gap between C. Sequences that generated evidence of pseudogene capitata and its nearest neighboring species, Dxy values copies, based on the presence of reading frame mu- were calculated between populations using window tations that resulted in premature stop codons, were sizes of 10, 50, 100, 200, 300, 400, and 500 bases and a removed from analysis. The edited consensus Þles constant step size of one base. These minimum and were submitted to GenBank with associated trace Þles. maximum values for intra-speciÞc and inter-speciÞc Unique haplotypes in the data set were identiÞed by variation were plotted against window size. Þrst generating a haplotype data Þle by using DnaSP The Dxy values estimate the mean variation be- v5.10 (Librado and Rozas 2009). These haplotypes tween populations but not between two specimens. As then were conÞrmed or further collapsed by compar- a result, the sliding window using Dxy values could fail ing pair-wise p-distances in MEGA. The program TCS to detect loss of a barcode gap between two speciÞc 1.21 (Clement et al. 2000) was used to generate sta- sequences. Therefore, the sliding window analysis ex- tistical parsimony networks of the haplotypes. The periment was repeated using p-distance (␲) estimates default connection limit of 95% was applied to all between the nearest neighboring specimens from dif- analyses. ferent species (i.e., minimum inter-speciÞc values). Intra- and inter-speciÞc variation values were esti- For comparison, intra-speciÞc variation for each spe- mated for each species by using MEGA. Intra-speciÞc cies was estimated using the most dissimilar specimens variation was estimated as a range using minimum and within a species (i.e., maximum intra-speciÞc varia- maximum genetic distances separating any two bar- tion). codes within a species and as the mean genetic dis- The resulting barcode database was tested against tance among barcodes of conspeciÞc specimens. To 27 unique COI sequences reported previously by Barr obtain inter-speciÞc estimates, all conspeciÞc pair- (2009) for C. capitata. These datasets were compared wise comparisons were excluded from estimates of using minimum genetic distances, character differ- mean, minimum, and maximum p-distance values. For ences between haplotypes, and connectivity of hap- example, the p-distances between a C. pinax barcode lotypes in a statistical parsimony network. The Barr sequence and each of the 234 non-C.pinax barcodes in (2009) sequences were shorter (467 bp) than the the alignment were calculated. This was repeated for sequences generated for this study (Table 1). Because the other four C. pinax barcodes in the database. The of this difference in length and the fact that the 2009 1,170 p-distance values were then used to identify the records were generated from specimens lacking any mean, minimum, and maximum inter-speciÞc varia- series or tissue vouchers, these sequences were not tion for C. pinax. An uncorrected p-distance estimate combined in the general analysis of diagnostic utility. based on the observed proportion of differences between sequences was selected for estimating variation Results in the COI data set instead of the more commonly used Kimura 2-Parameter (K2P) model. Based on previous General PCR and Sequencing Success. All 249 sam- research the p-distance estimate is an appropriate ples selected for barcode analysis generated visible measure for analysis of closely related species (Nei PCR product of the expected size (c. 650bp). The and Kumar 2000, Srivathsan and Meier 2011). majority of these PCR products generated usable The minimum inter-speciÞc distance value was used DNA sequences. Six of the C. capitata samples (011, to identify the “nearest neighboring species” in the data 022, 023, 024, 025, and 027), however, generated se- March 2012 BARR ET AL.: DNA BARCODING THE MEDITERRANEAN FRUIT FLY 343 quences with evidence of intra-individual polymorphism at many (Ն5%) sites. In addition, poor quality DNA sequences were generated from two C. caetrata samples (047 and 056), one Ceratitis copelandi De Meyer and Freidberg sample, and one Ceratitis cosyra sample (417). Eight Ceratitis cristata Bezzi samples generated poor sequences (507, 508, 511, 514, 516, 517, 518, and 519) using the Folmer primers. One Cappa- rimyia sp. sample (551) generated Ͼ1% polymorphic sites (8/603). The C. capitata, C. caetrata, and C. copelandi samples with evidence of high polymorphism levels and poor quality sequence were removed from the analysis. The Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 Capparimyia sp. sample was included in the analysis but will not meet the GenBank criteria as an ofÞcial barcode record according to NCBI standards. The C. cristata and C. cosyra samples were re-ampliÞed and sequenced using the Simon et al. (1994) COI primer Fig. 1. Network of CAP-CAET. The white circles rep- pair (see Materials and Methods). This primer sub- resent C. capitata haplotypes and gray circles represent C. stitution generated good quality barcode sequences caetrata haplotypes. for the C. cristata samples but not for the C. cosyra sample. 582; Supp. Table S2) that were insufÞcient to diagnose The C. cosyra PCR product that failed to generate the species because of shared characters between spe- good sequence data was cloned using the TA-cloning cies. kit (Invitrogen). From that cloning event, four trans- Consistent with the character-based diagnostics, formed colonies were selected for DNA sequencing. the C. capitata and C. caetrata sequences lacked a Three unique genotypes were observed in the four “barcode gap” using genetic distances (Fig. 2). The clones (JN715791Ð4). One genotype (JN715793) had minimum genetic separation between the two species two reading frame mutations. The other two geno- was 0.17%, and the maximum variation within either types (JN715791 and JN715792) had amino acid se- species was Ն1.5%. In comparison, a barcode gap was quences similar to other C. cosyra sequences (2Ð3 observed for the majority of tested Ceratitis species amino acid substitutions), but p-distances distinct (Fig. 2). For example, a strong barcode gap was ob- from the other con-speciÞc barcodes (Ͼ11%). These served for Ceratitis pinax, a close relative of C. capitata sequences were most similar to Ceratitis fruit ßy DNA and C. caetrata. using NCBI BLAST searchers. Based on these results Fig. 3 displays barcode gap information after rede- the specimen was not included in further barcode Þning taxonomic units in the data set for some of the analyses of the species. species. Recalculation of genetic distances when using The Þnal data set included 239 barcode sequences a taxon, called “CAP-CAET,” that collapses the species representing 44 species and resulted in a 603-bp align- C. capitata and C. caetrata resulted in a barcode gap ment containing no gaps. There were a few examples (Fig. 3). The maximum genetic variation in the CAP- of single amino acid substitutions within species (i.e., CAET taxon was 2% and the minimum separation C. capitata: I/T; C. perseus De Meyer and Copeland: between the taxon and its nearest neighbor, C. pinax, A/T; C. oraria De Meyer and Copeland: T/A; C. stictica was 6.7% (Supp. Fig. S3). Separation of the taxa CAP- Bezzi: M/V), but no evidence of reading frame mu- CAET and C. pinax was also supported by 32 Þxed tations. differences and two unconnected networks. C. capitata Identiﬁcation. The alignment included A sliding window analysis of genetic distance (Dxy) 34 C. capitata barcode sequences of which 27 repre- between the two taxa, CAP-CAET and C. pinax, dem- sented unique haplotypes. Based on the statistical par- onstrates that divergence values can vary across the simony network, the C. capitata haplotypes are con- COI gene barcode region (Fig. 4). The trend for in- nected in a network with 16 C. caetrata haplotypes creasing variation toward the 3Ј end is consistent with (Fig. 1). No haplotypes were shared between the two genetic diversity estimates (␲) for the entire COI data species. However, the positioning of haplotypes in the set (Fig. 4). Additional sliding window analyses show “C. capitataϩC. caetrata” (CAP-CAET) network does that a window size of 200 bases should generate a not suggest a clear separation of gene pools for the two barcode gap between the CAP-CAET and C. pinax species. taxa (Fig. 5). Although a barcode gap was observable There were no diagnostic character states in the using a 100 base window and Dxy estimates, a com- alignment that could distinguish C. capitata from C. parison of p-distances of the two most similar inter- caetrata. Wong et al. (2009) deÞne these characters as speciÞc sequences (C. capitata 015 and C. pinax 060) nucleotide diagnostics (ND). The two species lacked lacked the gap at this window size (Fig. 6). both simple ND (sND) and compound ND (cND) All 27 of the C. capitata COI haplotypes docu- characters. The cND search identiÞed eleven variable mented by Barr (2009) were genetically similar to sites (60, 99, 259, 342, 462, 486, 508, 537, 543, 579, and barcodes in the data base. Seven of the 2009 haplo- 344 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2 Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021

Fig. 2. Variation estimates and barcode gap for species. (Online Þgure in color.) types were identical to a C. capitata barcode in the cies, morphological evidence has not been docu- new data set. Although no COI sequences were shared mented to support revision of the taxon. between C. capitata and C. caetrata for the 50 se- The taxa Capparimyia sp. and Carpophthoromyia quences generated in the current study, one of the dimidiata Bezzi generated barcode gaps over 4%, but Barr (2009) C. capitata sequences was identical to a C. also exhibited relatively high levels of intra-speciÞc caetrata sequence. The other Barr (2009) sequences variation (6Ð7%). No congeners are included in the were Ͼ99% similar to the CAP-CAET barcodes. study for these two taxa, so it is uncertain if the ob- Barcoding of Other Ceratitis Species. Barcode gaps served barcode gaps are biologically meaningful. Al- were observed for 25 of the other species in the study. though De Meyer (2006) reports that Carpophthoro- The barcode gap value, calculated between nearest myia vittata (F.) has a parapatric distribution in Kenya neighboring taxa, was between 3 and 4% for ten spe- with C. dimidiata and the two species share a host cies and over 4% for 14 species. The smallest barcode range in Kenya, there is no evidence that this second gap (0.95%) was for Ceratitis cosyra and it resulted species is present in our collection. The possibility that from a high level of intra-speciÞc variation (9%). This this nominal species comprises multiple cryptic spe- large variation is the result of divergence between a cies, however, cannot be excluded without further single haplotype, observed in two individuals (403, study. 404), and the other six haplotypes for the species. As Ceratitis argenteobrunnea Munro also exhibited a expected, partitioning the haplotypes of the species high level of intra-speciÞc variation (6.6%) but the into genetic clusters (i.e., C. cosyra A and B) helped to value overlapped with the minimum inter-speciÞc di- generate a barcode gap (Fig. 3). This is consistent with vergence (5.14%). Similar to the variation results for the unconnected networks and large number of mu- C. cosyra, partitioning the haplotypes into genetic tations separating haplotypes of the species (Fig. 7). C. argenteobrunnea The two divergent C. cosyra genetic types were gen- clusters (i.e., A and B) helped to erated from expertly identiÞed material. Although it is generate a barcode gap (Fig. 3) and the haplotypes possible that the name C. cosyra includes cryptic spe- failed to generate a single network (Fig. 7). The other three species that failed to generate a barcode gap were C. fasciventris (Bezzi), C. anonae Graham, and C. rosa. Although the maximum intra- speciÞc variation was not exceptionally high (0.7Ð 2.2%) for these species, the minimum inter-speciÞc variation was low (0Ð0.5%). These three species are members of the FAR species complex (Barr and McPheron 2006, Virgilio et al. 2008) and expected to exhibit low levels of genetic separation. Haplotypes from FAR complex species connect into a single network (Fig. 8) and one haplotype is shared between C. fasciventris and C. rosa. Virgilio et al. (2008) report additional information on shared mitochondrial haplotypes and genetic similarity based on more extensive sampling of the three FAR species. Fig. 3. Variation and barcode gaps for taxa collapsed or Thirteen of the species in the study were repre- split. (Online Þgure in color.) sented by a single specimen and could not be tested March 2012 BARR ET AL.: DNA BARCODING THE MEDITERRANEAN FRUIT FLY 345 Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021

Fig. 4. Sliding window of total data set. 100 base window (one step) for all sequences (n ϭ 239) in study and a comparison between C. capitata/C. caetrata samples versus C. pinax samples. for barcode gaps. The inter-speciÞc values are re- clusters (i.e., C. cosyra A and B, and C. argenteobrunnea ported for these species in Supp. Fig. S4. A and B) helped to generate barcode gaps (Fig. 3). Results from parsimony network analyses support This is consistent with the unconnected networks and the barcode gap data. For example, the haplotypes of large number of mutations separating haplotypes of a species with low intra-speciÞc variation and barcode species (Fig. 7). Based on the data set, these intra- gaps Ն3% generated species-speciÞc haplotype net- speciÞc separations were similar to observed genetic works (Supp. Fig. S5), the haplotypes of species that distances that separate Ceratitis species. lacked adequate inter-speciÞc variation generated multi-species networks (Figs. 1 and 8), and the hap- Discussion lotypes of species that exhibited high levels of intra- speciÞc variation generated multiple (unconnected) Identiﬁcation of C. capitata Using a DNA Barcode. networks (Fig. 7). It is not possible to diagnose a C. capitata sample to When the three closely related FAR complex spe- the level of species by using only a COI DNA barcode cies were grouped and treated as a single taxon in the sequence. Consistent with previous genetic evidence barcode study, a small barcode gap was measurable (Barr and McPheron 2006), DNA barcodes were sim- (Fig. 3). Unlike the CAP-CAET taxon, the barcode gap ilar for specimens of C. capitata and its sister taxon, C. was small (Ͻ1%) relative to the intra-taxonomic di- caetrata. The two species lacked a “barcode gap” versity (2.5%) of FAR. (Meyer and Paulay 2005) for COI, precluding a phe- As expected, partitioning the haplotypes of species netic-based identiÞcation by using genetic distances. with high intra-speciÞc genetic diversity into genetic Comparison of the 27 C. capitata and 16 C. caetrata

Fig. 5. Barcode gap between CAP-CAET and C. pinax populations using different fragment sizes. Comparison of sliding window results for taxa as the window size increases. The mean genetic distance is reported for variation within C. pinax (pop-P, n ϭ 5) and C. capitata/C. caetrata (pop-C, n ϭ 59) samples and between the two taxonomic groups (pop-CvP, n ϭ 64). (Online Þgure in color.) 346 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2

Fig. 6. Barcode gap between CAP-CAET and C. pinax Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 individuals using different fragment sizes. Comparison of sliding window results for pairwise comparisons between specimens as the window size increases. The range of distances is reported for comparisons between two C. pinax (PvP) samples, one C. capitata and one C. caetrata sample Fig. 8. Network of FAR complex species. (CvC), and one C. capitata sample and one C. pinax sample (CvP). The PvP (071Ð072) and CvC (033Ð044) samples were connected all C. capitata and C. caetrata haplotypes in selected to maximize genetic distance and the CvP (015Ð069) a single network. samples were selected to minimize the distance between taxa. (Online Þgure in color.) Sarkar et al. (2002, 2008) and Wong et al. (2009) described procedures to identify diagnostic character states within a barcode data set. Using Wong et al.Õs haplotypes in our data set against 848,360 species bar- process, our data failed to identify a Þxed diagnostic code records in the Barcode of Life Database, BOLD difference between the two species using individual (Ratnasingham and Hebert 2007; www.barcodinglife. sites (i.e., simple nucleotide diagnostic, sND) or a org/), resulted in no species level matches because of combination of states at multiple sites (compound the high sequence similarity between these two spe- nucleotide diagnostic, cND). Wong et al. (2009) pro- cies [search performed 20 January 2011]. posed that in the absence of a Þxed cND, it is still Hart and Sunday (2007) proposed that statistical possible to identify private haplotypes called “condi- parsimony networks could provide an alternative ap- tional” NDs. Using our barcode data set, all haplotypes proach to distinguish species based on DNA barcode for these species are private (not shared between data. Using a 95% connection limit as deÞned by species) and conditional cNDs are present. However, Templeton et al. (1992), it is possible to statistically based on the haplotype network and genetic diversity sort haplotypes of different species into unconnected estimates, we suspect that the private haplotypes are networks (Chen et al. 2010). This approach, however, an artifact of sampling and not useful as diagnostic

Fig. 7. Haplotypes that did not connect into a single network according to taxonomy. March 2012 BARR ET AL.: DNA BARCODING THE MEDITERRANEAN FRUIT FLY 347 information. A comparison of our data with C. capitata polymorphism method of Barr et al. (2006) also re- COI sequences published previously by Barr (2009) quires analysis with multiple primer sets for most spe- identiÞed a C. capitata sequence that is identical to a cies in the tool. The rate of PCR-RFLP artifacts or C. caetrata barcode. failures using the restriction fragment-length poly- In contrast, the phenetic-based and character-based morphism method was around 2%, whereas 6% of the barcoding methods work well to identify a taxon com- samples included in the COI barcode study failed to prising C. capitata and C. caetrata. By merging these generate usable barcodes on the Þrst attempt using the species into one taxon, called CAP-CAET, the next Folmer et al. primers. most similar species is C. pinax. These lineages can be There is evidence of within-genome polymorphism distinguished by a barcode gap, unconnected net- in the COI gene for individuals of some species. In C. works, and Þxed character differences. cosyra, there was evidence of at least one pseudogene The barcodes reported in our study include Ϸ600 copy of the COI barcode. The presence of polymor- bases, but it is possible for data generated during a phisms and pseudogenes of mitochondrial DNA has Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 query of the database to be shorter. For example, poor been documented previously for various loci in the PCR or sequencing reactions (caused by poor quality genus for C. rosa (using the ND6 gene), C. marriotti sample or introduced error) could result in the ex- Munro (using the 3Ј half of COI gene), and C. venusta clusion of bases during the editing process or require (Munro) (using the 3Ј half of the COI gene) (Barr and the use of alternate primers to generate the data. To McPheron 2006). Magnacca and Brown (2010) discuss improve PCR or sequencing success rates with de- the problems associated with DNA barcoding poly- graded DNA of historical samples, Van Houdt et al. morphic and heteroplasmic markers. Although there (2009) proposed the use of several shorter (mini- were no conÞrmed instances of pseudogene copies in barcode) regions of the COI gene to generate fruit ßy the Barr et al. (2006) study, PCR-RFLP artifacts sug- barcodes. Consequently, the minimum fragment size gested polymorphism in the 12S rDNA and NADH- required to provide a correct identiÞcation of Ceratitis dehydrogenase 6 markers. capitata was calculated by testing sliding windows for The COI barcode and restriction fragment-length evidence of a barcode gap. Based on our data, any COI polymorphism diagnostic tools performed equally barcode fragment Ն300 bases should be adequate for well to diagnose most species. Neither method could providing a reliable identiÞcation of the CAP-CAET distinguish between C. capitata and C. caetrata or taxon. A comparison of correct identiÞcations across among the three species of the FAR complex (C. insect orders by using mini-barcodes also noted a fasciventris, C. anonae, and C. rosa). As expected, these reduction in performance when fragment size de- unresolved taxonomic clusters represent two lineages creased (Virgilio et al. 2010). of closely related species (Barr and McPheron 2006, Although it is not possible to provide an absolute Virgilio et al. 2008). As described previously, it is identiÞcation of C. capitata samples by using the COI possible to use either molecular method to diagnose barcode, the ability to limit the pool of possible Ce- the FAR taxon or the CAP-CAET taxon. tratitis species in a query to just two species (C. capi- Although it is possible to identify C. argenteobrun- tata or C. caetrata) is useful. If these are the only two nea and C. cosyra by using the restriction fragment- possible species, then the ITS-1 locus can be used to length polymorphism method, these species lack a distinguish the species (Barr et al. 2006). barcode gap suitable for a distance-based identiÞca- Based on differences in the biogeography and ecol- tion. This difference in identiÞcation capabilities is in ogy of C. capitata and C. caetrata, it is also possible to part because of the use of different gene regions and make a species identiÞcation that is conditional on how we deÞne the diagnostic characters. The restric- nonmolecular information. For instance, unlike the tion fragment-length polymorphism method is based invasive Mediterranean fruit ßy, C. caetrata has a more on the concept of private alleles or “conditional” hap- restricted host range that includes only indigenous lotypes (Wong et al. 2009), and the character differ- wild fruits (but not commercially grown fruits) and ences that separate species are not Þxed within each has not been detected outside of Kenya (De Meyer species. When DNA barcodes are analyzed as private 2001, De Meyer et al. 2002). The likelihood of inter- or conditional haplotypes, it is also possible to identify cepting C. capitata at U.S. ports of entry should be these two species. higher than that for C. caetrata. Conditional identiÞ- Using restriction fragment-length polymorphism, cations, however, are context dependent (i.e., the di- the Þve C. argenteobrunnea samples included in our agnostic process for one port or country may not be study generate three distinct mitochondrial haplo- the same for another port or country) and should be types. Using the COI barcode, the Þve samples gen- justiÞable with scientiÞc information. erate four distinct DNA sequence haplotypes. Al- Comparison of the COI Barcode and Original Re- though it is possible to estimate which restriction striction Fragment-length Polymorphism Method for fragment-length polymorphism haplotypes are most Identiﬁcation of Ceratitis species. From an operational similar, it is difÞcult to understand the evolutionary perspective, the proposed “universal” barcode primers signiÞcance of these similarities because of the limited of Folmer et al. (1994) did not work consistently for number of characters in the data. Based on the COI the genus Ceratitis because some species such as C. barcode haplotypes, it is clear that the most distant cristata required alternate primers to generate usable haplotype is separated by at least 37 mutational steps barcodes. Similarly, the restriction fragment-length (Ͼ6% divergence). This is unexpectedly high in com- 348 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2 parison to COI barcode variation measured in the is one large network for the species). Proper sampling other Ceratitis species. In the restriction fragment- will be important for deciding how to use barcodes for length polymorphism method, this divergent haplo- identiÞcation of Ceratitis species and other economic type is just another character, but in the barcode data pests. set this haplotype is ßagged as a problem because of When evaluated using character and distance in- the quantitative distance and haplotype network es- formation, the DNA barcode method performs as well timations. The ability to detect variation within spe- as the restriction fragment-length polymorphism cies is an advantage of the sequencing approach. DNA method to identify Ceratitis species. Although the re- barcodes provide a greater level of quantitative infor- striction fragment-length polymorphism method is mation than the restriction fragment-length polymor- still a more economical approach to molecular iden- phism method and an opportunity to estimate uncer- tiÞcation based on reagents and instrumentation, tainty in the reference database. other factors can affect the operational costs for high The restriction fragment-length polymorphism throughput analysis of samples. From a diagnostics Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 method documented two diagnostic forms for C. co- standpoint, the DNA barcode method enables better syra, but barcodes were generated only from speci- quantitative analysis, provides more information for mens with the dominant form. This is because the detecting and assessing false positives and false neg- sample with the uncommon restriction fragment- atives, and uses a data set that can be easily shared and length polymorphism form generated multiple copies accessed by the greater research community (Barr of the COI gene and was excluded from further anal- 2009). ysis. Although the other 10 C. cosyra samples shared an identical restriction fragment-length polymorphism haplotype, two of these ßies had barcode sequences at Acknowledgments least 52 mutational steps away from the other eight We thank Lisa Ledezma and Amanda Cook for technical ßies. Although these data may be useful for identifying assistance in the experiments, and Raul Ruiz and two anon- cryptic species and guiding systematic revisions ymous reviewers for suggestions that have improved the within the genus, it is important that the name C. manuscript. cosyra stay associated with the barcode records until a formal taxonomic change is made. DNA barcodes, like other molecular diagnostic methods, are designed References Cited to serve as surrogate characters for recognized species Armstrong, K. 2010. DNA barcoding: a new module in New (Carew et al. 2005). ZealandÕs plant biosecurity diagnostic toolbox. OEPP/ Hebert et al. (2003, 2004) proposed using a stan- EPPO Bull. 40: 91Ð100. dardized threshold value as a way to identify genetic Armstrong, K. F., and S. L. Ball. 2005. DNA barcodes for distances inconsistent with intra-speciÞc variation. Al- biosecurity: invasive species identiÞcation. Phil. Trans. R. though a threshold of Ϸ3% had been estimated by Soc. B 360: 1813Ð1823. Hebert et al. for several animal lineages, this value is Armstrong, K. F., C. M. Cameron, and E. R. Frampton. 1997. supposed to be estimated empirically for different Fruit ßy (Diptera: Tephritidae) species identiÞcation: a taxa. Despite problems with using a standard cut-off rapid molecular diagnostic technique for quarantine ap- threshold value to identify a large group of species plication. Bull. Entomol. Res. 87: 111Ð118. Ball, S. L., and K. F. Armstrong. 2006. DNA barcodes for (DeSalle et al. 2005, Cognato 2006, Meier et al. 2006, insect pest identiÞcation: a test case with tussock moths Wiemers and Fiedler 2007, Little and Stevenson 2007), (Lepidoptera: Lymantriidae). Can. J. For. Res. 36: 337Ð the utility of threshold values for diagnosing pests 350. likely will be context dependent (Armstrong 2010). Barr, N. B. 2009. Pathway analysis of Ceratitis capitata (Dip- Based on maximum intra-speciÞc variation, our data tera: Tephritidae) using mitochondrial DNA. J. Econ. lack a standard “cut-off” value for the genus. Using ten Entomol. 102: 401Ð411. times the mean intra-speciÞc value for the entire data Barr, N. B., and B. A. McPheron. 2006. Molecular phyloge- set, as proposed by Hebert et al. (2004), the threshold netics of the genus Ceratitis (Diptera: Tephritidae). Mol. would be near 8%. This is too high for most of our Phylogenet. Evol. 38: 216Ð230. Barr, N. B., and B. M. Wiegmann. 2009. Phylogenetic rela- sampled species. Wong et al. (2009) explained that the tionships of Ceratitis fruit ßies inferred from nuclear CAD 95% statistical parsimony connection limit that deÞnes and tango/ARNT gene fragments: testing monophyly of network connectivity also functions as a threshold. the subgenera Ceratitis (Ceratitis) and C. (Pterandrus). Therefore, when species with high intra-speciÞc vari- Mol. Phylogenet. Evol. 53: 412Ð424. ation (i.e., C. cosyra and C. argenteobrunnea) were Barr, N. B., R. S. Copeland, M. De Meyer, D. Masiga, H. G. divided into multiple operational taxonomic units Kibogo, M. K. Billah, E. Osir, R. A. Wharton, and B. A. (e.g., C. cosyra-A versus C. cosyra-B) based on net- McPheron. 2006. Molecular diagnostics of economically works, the maximum intra-speciÞc values for taxa important Ceratitis fruit ßy species (Diptera: Tephriti- were consistent with the proposed 3% threshold for dae) in Africa using PCR and RFLP analyses. Bull. En- tomol. Res. 96: 505Ð521. animal species. Additional information is required to Carew, M. E., V. Pettigrove, and A. A. Hoffman. 2005. The conÞrm that COI diversity within C. cosyra segregates utility of DNA markers in classical taxonomy using cyto- into two diagnosable networks (i.e., two genetic chrome oxidase I markers to differentiate Australian Cla- groups for the same species) or if genotypes linking dopelma (Diptera: Chironomidae) midges. Ann. Ento- these two networks have yet to be sampled (i.e., there mol. Soc. Am. 98: 587Ð594. March 2012 BARR ET AL.: DNA BARCODING THE MEDITERRANEAN FRUIT FLY 349

Chen, H., M. Strand, J. L. Norenburg, S. Sun, H. Kajihara, tions of Mediterranean fruit ßy (Ceratitis capitata) and A. V. Chernyshev, S. A. Maslakova, and P. Sundberg. Natal fruit ßy (Ceratitis rosa). J. Biogeogr. 35: 270Ð281. 2010. Statistical parsimony networks and species assem- DeSalle, R., M. G. Egan, and M. Siddall. 2005. The unholy blages in Cephalotrichid Nemerteans (Nemertea). PLoS trinity: taxonomy, species delimitation and DNA barcod- ONE 5: e12885. ing. Phil. Trans. Soc. B 360: 1905Ð1916. Cognato, A. I. 2006. Standard percent DNA sequence dif- deWaard, J. R., A. Mitchell, M. A. Keena, D. Gopurenko, ference for insects does not predict species boundaries. J. L. M. Boykin, K. F. Armstrong, M. G. Pogue, J. Lima, R. Econ. Entomol. 99: 1037Ð1045. Floyd, R. H. Hanner, et al. 2010. Towards a global bar- Copeland, R. S., R. A. Wharton, Q. Luke, and M. De Meyer. code library for Lymantria (Lepidoptera: Lymantriinae) 2002. Indigenous hosts of Ceratitis capitata (Diptera: Te- tussock moths of biosecurity concern. PLoS ONE 5: phritidae) in Kenya. Ann. Entomol. Soc. Am. 95: 672Ð694. e14280. Copeland, R. S., R. A. Wharton, Q. Luke, M. De Meyer, S. Douglas, L. J., and D. S. Haymer. 2001. Ribosomal ITS1 Lux, N. Zenz, P. Machera, and M. Okumu. 2006. Geo- polymorphism in Ceratitis capitata and Ceratitis rosa graphic distribution, host fruit, and parasitoids of African (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 94: 726Ð Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 fruit ßy pests Ceratitis anonae, Ceratitis cosyra, Ceratitis 731. fasciventris, and Ceratitis rosa (Diptera: Tephritidae) in Floyd, R., J. Lima, J. deWaard, L. Humble, and R. Hanner. Kenya. Ann. Entomol. Soc. Am. 99: 261Ð278. 2010. Common goals: policy implications of DNA bar- Clement, M., D. Posada, and K. A. Crandall. 2000. TCS: a coding as a protocol for identiÞcation of arthropod pests. computer program to estimate gene genealogies. Mol. Biol. Invasions 12: 2947Ð2954. Ecol. 9: 1657Ð1660. Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. De Meyer, M. 1996. Revision of the subgenus Ceratitis 1994. DNA primers for ampliÞcation of mitochondrial (Pardalaspis) Bezzi, 1918 (Diptera: Tephritidae: Cera- cytochrome c oxidase I from diverse metazoan inverte- titini). Syst. Entomol. 21: 15Ð26. brates. Mol. Marine Biol. Biotechnol. 3: 294Ð299. De Meyer, M. 1998. Revision of the subgenus Ceratitis Hart, M. W., and J. Sunday. 2007. Things fall apart: biolog- (Ceratalaspis) Hancock (Diptera: Tephritidae). Bull. En- ical species form unconnected parsimony networks. Biol. tomol. Res. 88: 257Ð290. Lett. 3: 509Ð512. De Meyer, M. 2000a. Phylogeny of the genus Ceratitis Hebert, P.D.N., A. Cywinska, S. L. Ball, and J. R. deWaard. (Dacinae: Ceratitidini), pp. 409Ð428. In M. Aluja and 2003. Biological identiÞcations through DNA barcodes. A. L. Norrbom (eds.), Fruit ßies (Tephritidae): phylog- Proc. R. Soc. Lond. B 270: 313Ð321. Hebert, P.D.N., A. Cywinska, S. L. Ball, and J. R. deWaard. eny and evolution of behavior. CRC, Boca Raton, FL. 2004. Biological identiÞcations through DNA barcodes. De Meyer, M. 2000b. Systematic revision of the subgenus Proc. R. Soc. Lond. B 270: 313Ð321. Ceratitis MacLeay s. s. (Diptera: Tephritidae). Zool. J. Huang, C.-G., J.-C. Hsu, D. S. Haymer, G.-C. Lin, and W.-J. Linn. Soc. 128: 439Ð467. Wu. 2009. Rapid identiÞcation of the Mediterranean De Meyer, M. 2001. Distribution patterns and host-plant fruit ßy (Diptera: Tephritidae) by loop-mediated isother- relationships within the genus Ceratitis MacLeay (Dip- mal ampliÞcation. J. Econ. Entomol. 102: 1239Ð1246. tera: Tephritidae) in Africa. Cimbebasia 17: 219Ð228. Kakouli-Duarte, T., D. G. Casey, and A. M. Burnell. 2001. De Meyer, M. 2005. Phylogenetic relationships within the Development of a diagnostic DNA probe for the fruit ßies fruit ßy genus Ceratitis MacLeay (Diptera: Tephritidae), Ceratitis capitata and Ceratitis rosa (Diptera: Tephriti- derived from morphological and plant evidence. Insect dae) using ampliÞed fragment-length polymorphism. J. Syst. Evol. 36: 459Ð480. Econ. Entomol. 94: 989Ð997. De Meyer, M. 2006. Systematic revision of the fruit ßy genus Kumar, S., M. Nei, J. Dudley, and K. Tamura. 2008. MEGA: Carpophthoromyia Austen (Diptera, Tephritidae). Zoot- a biologist-centric software for evolutionary analysis of axa 1235: 1Ð48. DNA and protein sequences. Brief. Bioinform. 9: 299Ð306. De Meyer, M., and R. S. Copeland. 2005. Description of new Lanzavecchia, S. B., J. L. Cladera, P. Faccio, N. Petit Marty, Ceratitis MacLeay (Diptera: Tephritidae) species from J. C. Vilardi, and R. O. Zandomeni. 2008. Origin and Africa. J. Nat. Hist. 39: 1283Ð1297. distribution of Ceratitis capitata mitochondrial DNA hap- De Meyer, M., and A. Freidberg. 2005a. Revision of the lotypes in Argentina. Ann. Entomol. Soc. Am. 101: 627Ð subgenus Ceratitis (Pterandrus) Bezzi (Diptera: Tephriti- 638. dae). Isr. J. Entomol. 35-. 36: 197Ð315. Librado, P., and J. Rozas. 2009. DnaSP version 5: a software De Meyer, M., and A. Freidberg. 2005b. Revision of the fruit for comprehensive analysis of DNA polymorphism data. ßy genus Capparimyia (Diptera, Tephritidae). Zoologica Bioinformatics 25: 1451Ð1452. Scr. 34: 279Ð303. Little, D. P., and D. W. Stevenson. 2007. A comparison of De Meyer, M., R. S. Copeland, S. A. Lux, M. Mansell, S. algorithms for the identiÞcation of specimens using DNA Quilici, R. Wharton, I. M. White, and N. J. Zenz. 2002. barcodes: examples from gymnosperms. Cladistics 23: Annotated check list of host plants for afrotropical fruit 1Ð21. ßies (Diptera: Tephritidae) of the genus Ceratitis. Doc- Lowenstein, J. H., G. Amato, and S.-O. Kolokotronis. 2009. umentations Zoologiques, Musee Royal de lÕAfrique Cen- The real maccoyii: identifying tuna sushi with DNA bar- trale 27: 1Ð91. codes Ð contrasting characteristic attributes and genetic De Meyer, M., R. S. Copeland, R. A. Wharton, and B. A. distances. PLoS ONE 4: e7866. McPheron. 2004. On the geographical origin of the med- Magnacca, K. N., and M.J.F. Brown. 2010. Tissue segrega- ßy, Ceratitis capitata (Wiedemann), pp. 45Ð53. In B. tion of mitochondrial haplotypes in heteroplasmic Ha- Barnes (ed.), Proceedings, the 6th International Sympo- waiian bees: implications for DNA barcoding. Mol. Ecol. sium on Fruit Flies of Economic Importance, 6Ð10 May Res. 10: 60Ð68. 2002, Stellenbosch, South Africa. Isteg ScientiÞc Publi- Meier, R., K. Shiyang, G. Vaidya, and P.K.L. Ng. 2006. DNA cations, Irene, South Africa. barcoding and taxonomy in Diptera: a tale of high intras- De Meyer, M., M. P. Robertson, A. T. Peterson, and M. W. peciÞc variability and low identiÞcation success. Syst. Mansell. 2008. Ecological niches and potential distribu- Biol. 55: 715Ð728. 350 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 105, no. 2

Meyer, C. P., and G. Paulay. 2005. DNA barcoding: error ural history collections of Tephritid fruitßies (Tephriti- rates based on comprehensive sampling. PLoS Biol. 3: dae, Diptera) using mini barcodes. Mol. Ecol. Res. 10: e422. 459Ð465. Nei, M., and S. Kumar. 2000. Molecular evolution and phy- Vera, M. T., R. Rodriguez, D. F. Segura, J. L. Cladera, and logenetics. Oxford University Press, New York. R. W. Sutherst. 2002. Potential geographical distribution Naro-Maciel, E., B. Reid, N. N. Fitzsimmons, M. Le, R. De- of the Mediterranean fruit ßy, Ceratitis capitata (Diptera: Salle, and G. Amato. 2010. DNA barcodes for globally Tephritidae), with emphasis on Argentina and Australia. threatened marine turtles: a registry approach to docu- Environ. Entomol. 31: 1009Ð1022. menting biodiversity. Mol. Ecol. Res. 10: 252Ð263. Virgilio, M., T. Backeljau, N. Barr, and M. De Meyer. 2008. Ratnasingham, S., and P.D.N. Hebert. 2007. BOLD: the Bar- Molecular evaluation of nominal species in Ceratitis fas- code of Life Data System (www.barcodinglife.org). Mol. civentris, C. anonae, C. rosa species complex (Diptera: Ecol. Notes 7: 355Ð364. Tephritidae). Mol. Phylogenet. Evol. 48: 270Ð280. Sarkar, I. N., P. J. Planet, T. E. Bael, S. E. Stanley, M. Siddall, Virgilio, M., T. Backeljau, B. Nevado, and M. De Meyer. R. DeSalle, and D. H. Figurski. 2002. Characteristic at- 2010. Comparative performances of DNA barcoding Downloaded from https://academic.oup.com/aesa/article/105/2/339/120778 by guest on 27 September 2021 tributes in cancer microarrays. J. Biomed. Inform. 35: across insect orders. BMC Bioinform. 11: 206. 111Ð122. White, I. M., and M. M. Elson-Harris. 1992. Fruit ßies of Sarkar, I. N., P. J. Planet, and R. DeSalle. 2008. CAOS soft- economic signiÞcance: their identiÞcation and bionom- ware for use in character-based DNA barcoding. Mol. ics. CAB, Wallingford, United Kingdom. Ecol. Res. 8: 1256Ð1259. Wiemers, M., and K. Fiedler. 2007. Does the DNA barcod- Scheffer, S. J., M. L. Lewis, and R. C. Josh. 2006. DNA ing gap exist? Ð a case study in blue butterßies (Lepi- barcoding applied to invasive leafminers (Diptera: Agro- doptera: Lycaenidae). Front. Zool. 4: 8. myzidae) in the Philippines. Ann. Entomol. Soc. Am. 99: Wong, E.H.-K., M. S. Shivji, and R. H. Hanner. 2009. Iden- 204Ð210. tifying sharks with DNA barcodes: assessing the utility of Simon, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. a nucleotide diagnostic approach. Mol. Ecol. Res. 9: 243Ð Flook. 1994. Evolution, weighting, and phylogenetic 256. utility of mitochondrial gene sequences and a compila- Yancy, H. F., T. S. Zemlak, J. A. Mason, J. D. Washington, B. J. tion of conserved polymerase chain reaction primers. Tenge, N.-L.T. Nguyen, J. D. Barnett, W. E. Savary, W. E. Ann. Entomol. Soc. Am. 87: 651Ð701. Hill, M. M. Moore, et al. 2008. Potential use of DNA Srivathsan, A., and R. Meier. 2011. On the inappropriate use barcodes in regulatory science: applications of the Reg- of Kimura 2-parameter (K2P) divergences in the DNA- ulatory Fish Encyclopedia. J. Food Prot. 71: 210Ð217. barcoding literature. Cladistics 27: 1Ð5. Yuval, B., and J. Hendrichs. 2000. Behavior of ßies in the Templeton, A. R., K. A. Crandall, and C. F. Sing. 1992. A genus Ceratitis (Dacinae: Ceratitidini), pp. 429Ð457. In cladistics analysis of phenotypic associations with haplo- M. Aluja and A. L. Norrbom (eds.), Fruit ßies (Tephriti- types inferred from restriction endonuclease mapping dae): Phylogeny and evolution of behavior. CRC, Boca and DNA sequence data. III. Cladogram estimation. Ge- Raton, FL. netics 132: 619Ð633. Van Houdt, J.K.J., F. C. Breman, M. Virgilio, and M. De Meyer. 2009. Recovering full DNA barcodes from nat- Received 7 June 2011; accepted 4 October 2011.