Species determination of Malaysian Bactrocera pests using PCR-RFLP analyses (Diptera: Tephritidae)

Transcription

1 Research Article Received: 15 January 2009 Revised: 21 August 2009 Accepted: 5 October 2009 Published online in Wiley Interscience: ( DOI /ps.1886 Species determination of Malaysian Bactrocera pests using PCR-RFLP analyses (Diptera: Tephritidae) Tock H Chua, a Yi Vern Chong a and Saw Hoon Lim b Abstract BACKGROUND: Identification of Bactrocera carambolae Drew and Hancock, B. papayae Drew and Hancock, B. tau Walker, B. latifrons Hendel, B. cucurbitae Coquillett, B. umbrosa Fabricius and B. caudata Fabricius would pose a problem if only a body part or an immature stage were available. Analysis of polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) of cytochrome oxidase I (COI) gene using primers COIR, COIF, UEA7 and UEA10 and restriction enzymes (MseI, RsaI and Alu1) was carried out. The banding profiles in the electrophoresis gel were analysed. RESULTS: The COI gene in six Bactrocera spp. was successfully amplified by COIR and COIF, as well as UEA7 and UEA10, while B. caudata was amplified successfully only by UEA primers. Using COI amplified PCR products and restriction enzymes, distinct banding profiles for B. tau,b. latifrons,b. cucurbitae and B. umbrosa were observed, but not for B. carambolae and B. papayae. However, using UEA7, UEA10 and RsaI, B. caudata could be identified, while B. carambolae and B. papayae might possibly be separated from one another. It was also shown that adult body parts or immature life stages of B. carambolae, B. papayae, B. latifrons and B. cucurbitae produced the same banding profiles as the adults. CONCLUSION: PCR-RFLP analyses are able to identify positively five Bactrocera species, while B. papayae and B. carambolae might possibly be separated from one another, even if immature life stages or adult body parts are used. c 2009 Society of Chemical Industry Keywords: Bactrocera spp.; PCR-RFLP; species determination; COI 1 INTRODUCTION The Tephritid fruit flies are almost ubiquitous and found in all regions of the world. 1 Some of these fruit flies possess a great potential to cause damage to agricultural and horticultural production. 1 3 The medfly, Ceratitis capitata (Wiedemann), the olive fruit fly, Bactrocera oleae (Rossi), and the melon fly, Bactrocera cucurbitae Coquillett, are some of the well-known examples. Bactrocera, a genus found mostly in tropical Asia, Australia and the South Pacific regions, is known to be a major tropical fruit pest causing heavy losses in fruit and vegetable cultivation. In Malaysia, there are possibly at least a hundred Bactrocera species, of which only about half have been recorded. 4 Of these, the melon fly, B. cucurbitae, the papaya fruit fly, B. papayae Drew and Hancock, the carambola fruit fly, B. carambolae Drew and Hancock, the nangka fruit fly, B. umbrosa Fabricius, and the Malaysian fruit fly, B. latifrons Hendel, are major agricultural pests. In quarantine work, it is quite common to encounter immature life stages or body parts of suspected pests that require species determination. The current method of identification, based on morphological characters of adult insects, 5 may encounter difficulty in separating sibling species such as B. papayae and B. carambolae or identifying immature stages. In these cases, molecular determination would be an advantage. Mitochondrial DNA is used as a common marker for identification of species in fruit flies. 6 8 However, a previous study using polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) failed to separate between B. papayae and B. carambolae. 8 In recent studies, researchers have studied species identification based mostly on 16S and ITS sequences using PCR-RFLP Similarly, the mitochondrial control region has been used as a marker to detect genetic variations in B. dorsalis. 12 Molecular diagnostics has also been employed by insect researchers to infer phylogenetic relationships and for identification to species level In this study, cytochrome oxidase (COI) gene was chosen as a marker, as it has not been widely used for PCR-RFLP analyses. A 1.3 kb variable region of the COI gene from seven Bactrocera pest species found in Malaysia and neighbouring countries was sequenced. These belong to two subgenera: Bactrocera and Zeugodacus. They were Bactrocera (Bactrocera) papayae, B. (B.) carambolae, B. (B.) umbrosa, B. (B.) latifrons, Bactrocera (Zeugodacus) cucurbitae, B. (Z.) tau Walker and B. (Z) caudata Correspondence to: Tock H Chua, School of Science, University of Monash, Sunway Campus, Jalan Lagoon Selatan, Bandar Sunway, Selangor Darul Ehsan, Malaysia. chua.tock.hing@sci.monash.edu.my a School of Science, University of Monash, Sunway Campus, Jalan Lagoon Selatan, Bandar Sunway, Selangor Darul Ehsan, Malaysia b Consultant, Transforming Science Sdn Bhd, 23, Jalan Putra Mahkota 7/3A, Putra Heights, Subang Jaya, Selangor Darul Ehsan, Malaysia Pest Manag Sci (2009) c 2009 Society of Chemical Industry

2 TH Chua, YV Chong, SH Lim Table 1. List of specimens of Bactrocera spp. collected in and used in the study: their collection sites, preservation methods and number of individuals. The number of individuals sequenced is indicated in brackets Subgenus Species Collection site Preservation method Number of individuals Bactrocera B. (B.) papayae Kerinchi 20 C freezer 19 (3) Pantai Hillpark Kota Kinabalu Seremban 20 C freezer B. (B.) carambolae Kerinchi 20 C freezer 16 (2) Pantai Hillpark B. (B.) umbrosa Damansara Jaya 20 C freezer 4 (2) Pantai Hillpark B. (B.) latifrons Petaling Jaya 20 C freezer 10 (2) Lukut 20 C freezer Zeugodacus B. (Z.) caudata Lukut 2 (2) B. (Z.) cucurbitae Kerinchi 20 C freezer 8 (2) Lukut 20 C freezer B. (Z.) tau Tanah Rata 20 C freezer 6 (2) Kerichi Gombak Table 2. Oligonucleotides used for DNA amplification and sequencing of the Bactrocera spp Primer Sequence Reference Set 1 COIF 5 -TACAATTTATCGCCTAAACTTCAGCC-3 Han and Ro 16 COIR 5 -CATTTCAAGTTGTGTAAGCATC-3 Set 2 UEA7 5 -TACAGTTGGAATAGACGTTGATAC-3 Lunt et al.; 17 Jamnongluk et al. 15 UEA10 5 -TCCAATGCACTAATCTGCCATATTA-3 Fabricius. The restriction site variation across the COI gene for the seven species was analysed through PCR-RFLP. Digestion of the PCR products with restriction enzymes produced different banding profiles in the gel that could be used for identification of the Bactrocera species. 2 MATERIALS AND METHODS 2.1 Fruit fly collection and handling Adult specimens of the seven Bactrocera species were collected from different parts of Malaysia using chemical lures (methyl eugenol and cuelure) (Table 1), while the immature stages were collected from host fruits. The specimens were either stored at 20 C in a freezer or preserved in prior to analysis. A total of 65 individuals were used in this analysis. 2.2 DNA extraction, amplification and sequencing Total DNA was extracted using DNeasy Tissue Kit (Qiagen Inc., USA), following the manufacturer s protocol for animal tissue with slight modifications to increase DNA yield. Either a whole body or one leg was used for each extraction. Two sets of primers were used for polymerase chain reaction (PCR) amplification of cytochrome oxidase I (COI) markers (Table 2). PCR amplifications for the first set of primers (COIR and COIF) were performed following the conditions as described by Han and Ro: 16 an initial denaturation at 95 C for 3 min, 40 cycles at 93 C for 1 min, 55 C for 1 min, extension at 72 Cfor2minandfinal extension at 72 C for 15 min. The PCR conditions for the second set of primers (UEA7 and UEA10) 17 were as follows: 18 an initial denaturation at 94 C for 3 min, 35 cycles at 94 Cfor1min,50 C for 1 min, extension at 72 C for 1 min and final extension at 72 C for 30 min. PCR products were electrophoresed on a 1% SeaKem LE agarose gel. The DNA fragment sizes were estimated by comparing with 1 kb, 50 bp and 25 bp commercial markers (Promega, USA). The band corresponding to the target PCR product was excised and purified using Qiagen QIAquick Gel Extraction Kit. All samples were then sent to First BASE Laboratories Sdn. Bhd. (Malaysia) for sequencing. To test whether this molecular method can be used to identify a species when only a small body part or an immature stage is available, which may be the case in quarantine work, DNA was also extracted from different body parts of B. papayae (head, thorax and legs) and from immature stages (egg, larvae and pupae) of B. carambolae,b.papayae,b.latifrons and B.cucurbitae for PCR-RFLP analyses. 2.3 Restriction analysis The sequences of seven Bactrocera spp. amplified using both primer sets were aligned using ClustalW 19 andexaminedfortheir recognition sites of 50 restriction enzymes using the BioEdit program. 20 Three restriction enzymes were then chosen for further work, and their recognition sites are depicted in Fig. 1. The restriction fragment lengths of PCR-amplified section were c 2009 Society of Chemical Industry Pest Manag Sci (2009)

3 DeterminingBactrocera spp. using PCR-RFLP Figure 1. Recognition sites of chosen restriction enzymes predicted for seven Bactrocera spp. from COI fragments amplified using the two primer sets. The primers and restriction enzymes were respectively for: (A) COIR, COIF, AluI; (B) COIR, COIF, MseI; (C) UEA7, UEA1, RsaI. then predicted (Table 3). Polymerase chain reaction products obtained from each species were directly used for the PCR- RFLP analyses, without further purification. For B. papayae and B. carambolae, however, the PCR-RFLP analyses were also carried out with purified polymerase chain reaction products. A quantity of 5 8 µl of the PCR product was incubated in a total reaction volume of 20 µl containing 1 U restriction enzyme, MseI, AluI (NEW ENGLAND Biolabs) or RsaI (Promega Corporation) for 2 h (as per manufacturer s protocol) or overnight (for confirmation purposes) at 37 C. For further confirmation of the banding patterns of B. papayae and B. carambolae, restriction enzyme digestion for amplified fragments (UEA7/10) of these two species was repeated Pest Manag Sci (2009) c 2009 Society of Chemical Industry

4 TH Chua, YV Chong, SH Lim Table 3. Major restriction fragment lengths (longer than 80 bp) of PCR-amplified section of Bactrocera spp. treated with restriction enzymes (MseI, AluI and RsaI), as predicted using Bioedit COIF/COIR UEA7/UEA10 Species MseI AluI RsaI B. papayae 253, 217, 156, 114, , 201, 176, 138, B. carambolae 253, 217, 156, 114, , 201, 176, 138, B. cucurbitae 367, 219, 122, 106, , 186, 124, , 198 B. tau 367, 219, 217, 111, , 252, 163, , 297 B. umbrosa 367, 310, 171, 123, , 288, 249, 84 B. latifrons 321, 232, 210, 143, 114, , 249, 217, 177, , 262 B. caudata 418 using 5 U of RsaI and incubated overnight. After restriction enzyme digestion, the restricted DNA fragments were electrophoresed in 2 or 3% SeaKem LE agarose gel in 1 TAE buffer at 6 7 V cm 1 for 3 h, and the gel was visualised under ultraviolet radiation. Cloning was also carried out for two smaller fragments of B. papayae obtained from digestion by RsaI. The fragments were cloned into pjet1/blunt cloning vector (Fermentas) and transformed into Escherichia coli TOP10F by the heat-shock method.thetransformantswerescreenedbypcrusingthevectortargeted primers. Positive clones with expected PCR product size were selected for plasmid extraction and further confirmation by PCR. The desired fragments were sequenced from both directions with the vector-targeted primers, using the purified plasmid as template. 2.4 Data analysis Raw nucleotide data obtained were analysed through BLAST, and the differences within each species were aligned using ClustalW. The sequences obtained from both primers were aligned with sequence data of Bactrocerapreviously reported in NCBI. Accession numbers of sequences used for alignment and restriction fragment analyses were B. papayae (NC009770, AB192438, AB192436), B. carambolae (AB192420, EF014414), B. cucurbitae (DQ116245, AY , EU ), B. tau (EU , EU , AY ), B. umbrosa (DQ116346, DQ006867, AY ), B. latifrons (FJ009203, DQ116296, AY ) and B. caudata (AF423109). 3 RESULTS 3.1 Sequence alignment The first primer set (COIF and COIR) had successfully amplified a 1.3 kb long sequence of the COI gene only from Bactrocera carambolae, B. papayae, B. tau, B. latifrons, B. cucurbitae and B. umbrosa. However, using the second primer set (UEA7 and UEA10), a shorter sequence of COI gene was amplified successfully for all seven species (i.e. including B. caudata). An approximately 700 bp long nucleotide sequence was obtained using this primer set. This primer was also used successfully to amplify DNA fragments from dried specimens that had been kept for more than 6 years at room temperature. All the sequences from both primers were deposited in GenBank. The accession numbers for the sequences are as follow: B. carambolae (FJ903489, FJ903495), B. papayae (FJ903487, FJ903494), B. umbrosa (FJ903488, FJ903499), B. cucurbitae (FJ903491, FJ903497), B. latifrons (FJ903490, FJ903498), B. tau (FJ903492, FJ903496) and B. caudata (FJ903493). The DNA sequences of the seven Bactrocera spp. displayed an average 86% of sequence similarity, while the intraspecific similarity was 98 99%. Nucleotide composition of the sequences showed an average A + T content of 65%, which was biased towards A + T. When the sequences of the Bactrocera species studied here were aligned with those of the same species from other countries (Thailand, Indonesia, Japan, China, the USA, Sri Lanka and Kenya; accession numbers given in Section 2.4), no more than 20 mutations were found, and the similarity between these sequences was %. 3.2 PCR-RFLP analysis In this analysis, restriction fragments shorter than 80 bp were ignored, as they could not be clearly resolved by agarose gel electrophoresis. For the amplified DNA fragments obtained with the primers COIR and COIF and subjected to digestion by the three restriction enzymes (MseI, AluI and RsaI), distinct differences in the banding of the COI gene for four Bactrocera species were observed. Five distinct banding profiles were obtained from digestion with MseI or AluI (Figs 2A and B), enabling B.cucurbitae,B.tau,B.umbrosa and B. latifrons to be separated. With RsaI, only four banding profiles resulted, and only B. umbrosa and B. latifrons can be differentiated, as B. cucurbitae and B. tau had identical profiles. In the digestion with these three enzymes, both B. carambolae and B. papayae appear to have identical banding profiles and cannot be differentiated one from another. For the PCR products from UEA7 and UEA10 primers, the digestion with RsaI produced a specific banding for B. caudata that provides discrimination from the other Bactrocera spp. (Fig. 3). Interestingly, the digested PCR products (both unpurified and purified) from B. carambolae consistently produced one fragment, while that of B. papayae consistently yielded three fragments. This was true even with 5 U of RsaI and overnight or 1 h incubation time. Cloning of the two smaller fragments of B. papayae indicated the presence of SNPs in the sequence at base 326T > C and 328C > T. The resultant GTAC (from GTAT) is restricted by RsaI, resulting in two fragments with lengths of 357 and 224 bp. The banding profiles obtained from using various body parts of B. papayae with the PCR products digested with AluI or RsaI were the same as those from using the whole insect. Banding profiles identical to the adult profiles were also observed for the immature stages of the four species tested (B. carambolae, B. papayae, B. latifrons and B. cucurbitae). c 2009 Society of Chemical Industry Pest Manag Sci (2009)

5 DeterminingBactrocera spp. using PCR-RFLP (A) MseI (B) AluI (C) RsaI M M M M2 Figure 2. Banding profiles of COI (primers: COIF/COIR) digested with MseI, AluI and RsaI for the six Bactrocera spp. Lane 1, B. papayae;lane2,b. carambolae; lane 3, B. cucurbitae; lane 4, B. tau; lane 5, B. umbrosa; lane 6, B. latifrons; M1, 25 bp ladder; M2, 1 kb ladder. Figure 3. Banding profiles of COI (primers: UEA7/UEA10) digested with RsaI for the three Bactrocera spp. Lane 1, B. papayae;lane 2,B. carambolae; lane 3, B. caudata;m,50bpladder. These restriction markers seem reliable, as no intraspecific polymorphism was detected. 4 DISCUSSION This study demonstrates the use of PCR-RFLP markers for differentiating the major Malaysian Bactrocera pests on the basis of COI sequences. The molecular technique described here is useful for identifying Bactrocera spp. accurately, especially in quarantine laboratories situated at national borders or airports. Currently, identification by quarantine officers using morphological characters can be time consuming, and technically difficult for closely related species. Diagnostic morphological characters for immature stages are also lacking for most tephritids. 1 Muraji and Nakahara 7 also employed PCR-RFLP methods for identification of Bactrocera spp. using mtdna. They have analysed 18 Bactrocera spp. using 16S, 12S and trna sequences and successfully identified most of the Bactrocera spp. except for two closely related species, B. carambolae and B. papayae. Here, PCR-RFLP analyses have been performed on Bactrocera spp. including three species not included in the study by Muraji and Nakahara. 7 The present results showed that five of the seven species studied here could be positively separated one from another by a combination of COIF/R and UEA7/10 primers and three restriction enzymes. However, the restriction fragment of the 1.3 kb COI sequence (from the COI primer sets) could not differentiate B. carambolae fromb. papayae, which could be partly due to the interspecific crossing and backcrossing 11 that occur in the wild. Predicting restriction sites with Bioedit for amplified fragments (from UEA primers) of B. papayae and B. carambolae also showed similar digestion sites for both species. Nevertheless, experimentally, the banding pattern of Bactrocera papayae did not correspond to the digestion sites obtained in silico from BioEdit, in that two additional bands consistently appeared in the gel. Cloning and sequencing of these two smaller fragments (357 and 224 bp) indicated that they resulted from SNPs (single nucleotide polymorphs) in the B. papayae sequence at base 326T > Cand 328C > T and digestion by Rsa. It appears conceivable that the combination of UEA primers and RsaI restriction enzyme could possibly be used to differentiate between B. carambolae and B. papayae. The banding profiles of B. papayae using PCR products obtained from different body parts or from the whole adult insect and then digested either with AluI or RsaI were the same. This indicates the robustness of using molecular techniques for species identification, even when only broken pieces or parts of the insect are available. Fleming et al. 21 have also shown that parts of an insect could produce sufficient PCR product for RFLP analyses. More importantly, it has been shown in the present work that the immature stages of B. papayae and B. carambolae can be separated using PCR-RFLP analysis. This would be a useful tool in quarantine work, as these stages are often encountered and species determination is usually required without delay or without having to wait for adult emergence. As the identification by molecular techniques presented here was conducted on samples from Malaysia, the possibility exists that they might not be applicable to individuals from outside this geographical area. Nevertheless, as great similarity between the present alignment exercise of COI sequences from Malaysia and those from other countries has been found, it is likely that this method might also be useful for individuals of the same species from other countries. Whether this method is applicable to other Bactrocera species not tested here remains to be seen, but the authors believe that the banding profiles produced from combination of the primer sets and restriction enzymes could help in identifying at least five species, and possibly B. carambolae and B. papayae as well. ACKNOWLEDGEMENTS The authors thank the Ministry of Science and Technology for funding the research (E Science Fund: ) and Monash University for providing general facilities for conducting Pest Manag Sci (2009) c 2009 Society of Chemical Industry

6 TH Chua, YV Chong, SH Lim the research. They would like to thank their colleague Dr Song Beng Kah for helping in the analysis. They are also grateful to the anonymous reviewer, who gave several constructive suggestions. REFERENCES 1 White IM and Elson-Harris M, Fruit Flies of Economic Significance: Their Identification and Bionomics. CAB International, Wallingford, Oxon, UK (1992). 2 Drew RAI and Hancock DL, The Bactrocera dorsalis complex of fruit flies (Diptera: Tephritidae: Dacinae) in Asia. Bull Entomol Res Suppl 2:68 (1994). 3 Clarke AR, Armstrong KF, Carmichael AE, Milne JR, Raghu S, Roderick GK, et al, Invasive phytophagous pests arising through a recent tropical evolutionary radiation: the Bactrocera dorsalis complexoffruitflies. Ann Rev Entomol 50: (2005). 4 Chua TH, Annotated checklist of the Tribe Dacini, Subfamily Dacinae in Malaysia (Diptera: Tephritidae). Serangga 3:49 60 (1998). 5 White IMandHancock DL,CABIKEYtotheDacini(Diptera:Tephritidae)of Asian,PacificandAustralasiaRegions. CABInternational,Wallingford, Oxon, UK (1997). 6 Han HY and McPheron BA, Molecular phylogeny of Tephritidae (Insecta: Diptera) using partial sequences of mitochondrial ribosomal DNA. Mol Phylo Evol 7:17 32 (1997). 7 Muraji N and Nakahara S, Phylogenetic relationships among fruit flies, Bactrocera (Diptera: Tephritidae), based on the mitochondrial rdna sequences. Insect Mol Biol 10: (2001). 8 Muraji M and Nakahara S, Discrimination among pest species of Bactrocera (Diptera: Tephritidae) based on PCR RFLP of the mitochondrialdna. Appl Entomol Zool 37: (2002). 9 Salazar M, Theoduloz C, Vega A, Poblete F, González, Badilla R, et al, PCRRFLP identification of endemic Chilean species of Rhagoletis (Diptera: Tephritidae) attacking Solanaceae. Bull Entomol Res 92: (2002). 10 Muraji M, Kawasaki K, Shimizu T and Noda T, Discrimination among Japanese species of the Orius flower bugs (Heteroptera: Anthrocoridae) based on PCR-RFLP of the nuclear and mitochondrialdnas. JARQ 38:91 95 (2004). 11 Ebina T and Kenji O, Morphological characteristics and PCR-RFLP markers in the interspecific hybrids between Bactrocera carambolae and Bactrocera papaya of the B. dorsalis species complex (Diptera: Tephritidae). ResBullPlProtJap43:23 34 (2006). 12 Nakahara S, Kobashigawa Y and Muraji M, Genetic variations among and within populations of the Oriental fruit fly, Bactrocera dorsalis (Diptera: Tephritidae), detected by PCR-RFLP of the mitochondrial controlregion. Appl Entomol Zool 43: (2008). 13 Chua TH, Tan YL, Harikrishana J and Lim SH, Phylogenetic relationships of Malaysian Bactrocera species based on mitochondrial cytochrome oxidase I sequences and morphological characters (Tephritidae: Diptera). Malayan Nat J 61:34 47 (2009). 14 Smith JJ and Bush GL, Phylogeny of the Genus Rhagoletis (Diptera: Tephritidae) inferred from DNA sequences of mitochondrial cytochromeoxidaseii. Mol Phylo Evol 7:33 43 (1997). 15 Jamnongluk W, Baimai V and Kittayapong P, Molecular evolution of tephritid fruit flies in the genus Bactrocera based on the cytochrome oxidaseigene. Genetica 119:19 25 (2003). 16 Han HY and Ro KE, Molecular phylogeny of the superfamily Tephritoidea (Insecta: Diptera): new evidence from the mitochondrial12s, 16SandCOII. Mol Phylo Evol 34: (2005). 17 Lunt DH, Zhang DX, Szymura JM and Hewitt GM, The insect cytochrome oxidase I gene: evolutionary profiles and conserved primersforphylogeneticstudies. Insect Mol Biol 5: (1996). 18 Jamnongluk W, Baimai V and Kittayapong P, Molecular phylogeny of tephritid fruit flies in the Bactrocera tau complex using the mitochondrialcoisequences. Genome 46: (2003). 19 Thompson JD, Higgins DG and Gibson TJ, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrixchoice. Nucleic Acid Res 22: (1994). 20 Hall T, Bioedit, Version [Online]. Ibis Biosciences, Carlsbad, CA ( ). Available from bioedit.html [10 April 2008]. 21 Fleming CC, Rao J, Moreland B, Craig D and Turner SJ, Diagnostics of cyst nematodes and tephritid fruit flies using mitochondrial and ribosomal DNA. OEPP/EPPO Bull 30: (2000). c 2009 Society of Chemical Industry Pest Manag Sci (2009)