1、A high-throughput detection method for invasive fruit fly(Diptera: Tephritidae) species based on microfluidic dynamicarrayFAN JIANG,* WEI FU, ANTHONY R. CLARKE, MARK KURT SCHUTZE, AGUS SUSANTO,SHUIFANG ZHU andZHIHONG LI*College of Plant Protection, China Agricultural University, Beijing 100193, Chin
2、a, Institute of Plant Quarantine, ChineseAcademy of Inspection and Quarantine, Beijing 100176, China, School of Earth, Environmental and Biological Sciences,Queensland University of Technology (QUT), G.P.O. Box 2434, Brisbane 4000, Qld, Australia, Faculty of Agriculture,Padjadjaran University, Jatin
3、angor, 40600 West Java, IndonesiaAbstractInvasive species can be detrimental to a nations ecology, economy and human health. Rapid and accurate diagnos-tics are critical to limit the establishment and spread of exotic organisms. The increasing rate of biological invasionsrelative to the taxonomic ex
4、pertise available generates a demand for high-throughput, DNA-based diagnostics meth-ods for identification. We designed species-specific qPCR primer and probe combinations for 27 economically impor-tant tephritidae species in six genera (Anastrepha, Bactrocera, Carpomya, Ceratitis, Dacus and Rhagol
5、etis) based on935 COI DNA barcode haplotypes from 181 fruit fly species publically available in BOLD, and then tested the speci-ficity for each primer pair and probe through qPCR of 35 of those species. We then developed a standardization reac-tion system for detecting the 27 target species based on
6、 a microfluidic dynamic array and also applied the method toidentify unknown immature samples from port interceptions and field monitoring. This method led to a specific andsimultaneous detection for all 27 species in 7.5 h, using only 0.2 lL of reaction system in each reaction chamber. Theapproach
7、successfully discriminated among species within complexes that had genetic similarities of up to 98.48%,while it also identified all immature samples consistent with the subsequent results of morphological examinationof adults which were reared from larvae of cohorts from the same samples. We presen
8、t an accurate, rapid and high-throughput innovative approach for detecting fruit flies of quarantine concern. This is a new method which hasbroad potential to be one of international standards for plant quarantine and invasive species detection.Keywords: biological invasion, DNA barcodes, Fluidigm s
9、ystem, molecular identification, plant quarantine, species-specificReceived 29 January 2016; revision received 14 April 2016; accepted 6 May 2016IntroductionBiological invasions are increasingly common with theacceleration of global economic integration, and thisposes serious threats to internationa
10、l biosecurity, ecose-curity, economy and human health. Preventing the intro-duction of a potential invasive species is the first and themost effective way to minimize the harm caused by inva-sive organisms (Davis 2009). Strict quarantine protocolsand regulations are needed to accomplish this, sup-po
11、rted by diagnostic tools that can accurately identifyorganisms of biosecurity concern within a time framethat does not put undue restrictions on normal transportand trade. Quarantine delays of even 24 h for fresh com-modities may lead to loss of fruit and vegetable quality,and so, for the horticultu
12、ral sector, it is particularlyimportant that diagnostics are not only accurate, but alsovery rapid. In addition, speed and accuracy are alsoimportant for postborder detections to minimize oppor-tunity for establishment.Fruit flies (Diptera: Tephritidae) are one of the mosteconomically important pest
13、 groups in the world, caus-ing significant damage globally to fruit and vegetableproduction and international trade (White & Elson-Harris 1992). There are over 5000 described fruit flyspecies, of which approximately 350 are consideredeconomically important and belonging primarily to fivegenera: Anas
14、trepha Schiner, Bactrocera Macquart, CeratitisCorrespondence: Zhi-hong Li, Fax: +86-10-62733000; E-mail: and Shui-fang Zhu, Fax: +86-10-53897660;E-mail: 2016 John Wiley & Sons LtdMolecular Ecology Resources (2016) 16, 13781388 doi: 10.1111/1755-0998.12542Macleay, Dacus Fabricius and Rhagoletis Loew
15、(White &Elson-Harris 1992; Van Houdt et al. 2010). Tephritid fruitflies are known invasive organisms, with species such asthe Mediterranean fruit fly, Ceratitis capitata and Orientalfruit fly, Bactrocera dorsalis, being of truly global impor-tance as exotic invaders of quarantine concern (Whiteet al
16、. 2000; Duyck et al. 2004; Malacrida et al. 2007; Kha-mis et al. 2009; Papadopoulos et al. 2013).China is a globally significant importer of fresh pro-duce; hence, fruit flies are of particular concern. The verylarge trade of fresh commodities into China, the coun-trys range of climatic conditions w
17、hich make it suitablefor most exotic fruit fly species, and the economic andsocial importance of its domestic fresh-commodity pro-duction mean that exotic pest fruit flies pose a continualand high-profile threat to the country. Intercept informa-tion from Chinese ports in the last 10 years shows tha
18、tfruit flies have been intercepted from baggage and cargoon a staggering 19 000 occasions, with each interceptionconsisting of one or more individuals. These records aredominated by interceptions of Bactrocera species (97.7%),but there have also been 262 interceptions of Ceratitisspp, 43 interceptio
19、ns of Dacus species, 15 interceptions ofAnastrepha species and three interceptions of Rhagoletisspecies; at least two species from each of these generahave been intercepted. The number and diversity of spe-cies intercepted causes significant problems for quaran-tine officers at the ports as most ide
20、ntifications are basedon adult morphology; therefore, identification of inter-cepted fruit fly eggs, larvae or pupae involves rearingthrough to adults before a diagnosis is possible. This istime-consuming and sometimes unsuccessful (Arm-strong et al. 1997). The identification of adult intercep-tions
21、 based on morphology alone can also beproblematic if the adult insect is damaged or the speci-men belongs to a group of morphologically identical, ornear identical, sibling species which are common withinthe Tephritidae (Clarke & Schutze 2014).Molecular diagnostics represents an important steptoward
22、s rapid fruit fly identification without life-stagerestrictions. PCR or real-time PCR, based on species-spe-cific primers and/or probes, allows rapid identificationby overcoming the need for post-amplification digestion,as occurs in restriction fragment length polymorphism(RFLP) or DNA barcode seque
23、ncing techniques (whichcan take between several hours to days) (Yu et al. 2004,2005; Barr et al. 2006, 2012; Chua et al. 2010; Jiang et al.2013, 2014a). However, PCR with species-specific mark-ers is generally only used for single species, or species-pair discrimination, while multiplex PCR has, at
24、most,been successfully applied to the amplification of sevenspecies in a single reaction (Hosseini et al. 2007; Sint et al.2012).Due to the increasing number and diversity of inter-cepted fruit flies, a high-throughput method combinedwith high accuracy, efficiency and reagent-saving hasbecome a nece
25、ssity not only for China, but all major trad-ing countries. The BioMark system from Fluidigm com-pany (https:/ is a high-throughput PCR platform based onmicrofluidic dynamic arrays (Unger et al. 2000) and iscomposed of the BioMarkTM real-time PCR system, Inte-grated Fluidic Circuit (IFC) chip, IFC c
26、ontroller and analy-sis software. Available chips can perform as many as9216 reactions in a single run (Liu et al. 2003). Whenusing this system, the assays and samples are individu-ally pipetted into corresponding inlets and loaded andmixed in the chip, and then, the chip is carried out as aregular
27、PCR cycling step such that each assay can reactwith each sample. To date, the microfluidic dynamicarray process has been used in a number of novel appli-cations, such as medical diagnosis, gene expression,genotyping and GMO analysis (Spurgeon et al. 2008;Wang et al. 2009; Yung et al. 2009; Brod et a
28、l. 2014). Thistechnique has not been applied for species detection inplant quarantine and invasion biology, where it offersgreat potential for high-throughput and high-speciesdiversity screening.The objective of this study was, therefore, (i) todevelop a standardized high-throughput diagnostic sys-t
29、em for the detection of invasive fruit fly species usingreal-time PCR in a microfluidic dynamic array and (ii) totest the applicability of the system to detect unknownimmature fruit fly samples collected from interceptionsin ports and field monitoring.Materials and methodsSample collectionWe collect
30、ed 35 fruit fly species from six genera basedlargely on port interceptions, from 54 geographical pop-ulations from 16 countries across six continents (AsiaChina, Laos, Myanmar, Thailand, Vietnam, Malaysia,Indonesia; Australia; Africa Kenya, Burundi, SouthAfrica; Europe Italy; North America Mexico, G
31、uate-mala, Belize; and South America Suriname) (Table S1,Supporting information). All samples were preserved in100% ethanol and stored at C020 C in the Plant Quaran-tine and Invasion Biology Laboratory of China Agricul-tural University (CAUPQL). Each adult specimen wasidentified based on available t
32、axonomic keys, and diag-nostic images were captured using Optical ConfocalMicroscopy Imaging System (UV-CTS) prior to molecu-lar experiments (see figures of each species in Fig. S1,Supporting information) (White & Elson-Harris 1992; 2016 John Wiley & Sons LtdHIGH-THROUGHPUT DETECTION FOR FRUIT FLIES
33、 1379Liang et al. 1996; Wang 1996). A subset of these fliesbecame our test organisms.DNA extraction and detectionThree individuals from each geographical population foreach species were used for DNA extraction and subse-quent primer and probe selection in a microfluidicdynamic array for species dete
34、ction. Total genomic DNAwas extracted from a single leg of individual adultsusing the TIANamp Genomic DNA Kit (DP304, TIAN-GEN) following the manufacturers protocol for animaltissue with slight modifications: each leg was digestedby 15 lL of proteinase K and the template DNA dis-solved in 50 lL of T
35、E buffer. DNA concentrations wereestimated by spectrophotometry (UV-Vis spectropho-tometer Q5000, QUAWELL). The rest of each body andthe template DNA were stored at C020 C for morpholog-ical and molecular verification in CAUPQL.Primer and probe design and specificity testWe focused on the following
36、27 economically importantfruit flies and closely related species from our 35 freshlycollected species: A. ludens, A. obliqua, B. (B.) albistrigata,B. (B.) correcta, B. (B.) carambolae, B. (B.) dorsalis, B. (B.) lat-ifrons, B. (B.) rubigina, B. (B.) tryoni, B. (B.) tuberculata, B.(B.) umbrosa, B. (B.
37、) zonata, B. (D.) oleae, B. (T.) minax, B.(T.) tsuneonis, B. (Z.) bezziana, B. (Z.) cilifera, B. (Z.) cucur-bitae, B. (Z.) scutellata, B. (Z.) tau, C. vesuviana, C. capitata,C. cosyra, C. rosa, D. bivittatus, R. cerasi and R. pomonella.Our choice of specimens was based on the Chinesenational lists o
38、f important pests of plant quarantine andhistorical intercept data from Chinese ports. Species-specific primers and probes were developed for these27 species based on COI barcodes from species within fiveeconomically significant genera (Anastrepha, Bactrocera,Ceratitis, Dacus and Rhagoletis,alsoincl
39、udingC. vesuviana)publically available in the Barcode of Life Database (BOLD,http:/www.boldsystems.org, Ratnasingham & Hebert2007); some of the sequences in BOLD were from our previ-ous work (Jiang et al. 2014b). Each sequence was checkedbefore sequence alignment to ensure it was correct accord-ing
40、to published criteria (Jiang et al. 2014b), and then usedDNASP 5.10 software (Librado & Rozas 2009) to generatehaplotypes for each species.The sequence alignment of 935 COI haplotypes from181 fruit fly species from six genera (Table S2, Support-ing information) was performed using BIOEDIT 7.0.0 soft
41、-ware (Hall 2004). Sites with intraspecific conservationand interspecific variation were manually identified. Foreach species, species-specific primers were manuallydesigned to cover species-specific sites. We placed thespecies-specific sites in the 30-end of the forward andreverse primer to ensure
42、the specificity of each species-specific primer pair. Oligonucleotide sequences werechecked using OLIGO 7.0 software (Rychlik 2009). Criteriafor evaluating primers were as follow: (i) the absolutevalue of 30MG cannot be higher than nine or three consec-utive G/C at 30-end should be avoided; (ii) the
43、 dimershould be unstable and the absolute value of MG cannotbe higher than 10; (iii) the absolute value of MG of hair-pin structure cannot be higher than 4.5; (iv) GC contentsshould be 3070%, and the difference of Tm between for-ward and reverse primer should not be very great usu-ally less than 5 C
44、; (v) the false priming efficiency shouldbe lower than 100; and (vi) the length of primer shouldbe 1830 bases. Primers were synthesized by InvitrogenTrading (Shanghai) Co., Ltd.DNA extracted from freshly collected 35 fruit fly spe-cies was used by performing PCR assays to select thespecies-specific
45、primers. The principle for primer selec-tion was that each species-specific primer could onlyamplify its target species, and no bands were evident forany of the other species tested through agarose gel elec-trophoresis. PCR amplification in a total reaction volumeof 25 lL was performed using the fol
46、lowing compo-nents: 12.5 lLof29 Taq PCR MasterMix (KT201, TIAN-GEN); 0.5 lL of forward and reverse primer (0.01 mM);1 lL of template DNA; and 10.5 lL of ddH2O. The PCRcycling conditions were an initial denaturation at 95 Cfor 3 min, followed by 30 cycles of denaturation at 95 Cfor 15 s, annealing an
47、d extension at 60 C for 1 min; anda final extension at 60 C for 1 min. The PCR products(5 lL) were analysed on a 2% agarose gel in 1 9 TAEbuffer, and then visualized under UV light after ethid-ium bromide staining. In addition, each positive bandproduct was sequenced separately using each PCR pri-me
48、r by Beijing Aoke Biotechnology Co. Ltd. to confirmthe amplified products generated by the species-specificprimers; results were compared with known sequencesby searching the NCBI or BOLD database.TaqMan MGB probes were designed between the spe-cies-specific primer pairs which had been selected anda
49、lso covered by species-specific sites. PRIMER EXPRESS 3.0software (Applied Biosystems, 2005) was used for probedesign. The principles referenced for TaqMan MGBprobe design were as follows: (i) 50-end cannot be a Gresidue; (ii) three or more G in the first four bases at 30-end should be avoided; (iii) the shorter distance betweenprobe and primer the better; (iv) GC% should be 3070%and Tm should be 6567 C; (v) probe length should be1325 bases; (vi) repeating oligonucleotides should beavoided, especially for four or more G residues and sixconsecutive A residues anywhere in the probe; and (vii)pr
