57

7 Identification

Overview

The National Handbook for the Identification of Fruit Flies in Australia (overview presented in Figure 3 and Figure 4) proposes that primary identification is undertaken using conventional taxonomy with the support of molecular genetic techniques for some species or immature stages. The diagnostic methods available for each species are presented in Table 3 and covered in greater detail in sections 7.2 (Morphological), 7.3 (Molecular) and Section 8 (data sheets with the specific morphological and molecular diagnostic information for each species). These techniques are currently in use in Australia and form the basis of this national protocol.

Tephritid fruit fly adult specimens are primarily identified through an examination of diagnostic morphological characters. Other life stages are more problematic, with only third instar larvae (and sometimes pupae) of some species usually identified through visual examination (e.g. White & Elson-Harris 1992). Identification of earlier life stages (early instars, eggs), and adults of morphologically ambiguous species, generally requires the use of molecular techniques.

Molecular techniques are best used to support or augment morphological identification. They can be used to identify early larval stages (which are hard to identify reliably on morphological features) and eggs. They also can be used for incomplete adults that may be missing specific anatomical features required for morphological keys, or specimens that have not fully developed their features (especially colour patterns). It should be recognised, however, that the success of a molecular diagnosis can be impacted by factors such as life stage, specimen quality or any delays in processing. As a result, the suitability of each method has been identified (See Section 7).

The molecular protocols require a laboratory to be set up for molecular diagnostics, but can be conducted by almost any laboratory so equipped. Access to on-line published sequences is required for the DNA barcoding protocol.

Most molecular techniques presented in this standard involve the amplification of particular region(s) of the fly genome using a polymerase chain reaction (PCR), while the other technique covered (allozymes) examines protein variation. The PCR target is either the mitochondrial gene for cytochrome oxidase subunit I (COI), known as the DNA barcode region, or a region of the ribosomal RNA operon, either just the first internal transcribed spacer (ITS1) or part of the 18S subunit plus the ITS1.

DNA barcoding is now available in this manual as an alternative to the original ITS-based techniques. This technology, in contrast, incorporates population variation and has an international, independently growing reference dataset. However, difficulties can still arise among closely related species within species complexes.

For the ITS1, the size of the PCR amplicon is useful for identification of a few species. However, restriction digestion of the ITS1 PCR amplicon, which denotes the actual sequence in defined regions of the amplicon, is recommended for all analyses as a more robust method of identification. This is referred to as restriction fragment length polymorphisms (RFLP) analysis. Reference data has been developed for the economically important species. However, the RFLP does not necessarily eliminate non-economic fruit flies for which reference data have not been developed.

This national protocol is presented on the premise that most species can generally be resolved using traditional taxonomy without ambiguity. The molecular methods described here are recommended for use alongside morphological methods where there is any ambiguity; although they may be equally ambiguous in the case of species complexes. They can be used on their own for morphologically cryptic immature life stages, or when only part of a specimen is available, or when access to the appropriate taxonomic expertise is lacking.

58

Figure 3. Overview of fruit fly diagnostic procedures (adult specimens).

59

Figure 4. Overview of fruit fly diagnostic procedures (larval specimens).

60

Table 3. Diagnostic methods used to identify fruit fly species

Scientific name Morphological description

PCR-RFLP1

PCR-RFLP 2

PCR- DNA Barcoding4

Allozyme electrophoresis

qPCR11

Anastrepha fraterculus 49/141e*

Anastrepha distincta 0/27e*

Anastrepha grandis 3/35

Anastrepha ludens b 68/174e*

Anastrepha obliqua 52/133e*

Anastrepha serpentina 8/52*

Anastrepha striata 18/59

Anastrepha suspensa b 5/76e*

Bactrocera albistrigata 24/24

Bactrocera aquilonis a

c 3/38f

Bactrocera atrisetosa

Bactrocera bryoniae 1/11

Bactrocera carambolae 110/111

Bactrocera caryeae 1/1g

Bactrocera correcta 54/54

Bactrocera cucumis 2/11

Bactrocera cucurbitae 244/289*

Bactrocera curvipennis 2/2

Bactrocera decipiens

Bactrocera depressa

Bactrocera dorsalis s.s. 12 >1000/1000g

Bactrocera facialis 1/1

Bactrocera frauenfeldi d 2/16

Bactrocera jarvisi 3/7

Bactrocera kandiensis 17/17g

Bactrocera kirki13 d,3 5/5

Bactrocera kraussi 3/3

11 Real-time PCR 12 Includes B. papayae, philippinensis and invadens since their synonymisation with dorsalis sensu stricto as per Schutze et al. (2015). This is a taxonomic update of the previous list as opposed to an omission of these species. 13 B. kirki can be distinguished from B. frauenfeldi and B. trilineola using an additional enzyme Acc I

61

Scientific name Morphological description

PCR-RFLP1

PCR-RFLP 2

PCR- DNA Barcoding4

Allozyme electrophoresis

qPCR11

Bactrocera latifrons 82/86

Bactrocera melanotus 3/3

Bactrocera minax 58/58

Bactrocera musae14 21/255

Bactrocera neohumeralis a

c 2/4f

Bactrocera obliqua

Bactrocera occipitalis 31/34g

Bactrocera oleae 109/114

Bactrocera passiflorae 0/1

Bactrocera psidii 2/2

Bactrocera pyrifoliae

Bactrocera tau 85/100*

Bactrocera trilineola d 21/21

Bactrocera trivialis 3/3

Bactrocera tryoni a

c 48/55f

Bactrocera tsuneonis 21/21

Bactrocera tuberculata 5/5

Bactrocera umbrosa 39/39

Bactrocera xanthodes 6/7

Bactrocera zonata 36/61

Ceratitis capitata 208/292

Ceratitis rosa 46/50*

Dacus longicornis 2/2

Dacus solomonensis

Dirioxa pornia 3/3

14 The 17 species within the B. musae complex (Drew et al. 2011) have not been tested by the RFLP method, but DNA barcodes will distinguish B. musae s.s. from at least the non-pest species B. conterminal, B. prolixa, B. rufivitta, and B. tinomiscii; adults of all species can be distinguished morphologically, see section 7.2.

62

Scientific name Morphological description

PCR-RFLP1

PCR-RFLP 2

PCR- DNA Barcoding4

Allozyme electrophoresis

qPCR11

Rhagoletis cingulata 24/29

Rhagoletis completa 21/21*

Rhagoletis fausta 1/1

Rhagoletis indifferens

Rhagoletis mendax 2/2h

Rhagoletis cerasi 17/29

Rhagoletis pomonella 23/24h

1 Species with the same superscript letter cannot be distinguished from each other with this test. 2 This test is best used when the unknown samples may be suspected as one for which the PCR-RFLP test 1 has not been developed; all RFLP patterns available in Armstrong & Cameron (1998) are listed in Appendices 3 and 4, and includes other species not listed in this table, A. bistrigata, A. sorocula, A. zenildae and B. quadriestosa.

3 Numbers in brackets (x/y) refer to the x number of individuals of that species with publically accessible DNA (COI) barcodes on the Barcode of Life (BOLD) website and y number of total sequences available if using the BOLD identification engine; the difference represents private barcodes unavailable for download (www.boldsystems.org/views/taxbrowser.php?taxid=439) as of 18th September 2015. Where there are no specific DNA barcodes available, others for other species within the genus could be used to at least determine genus-level identification; as of 18th September 2015, there are 203 species of Bactrocera, 66 of Ceratitis, 82 of Dacus, 28 species of Rhagoletis and 1 of Toxotrypana that have barcodes available. Species with the same superscript letter cannot be distinguished from each other with DNA barcodes, or can only be identified to the level of a species complex. Species with * are not well resolved amongst a number of other species in BOLD, although low divergence clades may be apparent. Used in conjunction with other information about the specimen this may still be sufficient to make a confident identification. Complimentary identification by ITS1 PCR-RFLP could also be considered.

63

Morphological identification

Approximately 90% of the dacine pest species can be identified accurately, and quickly, by microscopic examination of the adult. For these species there is no need for supporting evidence. The remaining 10% (mainly some dorsalis complex species) can be identified with this same method but require expert examination and may require additional supporting evidence such as the molecular diagnosis or host association records. Methods here are updated according to Drew & Romig (2013).

Only morphological diagnostic procedures and information for adult fruit flies are contained in this document. Aside from molecular techniques, larval diagnosis has been excluded from this protocol.

For routine morphological identification: Collect flies in dry traps. When clearing traps collect samples into a tissue. Put tissue in a small box with collection

details on the outside. Samples in tissue can also be collected into a vial though this is less preferred particularly in the tropics as samples can sweat causing specimens to deteriorate. If collecting into vial a pencil or permanent pen label (not biro as it runs) should be put inside as writing on the outside can rub off.

Store samples in freezer until ready for identification. Sort dry specimens in a petri dish under a binocular microscope. If keeping specimens after identification store in freezer to prevent deterioration. Do not store specimens in ethanol/alcohol/propylene glycol unless being kept for DNA

analysis. They leach diagnostic colours and patterns necessary for morphological identification.

For suspect specimens requiring further identification Store the specimen in a small vial with tissue to protect it until ready to pin or ship. Add a

pencil or permanent ink label (not biro as it runs) detailing collection location, collection date, collection method, collector, tentative identification and identifier.

As manipulating loose specimens with forceps tends to damage them, suspect specimens should ideally be pinned to keep them as intact as possible. If the specimen is a suspect exotic and needs to be shipped to a specialist ASAP they can be sent unpinned in a tube as above.

For pinning: Put specimen in relaxing chamber with thymol (to prevent mould growth) for 6-12 hours. Using a micropin, pin through RHS of scutum. Mount micropinned specimen on a pith stage

on a pin. Add label to pin. Store pinned specimens in reference collection conditions i.e. 21OC and 50% RH.

SAMPLE PREPARATION

7.2.1.1 Procedure

The following apparatus and procedures should be used to prepare the specimen for identification (adapted from QDPIF 2002):

Apparatus:

Stereoscopic microscope or Stereomicroscope with magnification range of 7X to 35X.

Light source

90mm diameter petri dishes

64

Forceps (Inox #4)

Preparation procedure: 1) Ensure the workstation is clean and clear of all flies before commencing. 2) Adjust chair height and microscope, and turn on the light source (refer to specific operating

procedures for the microscope in use). 3) If applicable, record the lure and trap type or host material in which the specimen was found. 4) Carefully place the fruit fly into a plastic petri dish. If examining more than one fly at once

ensure there is a single layer of flies only, with room to move flies from one side of the dish to the other.

5) While looking through the microscope check each fly individually. Manipulate them with the forceps so that diagnostic features are visible.

7.2.2 Identification

Key features (Figure 5, Figure 6, Figure 7 and Figure 8) used for the morphological diagnosis of adult fruit flies include:

Overall colour and colour patterning

Wing morphology and infuscation

Presence, shape and colour of thoracic vittae. A vitta is a band or stripe of colour.

Presence or absence of various setae, and relative setal size (for Trypetinae). (Note: chaetotaxy, the practice of setal taxonomy, is not as important for Dacinae, i.e Bactrocera and Dacus)

Use the morphological diagnostic key and descriptions contained in Section 8 to identify the species of fruit fly under microscopic examination.

If identification cannot be made using this diagnostic procedure and/or the specimen is suspected to be of quarantine concern, it should be referred to either a State or National authority (see section 9.1 Key contacts and facilities). If the specimen is identified as an exotic fruit fly, it should be referred to a National Authority within 24 hours and the appropriate National Authority notified as required in PLANTPLAN.

65

Figure 5. Adult morphology; head (top) and wing (bottom) (White and Elson-Harris 1992).

ar – arista comp eye – compound eye fc – face flgm 1 – 1st flagellomere fr – frons fr s – frontal setae gn – gena (plural: genae)

gn grv – genal groove g ns – genal seta i vt s – inner vertical seta lun – lunule oc – ocellus oc s – ocellar seta o vt s – outer vertical seta orb s – orbital setae

pafc – arafacial area ped – pedicel poc s – postocellar seta pocl s – postcular setae ptil fis – ptilinal fissure scp – scape vrt – vertex

66

Figure 6. Adult morphology, Thorax; Dorsal features (White and Elson-Harris 1992).

a npl s – anterior notopleural seta a sctl s – apical scutellar seta a spal s – anterior supra-alar seta a spr – anterior spiracle anatg – anatergite anepm – anepimeron anepst – anepisternum anepst s – upper anepisternal seta b sctl s – basal scutellar seta cx – coax dc s – dorsocentral seta

hlt – halter or haltere ial s – intra-alar seta kepst – katepisternum kepst s – katepisternal seta ktg – katatergite npl – notopleuron p npl s – posterior notopleural seta p spal s – posterior supra-alar seta p spr – posterior spiracle pprn lb – postpronotal lobe pprn s – postporontal seta

prepst – propisternum presut area – presutural area presut spal s – preutural supra-alar seta psctl acr s – prescutellar acrostichal seta psut sct – postcutural scutum sbsctl – subscutellum scape – scapula setae sctl – scutellum trn sut – transverse scuture

67

Figure 7. Adult morphology, thorax; lateral features (White and Elson-Harris 1992).

See Figure 5 for abbreviations.

Figure 8. Adult morphology, abdomen; male with features of typical dacini (left), Female, with extended ovipositor (right) (White and Elson-Harris 1992).

acul – aculeus ev ovp sh – eversible ovipositer sheath

ovsc – oviscape st – sternites numbered 1-5 in the male and 1-6 in the female

tg – tergites where 1+2 are fused to form syntergosternite 1+2, followed by tergites 3-5 in the male and 3-6 in the female

68

Molecular identification

Three methods are presented here, all of which can use a common sample storage and handling technique (Section 6), DNA extraction method (Section 7.3.1), and are based on PCR (Polymerase Chain Reaction) analysis.

Choice of method: The DNA barcoding method (Section 7.3.2) is generally recommended over the PCR-RFLP (restriction fragment length polymorphism) methods (Section 7.3.3) because:

(a) DNA barcoding can produce better resolution between species as it utilises variation in the

complete sequence amplified. PCR-RFLP is limited to variation at just a few or several 4-6 bp restriction sites within the amplicon, the suite of which are dependent on the nature of the restriction enzymes used

(b) DNA barcoding uses a very large reference sequence database. This is international, publicly accessible and constantly being added-to by unrelated institutions and projects. Consequently there is greater inclusion of taxonomically comparative species and population data to improve confidence in identification. PCR-RFLP generally relies on in-house developed reference restriction patterns; therefore comparative species and population data are incorporated at a significantly reduced rate without contributions from independent and international laboratories

(c) DNA barcoding is quantifiable and accessible to bioinformatics analyses. PCR-RFLP is essentially qualitative, relying on visual inspection against control samples and molecular weight markers.

If access to a laboratory with DNA sequencing equipment is difficult, PCR-RFLP is a useful alternative for the majority of species. There are also some species for which identification is ambiguous with DNA barcoding but not for PCR-RFLP (Table 3); however, while this may be a function of the different gene regions used, it may also be a result of the many more species and populations included in the DNA barcode database covering more of the variation.

Gene regions used: DNA barcoding utilises a mitochondrial locus within the cytochrome oxidase I (COI) gene, while the RFLP methods both utilise a ribosomal DNA (rDNA) gene region that includes the first internal transcribed spacer (ITS1). Both gene regions have been chosen for the suitability of their sequences to be distinct between species (Jinbo et al 2011; Wang 2015).

Preparation of general reagents is provided in Appendix 2.

69

7.3.1 DNA extraction for adults and immature stages

The preferred method presented here uses the DNeasy® Blood and Tissue kit (Qiagen). This produces DNA of very high quality suitable for archiving, should any future work be needed. Other cheaper and more rapid methods, such as prepGEM (ZyGEM, AKL, NZ) and ZR Tissue & Insect DNA MicroPrep (Zymo Research, CA, USA) kits, are also highly suitable for diagnostic purposes, but the DNA is of poorer quality and not so well suited to long-term archiving.

DNA should ideally be obtained using relatively non-destructive techniques, to ensure that a voucher specimen is available for future morphological examination (Floyd et al. 2010). For fruit fly adults a leg or the head of a specimen can be used, therefore retaining many other valuable morphological features of the specimen (such as wings, thorax and abdomen). For larvae, anterior and posterior sections can be sectioned off and retained, preserving the morphologically valuable mouthparts and spiracles. Alternative even less destructive methods, involving Proteinase K digestion of internal tissues, are also very effective (e.g. Gilbert et al. 2007).

DNA should be extracted in a room or a bench separated from PCR set up, and from post-PCR product analysis, i.e. electrophoresis, sequencing or RFLP tests (Sections 7.3.4 and 7.3.3).

Suitable DNA can be obtained from dry, frozen or ethanol / propylene glycol preserved specimens, (see Section 6.4.1).

Equipment and/or material needed • Blotting paper or Kimwipe tissues

• Scalpel blades (if sub-sampling each sample)

• Benchtop Microcentrifuge

• Microcentrifuge tubes (1.5 mL), (certified DNA free, or autoclaved)

• QIAGEN extraction kit (DNeasy® Blood and Tissue kit)

• Heating block or water bath to 56°C

• Forceps

• Sterile micro pestle

• Vortexer

• Bead-Mill with 3 mm solid glass beads, or plastic micropestle and matching microcentrifuge tube, (certified DNA free, or autoclaved)

• QIAGEN DNeasy® Blood and Tissue Kit Handbook, July 2006 (for reference if required)

METHOD

The DNA is extracted by following the manufacturer’s instruction in the DNeasy Blood and Tissue kit handbook (July 2006) for animal tissue (spin column). Slight modifications are highlighted here, with additional notes, for DNA used in the real-time or quantitative PCR (qPCR) assay.

Prior to starting the extraction: Check if the ATL and AL buffers have precipitated during storage and, if so, warm them until

the precipitate has fully dissolved. Ensure ethanol has been added to the AW1 and AW2 buffers as specified on the reagent

bottles. Heat the heating block or water bath to 56°C Pipette 100 µl x (n+1) of AE buffer into a clean 1.5 mL Eppendorf tube, and place the tube in

the 56°C heating block. (n=the number of samples to be extracted). Eluting the DNA with pre-

70

warmed AE buffer can increase DNA yield and this is recommended for use with the real-time PCR protocol.

Starting the extraction:

1) Allow several hours for processing (Proteinase K digest time dependent).

2) If using a “Bead-Mill” to homogenise the tissue add two 3 mm solid glass beads (acid-washed in 10 % HCl prior to use) to a clean microcentrifuge tube and add 20 µL of proteinase K. If using a micro pestle to homogenise the tissue, add 20 µL of Proteinase K directly to the paired microfuge tube.

3) Remove samples (e.g. larva, adult head or leg) from ethanol and dry on blotting “tissue” until ethanol evaporates (approximately 1 min). If samples were “dry” frozen or air dry omit this step. Eggs can be used, but usually require multiple eggs (>5) to be processed together to obtain enough DNA (K. Armstrong pers. comm.). Note: 1-2 eggs are sufficient for use in the real-time PCR assay.

4) Add sample to the Proteinase K tube, cleaning forceps (ethanol wipe) in between samples to prevent cross-contamination. Homogenise the specimen, either in a Bead-Mill (1 min @ 30 MHz) or with plastic micro pestle in a microfuge tube.

5) Quick-spin in centrifuge (up to 10,000 x g / rpm).

6) If processing large numbers of specimens, rotate previously “outer” samples to “inner” position of the Bead-Mill (i.e. swap the inner microcentrifuge tube insert around). Repeat Bead-Mill shaking (1 min @ 30 MHz).

7) Repeat Bead-Mill and centrifuge steps until samples contain no large visible fragments. If using a plastic pestle, initially grind until this point is reached.

8) Add 180 µL of Buffer ATL and vortex. Note: the tissue can alternatively be homogenised in 180 µl of ATL buffer plus the 20 µl of proteinase K together.

9) Incubate for 0.5-1 ½ h at 56oC, or possibly overnight for complete digestion. For empty pupal cases, or aged and dried samples, an overnight incubation is recommended.

10) Vortex, quick-spin to remove liquid from inner lid of microcentrifuge tubes.

11) Add 200 µL of Buffer AL and 200 µL of ethanol (Buffer AL and ethanol can be premixed in a large tube for multiple samples, and then dispensed [400 µL] to each sample). Vortex.

12) Pipette mixture to QIAGEN kit column to bind DNA and centrifuge at ~6000 x g for 1 min. Discard lower collection tube.

13) Place column into a new collection tube. Add 500 µL of Buffer AW1 to wash. Centrifuge at ~6,000 x g for 1 min. Discard lower collection tube.

14) Place column into a new collection tube. Add 500 µL of Buffer AW2 to wash. Centrifuge at ~20,000 x g (17,000 x g is acceptable) for 3 min. Discard lower collection tube (making sure no AW2 Buffer splashes onto the base of column).

15) Label the top and side of clean 1.5 mL microcentrifuge tube with the sample’s identification / database number.

16) Place column into the new microcentrifuge tube. Add 100 µL of AE buffer (this buffer must come into direct contact with the column filter). Incubate at room temperature for 1 min, then centrifuge for 1 min at 6000 x g to elute the DNA from the column. Note: 50 µl of pre-warmed AE buffer should be used for DNA to be used in the real-time PCR protocol.

17) Repeat elution a second time.

18) Discard column and retain microcentrifuge tube containing ~200 µL (or ~100 µL) of DNA in AE Buffer.

20) Store DNA in -20°C or -80oC freezer. Ideally, DNA can be aliquoted for long-term storage.

71

7.3.2 DNA Barcoding

This test was developed by Mark Blacket, Linda Semeraro and Mali Malipatil, Victorian Department of Economic Development, Jobs, Transport and Resources (Blacket et al. 2012). This section (7.3.2) was written by Mark Blacket and Karen Armstrong.

INTRODUCTION

Tephritid fruit fly adult specimens are primarily identified through an examination of diagnostic morphological characters (Table 2). Other life stages are more problematic, with only third instar larvae (and sometimes pupae) of some species usually identified through visual examination (e.g. White & Elson-Harris 1992). Identification of earlier life stages (early instars, eggs), and adults of morphologically ambiguous species, generally requires the use of molecular techniques.

DNA barcoding is a diagnostic molecular method that is routinely applied at the DEDJTR (Vic.) and Ministry for Primary Industries (MPI) laboratories (NZ) to identify morphologically problematic specimens and confirm new fruit fly incursions. The Australian Department of Agriculture and Water Resources (DAWR) also regularly uses it to screen immature stages (e.g. eggs) collected from fruit to exclude the non-tephritid exotic pest Drosophila suzukii (A. Broadley pers. comm.). DNA barcoding of an unknown insect specimen involves obtaining a DNA sequence of a specific region of the mitochondrial Cytochrome Oxidase I (COI) gene, and then comparing it with a database of sequences from positively identified reference specimens.

There are currently many reference DNA barcoding sequences available. The Barcode of Life Data Systems (BOLD) (http://www.boldsystems.org, accessed June 2015) holds 833 x Tephritidae, including 198 x Bactrocera “species with barcodes”. This also includes data from NCBI GenBank and of regional studies covering a broad range of endemic or intercepted tephritid species – e.g. 60 species from Oceania (Armstrong and Ball 2005), 153 species from Africa (Virgilio et al. 2012), 135 species from Europe (Smit et al. 2013). However, there is no single peer-reviewed published DNA barcoding laboratory protocol that covers all of the targeted tephritid species listed in Table 3.

DNA barcoding should ideally obtain DNA using relatively non-destructive techniques, to ensure that a voucher specimen is available for future morphological examination (Floyd et al. 2010). For fruit fly adults a leg or the head of a specimen can be used, therefore retaining many other valuable morphological features of the specimen (such as wings, thorax and abdomen). For larvae, anterior and posterior sections can be sectioned off and retained, preserving the morphologically valuable mouthparts and spiracles. Alternative even less destructive methods, involving Proteinase K digestion of internal tissues, are also very effective (e.g. Gilbert et al. 2007).

There are a number of potential issues that should be taken into consideration when applying a DNA barcoding approach to fruit fly species identification:

1) Many studies have not used the same gene region rendering the data unusable (Boykin et al. 2012);

2) Some commonly applied combinations of standard polymerase chain reaction (PCR) primers (e.g. LCO/HCO) may amplify a nuclear copy (a numt pseudogene) of the COI barcode region in some fruit flies (Blacket et al. 2012);

3) The presence of fruit fly species complexes can limit precise species identification (e.g. Blacket et al. 2012, Jiang et al. 2014), enabling identification only to the complex level (e.g. B.

tryoni complex);

4) The confidence of assigning an unknown sequence to a reference species can be problematic if all relevant potential taxa are not able to be included in the reference dataset (e.g. Virgilio et

al. 2012). The degree of identification uncertainty is dependent upon a number of factors,

72

including the accuracy of the taxonomic reconstruction (Lou & Goulding 2010), how recently the sister species diverged (e.g. van Velzen et al. 2012), and the geographic scale of reference samples (Bergsten et al. 2012).

Other applications of DNA barcoding data in addition to species identification are beginning to be explored. For example, in a biosecurity context the DNA sequence data generated can also be used to trace lineages (Blacket 2011) and possible source populations (Barr et al. 2014).

AIM

This test aims to identify fruit fly species, from any life stage, through DNA sequencing and comparison with reference sequences of the DNA barcoding region using the publicly available reference information in BOLD.

TARGETS

Almost all of the relevant species of fruit flies from the Australasian region are represented in BOLD (Table 3). The small number of species that have not been sequenced to date belong to genera where many other species have been examined; this allows a DNA barcoding approach to at least place these species to the appropriate genus. Note: for some species complexes COI or ITS DNA sequencing is unable to differentiate individual species (see Table 3).

PROCEDURE

This document provides supporting information for a three-step DNA barcoding process involving:

1) DNA extraction for fruit flies (adults and immature stages)

2) PCR of the mitochondrial COI barcode gene region from the fruit fly DNA.

3) Data analysis of the COI barcode gene region for species assignment using BOLD.

7.3.2.1 PCR OF THE MITOCHONDRIAL COI BARCODE GENE REGION FROM FRUIT FLY DNA

The DNA barcode region of the COI gene is PCR amplified using a fruit fly-specific priming system, producing a ~550 bp amplicon for sequencing. The PCR Master Mix reagents and set up must be in a

separate storage and at a separate bench (respectively) to the DNA extraction and post-PCR product

analysis (electrophoresis or RFLP, Sections 7.3.3 and 7.3.4).

Equipment and/or material needed

Primers:

o Forward: FruitFlyCOI-F (FFCOI-F) 5’-GGAGCATTAATYGGRGAYG-3’ (Blacket et al. 201215)

o Reverse: HCO 5’- TAAACTTCAGGGTGACCAAAAATCA-3’ (Folmer et al. 1994)

15 This primer is a fly-specific primer that was initially successfully tested on Bactrocera, Ceratitis (Tephritidae) and Calliphora (Calliphoridae) species.

73

PCR Master Mix:

o BSA [1X] (diluted with distilled H2O from 100X BSA stock)

o NEB 10X Buffer (Cat# M0267S)

o dNTP’s [2.5 mM]

o MgCl2 [25 mM]

o Primers [10 µM] (working primer concentration is 10 µM, store stocks at 100 µm, -20oC)

o NEB Taq DNA Polymerase (Cat# M0267S)

o QIAGEN QIAquick Spin Handbook, March 2008 (for reference if required)

DNA template of unknown specimen(s) (see Section 7.3.1)

Fruit fly DNA to use as a positive control (to confirm that the PCR amplification is working)

AE Buffer (or TE) from the DNA extraction kit can be used to replace DNA as a negative control (to confirm that no PCR reagents are contaminated).

Thermal cycler, e.g. “T800”, Eppendorf (epgradient S) Thermocycler

Additional equipment / materials: Submerged gel electrophoresis power pack, gel tank, TBE buffer, agarose etc.

Methods

1) Extract fruit fly DNA for use as template (see Section 7.3.1). 2) Set up Master mix (see below), keeping all reagents on ice during setup and adding the DNA

template last. 3) The Master Mix consists of 23 µL reaction volume for each sample (plus a negative and

positive control). Note, if dealing with large numbers of samples (e.g. >10) an extra reaction may be required to account for retention of liquid in pipette tips:

Reagent

x 1 reaction (µL)

MasterMix

x 3 reactions (µL)

1X BSA 15.3 45.9

10X Buffer 2.5 7.5

dNTP’s 2 6

MgCl2 0.5 1.5

FFCOI-F 1.25 3.75

HCO 1.25 3.75

NEB Taq 0.2 0.6

DNA Template 2 -

TOTAL: 25 µL 69 µl (x 23 µL per reaction)

74

4) PCR Conditions (“T800”, Eppendorf (epgradient S) Thermocycler):

1 x cycle 94 °C, 2 min

40 x cycles 94 °C, 30 s

52 °C, 30 s

72 °C, 30 s

1 x cycles 72 °C, 2 min

1 x hold 15 °C, indefinitely

5) After PCR is complete, load 5 µL of the PCR product (plus 2 µL loading dye) onto a 2% agarose checking gel (use 5 µL of SYBR Safe [Cat# S33102] per 50 mL liquid gel mix, before casting gel). Mix PCR product and dye together in plastic “gel loading” plate, using a new pipette tip for each sample. Run agarose gel at 100 V, for 30 min. Visualise and photograph gel on a UV light box. Estimate PCR product concentration for DNA sequencing from the agarose gel photo (weak PCR ~20 ng µL-1, strong PCR >100 ng µL-1).

6) If necessary, clean successful PCR products using QIAquick PCR Purification Kit (QIAGEN, Cat# 28104), elute final volume in 30 µL of EB Buffer (some external sequencing companies, e.g. Macrogen do this step).

7) Send to external facility (e.g. Micromon, Monash University or Macrogen, Korea) for DNA sequencing in both directions using the PCR primers above (Note, depending on each laboratories procedures, and quality of the sequence obtained therein, sequencing may be conducted in one direction only).

8) After high quality DNA sequences have been obtained (preferably with a QV or Phred score of greater than 20), and with PCR primer sequence removed and the forward and reverse consensus sequence determined (if sequencing both directions), they can be compared with a public database (i.e. BOLD, Barcode of Life website16) to identify species as outlined in Section 7.3.2.2.

16 http://www.boldsystems.org

75

7.3.2.2 DATA ANALYSIS – DNA BARCODING SPECIES IDENTIFICATION

The DNA sequence data for each specimen is compared with other publicly available reference sequences for species identification. Some species, such as Bactrocera tryoni, are members of a species complex, where the very closely related species within the complex cannot be distinguished by their DNA barcode sequence; these specimens are thus reported as being identified to a species group (e.g. B. tryoni complex, rather than B. tryoni).

Method 1) Go to the Barcode of Life website (http://www.boldsystems.org/). 2) Click the “Identification” tab (http://www.boldsystems.org/index.php/IDS_OpenIdEngine). 3) Paste the DNA sequence (use only the high quality section of the DNA sequence) into the

“Enter sequences in fasta format:” box (the sequence can just be entered as simple text and does not need to be in FASTA format).

4) Select the “Species Level Barcode Records” (default); this selects reference sequences that are >500 bp from named (identified) specimens, but some species may be represented by only one or two specimens or be included with interim taxonomy.

5) Click the “Submit” button. 6) The top 20 matches are displayed (default), 99 best matches are available, together with the

“Specimen Similarity” score (as a percentage). 7) The matches with the highest percentage similarity (listed from highest to lowest) are the

reference sequences that best match the unknown specimen being identified. The best matches are listed under “Search Results” and in a “Identification Summary” table together with the percentage “Probability of Placement” (Figure 9).

8) The matches should also be viewed as a phylogenetic tree using the “Tree based Identification” button.

9) Click “View Tree” to view a PDF of the phylogenetic tree. 10) On the tree the specimen being identified is referred to as the “Unknown Specimen” (written in

red) on the tree (indicated with arrows in Figure 10 and Figure 11), and is shown closest to the reference specimens that it best matches.

11) The specimen can now be assigned to the species that it is most similar to, with three caveats: a. Some specimens can only be assigned to a species group (i.e. a closely related

complex of species, Figure 11), where the species within it are unable / too similar to be distinguished using DNA barcoding. The indications for this are when different specimens of the same species are mixed among other species and there is no clear single-species clade (Figure 11); this is known to occur for the complexes of B. tryoni, B. dorsalis and Anastrepha fraterculus.

b. The assignments made by BOLD are based on a generic <2% within-species divergence (i.e BINs, Ratnasingham & Hebert 2013). This may not be appropriate for all of the species being analysed. For example Figure 12 shows divergences within the B. tryoni complex are very similar (<2%) to another closely related distinct species, B. curvipennis.

c. The closest match is not always with the correct species. BOLD recommends that an “identification is solid unless there is a very closely allied congeneric species that has not yet been analysed”. While most pest fruit fly species of interest are represented in BOLD (e.g. Table 2), closely related congeners may not. Indications for this are difficult to observe and this caveat is dependent on knowledge of the local taxonomy.

76

Figure 9. BOLD generated identification results for the closest matches for an Island Fly (Dirioxa pornia) COI sequence (accessed June 2015).

77

Figure 10. Specimen confidently assigned to species (Lamprolonchaea brouniana), (BOLD tree June 2015).

78

Figure 11. Specimen only confidently assigned to species group (Bactrocera tryoni complex), due to three closely related species (B. tryoni, B. aquilonis, B. neohumeralis) being “mixed together” (i.e. non-monophyletic) on the phylogenetic tree (BOLD tree June 2015).

79

Figure 12. Closely related species (B. curvipennis) genetically distinct from the species complex (B. tryoni) but within the <2% assignment limit set by BOLD (BOLD July 2015).

80

7.3.3 PCR-RFLP

OVERVIEW

Two tests are described. Both are based on fruit fly-specific PCR of the rDNA ITS region (Figure 13). Differences in method details (sections 7.3.3.1 and 7.3.3.2) reflect subtle differences in the standard operating procedures of the original laboratories.

Test 1, developed by McKenzie et al. (2004), utilises a 0.6 to 1.2 kb PCR fragment of the ITS1 only (Figure 13). As the ITS1 can be variable in length, for some species the size of the PCR fragment can be an additional diagnostic character. Separate aliquots of the PCR amplicon are digested with each of up to six different restriction enzymes to produce a species-specific RFLP profile. This number may be reduced if the potential species could be narrowed down by knowledge of likely origin and/or host, and by reference to Table 4 which illustrates which enzymes distinguish which species. This test can be used for identification of 26 fruit fly species (Table 3).

Test 2, originally developed by Armstrong and Cameron (1998), is similar to Test 1 but amplifies a larger 1.5-1.8 kb DNA fragment, encompassing the 18S and the ITS1 gene regions (Figure 13). Because of this, the variability in PCR fragment length is not so discernible and is therefore not useful as an additional character. Various combinations of three or four out of 10 restriction enzymes are recommended to produce a species-specific RFLP profile. The specific suite will depend on the likely species identity if it can be narrowed down by likely country of origin or the fruit. This test can be used for identification of 33 fruit fly species (Table 3); details are available in Armstrong & Cameron (1998). In the specific circumstance to distinguish just B. tryoni complex from C. capitata the enzymes AluI, DdeI, RsaI and SspI are diagnostic.

Figure 13. Part of the ribosomal RNA operon with the location of primer positions for Tests 1 and 2

18S ITS1 5.8S ITS2 28S

Test 1 baITS1f baITS1r

Test 2

The general workflow is presented in Figure 14.

NS15 ITS6

81

Figure 14. Workflow of PCR-RFLP procedures for fruit fly identification

1. Choose standards to run in conjunction with unknown and up to four restriction enzymes for

diagnosis

2. DNA extraction from unknown sample

Section 7.3.3

3. PCR amplification

Section 7.3.3.1 or 7.3.3.2

5. Electrophoresis test for amplified DNA

7. Test 1. Record if ITS1 fragment length is

diagnostic (Table 4).

6. Has DNA been amplified?

11. Compare fragment number and length with

reference data

Test 1 Table 5

Test 2 Appendix 4

10. Set up fresh reactions; incubate for longer

12. Document fruit fly identification

End

Start

Yes

No

Yes

No

8. Test 1 & 2 Restriction analysis. Section 7.3.3.1 or

7.3.3.2

Yes

13. Consider additional enzymes, or conclude it is a different species to

any of the reference species

No 9. Is identity consistent across enzymes; does identity make sense re other information on fruit host or origin?

Yes

9. Have restriction reactions gone to

completion?

82

SPECIES TARGETS

Test 1 can distinguish 25 of 27 target species within the current list of 59 species (Table 1); three species within the Bactrocera tryoni group (including B. tryoni, B. neohumeralis and B. aquilonis) cannot be distinguished from each other.

Test 2 can distinguish 24 of 33 target species within the current list of 59 species (Table 1); nine species form four species groups.

Test 2a distinguishes B. tryoni complex from C. capitata only.

All tests should be considered in terms of the host fruit (for immature life stage samples) and likely country/place of origin to (a) narrow the possible species at the outset for choice of restriction enzymes and (b) to check the results for reliability. Host records (Section 7) for the target taxa may assist in the elimination of possible non-target species.

CHOICE OF RFLP TEST

Test 1 is more rapid, the RFLP patterns less complex to interpret and the variation in size of the ITS1 region easier to detect than Test 2. Therefore this test is recommended when the potential species is within that listed for this test (Table 3).

Test 2 takes slightly longer, the RFLP patterns are sometimes more complex to interpret and the variation in size of the ITS1 region not as easy to detect is in Test 1. However, a slightly modified version (Semeraro & Malipatil 2005) provided below, is recommended instead of Test 1 specifically for the diagnosis of the high-risk pest species B. tryoni (although indistinguishable from B. neohumeralis and B. aquilonis within the species complex) and Ceratitis capitata. The original method provided in Armstrong and Cameron (1998) (not included in this Handbook) is also recommended for a further 11 species not included in Test 1 but within the now extended species list in this version of the Handbook (Table 3); the diagnostic data for Test 2 is however provided in Appendices 3 and 4.

Potential for misidentification: False positives are possible if other less economically important species, not included during

development of the tests, have the same RFLP pattern. This scenario is reduced by increasing the number of restriction sites screened by using several restriction enzymes for an identification. It is also unlikely for exotic fruit fly interceptions given the low potential for non-pest species to reach Australia. For Australian methyl eugenol or cue lure trapped specimens, the range of taxonomically close local non-target species has not been included in development of either test. Therefore care is needed in interpretation of the results in light of other information about the sample, e.g. data on host or geographic origin (Section 8).

False negatives could arise if a non-conforming RFLP pattern is produced by a population of a species that has an aberrant polymorphism at a diagnostic restriction site. This is unlikely given the range of populations included during the development of these tests. Also, the ITS1 is generally not useful for detecting population-level variation (cf COI barcodes, Section 7.3.3). False negatives could also arise if the restriction fails or fails to go to completion. DNA from a positive control (known species) should therefore be included to detect this.

CHOICE OF STANDARDS

All analyses should incorporate at least one standard, i.e. DNA from a known species. This can include one of the anticipated species for identity of the unknown specimen, which is useful as a positive control to help with sizing the restriction fragments of the unknown. Alternatively, it can be from a completely unrelated species, which is useful to minimise any question of contamination or mix up. Standards are essential to provide confidence that the PCR and restriction steps are operating correctly.

83

7.3.3.1 PCR-RFLP TEST 1

A) PCR

Equipment Pipettors and tips Sterile disposable microcentrifuge tubes Microcentrifuge Gel tank and power pack Latex or Nitrile gloves Microwave UV transilluminator with camera Thermocycler Personal protective equipment including lab coat, eye protection, gloves

Reagents (see APPENDIX 2 for reagent compositions)

Manufacturer’s polymerase enzyme buffer 10X Taq polymerase enzyme (5U μL-1) Primer (McKenzie et al. 1999) are:

o baITS1f 5’ GGA AGG ATC ATT ATT GTG TTC C 3’, 10 μM o baITS1r 5’ ATG AGC CGA GTG ATC CAC C 3’, 10 μM

dNTP’s (2 mM) MgCl2 (50 mM) Sterile water 1X TBE buffer 1% (w/v) agarose gel: 1 g DNA grade agarose per 100 mL 1X TBE 6X Loading dye DNA molecular weight marker (aka 100 bp ladder) Gel staining solution, e.g. Ethidium bromide (final concentration 800 ng/μL), or non-toxic options

such as Syber Safe, or Red Safe according to manufacturer’s instructions

Method In a pre-PCR cabinet:

1) label sterile 0.2 ml PCR tubes 2) make a Reagent Master Mix for the total number of samples to be analysed according to the

table below, all reagents except for the polymerase (note, recommend adding an extra volume for more than 10 samples to allow for any loss of solution via adherence to the outside of the pipette tip):

Reagent Final concentration Vol per reaction (μl)

Manufacturer’s reaction buffer (10X) 1X 5

MgCl2 (50 mM) 1.5 mM 1.5

dNTP’s (2 mM) 200 μM 5

Forward primer (10 μM) 1 μM 5

Reverse primer (10 μM) 1 μM 5

H2O 20.25

Taq polymerase enzyme (5U μL-1) 0.25

Total volume 42

84

3) Store Master Mix on ice in sterile 1.5 mL centrifuge tube. 4) Add 8 μL of sterile distilled H2O or DNA buffer to the first negative control tube.

In a laminar flow hood or PCR cabinet:

5) Add the Taq polymerase to the Master Mix 6) Aliquot 42 μL Taq Master Mix to each PCR tube 7) Add 8 μL of DNA extract/control to each sample tube as appropriate 8) PCR amplify the DNA in a thermal cycler using the following program:

Cycle 1 - denature 94°C 2 min

Cycles 2 to 35 - denature 94°C 1 min, anneal 60°C 1 min, extend 72°C 1 min

Cycle 36 - extend 72°C 5 min

9) Place reaction products on ice or freeze until ready to analyse. 10) Mix 3 μL of each PCR sample with 2 μL loading dye. 11) Load samples and 100 bp DNA ladder onto separate wells of 1% (w/v) agarose gel in 1X TBE. 12) Electrophorese in 1X TBE buffer at 100 V for around 40 min. 13) Stain the gel in ethidium bromide or other stain, according to local Standard Operation

Procedure. 14) Visualise bands and capture image using the Gel Documentation System.

Diagnostic use of ITS1 amplicon size The expected size of the amplified product is between 500 and 1000 bp, depending on the species (Table 4). For some species the size can be an additional diagnostic character. However, relying on the resolution available on a gel to distinguish the majority of these amplicon sizes is not advisable as the only diagnostic character, but is useful in conjunction with restriction analysis as described below. Flies producing fragments of less than 700 bp or greater than 900 bp are segregated and then restriction enzymes are used in series to differentiate the species. Sizes are given as a range to reflect that sizing is approximate when using low-resolution gel electrophoresis systems as here.

85

Table 4. Approximate size of the ITS1 amplicon for each species. Potential diagnostic size ranges are contained in a block

Species Fragment range

ITS1 amplicon ~size range (bp) Potential species

500-520 D. pornia

590-610 B. cucurbitae

640-680 A. ludens

650-690 A. obliqua

670-700 B. xanthodes

740-760 A. serpentina

740-780 R. pomonella

750-780 B. umbrosa, B. facialis

760-770 B. cucumis

760-780 B. latifrons

770-790 B. musae

770-800 B. endiandrae

780-800 B. psidii

790-840 B. bryoniae, B. tryoni sp. complex

800-820 B. moluccensis

800-840 B. dorsalis sp., B. jarvisi

810-840 B. passiflorae

820-850 B. zonata

830-860 B. carambolae, B. curvipennis, B. frauenfeldi

840-860 B. albistrigata, B. kirki

890-900 C. capitata

1000-1040 C. rosa

86

B) RESTRICTION DIGESTION OF PCR PRODUCT

The use of a combination of enzymes, in series, allows definitive identification of the majority of the species. This also eliminates the reliance on discrete restriction sites and minimises the likelihood of false negatives that may arise through a rare recombination event.

Restriction endonucleases used are VspI, HhaI, SspI, HinfI, BsrI, SnaBI and/or Sau3aI. During the development of this test standard enzymes purchased from New England Biolabs were used but other brands would work equally well. Enzymes were also selected based on the requirement for differences in fragment sizes to be easily detected by visual examination of an agarose gel.

If the likely species can be narrowed down, e.g. by fruit or geographic region, then a reduced number of enzymes could be used. However, for robust diagnoses, it is recommended that at least four enzymes are used, even if there is a diagnostic pattern for one enzyme that distinguishes a species from all others.

Equipment

Pipettors and tips Sterile disposable microcentrifuge tubes Microcentrifuge Dry heating block, waterbath or similar Gel tank and power pack Latex or nitrile gloves Microwave UV transilluminator with camera and image capture and analysis software Personal protective equipment including lab coat, eye protection, gloves

Reagents Sterile distilled water Bovine serum albumin (BSA, 10 μg/μL) (comes supplied with NEB enzymes) Restriction enzymes VspI, HhaI, SspI, HinfI, BsrI, SnaBI, and Sau3aI Restriction buffer supplied with enzyme Ethidium bromide solution, 800 ng/μL final concentration, or non-toxic options such as Syber

Safe, or Red Safe according to manufacturer’s instructions

Method

1) Label microcentrifuge tubes, including one for the positive control.

2) To each centrifuge tube add:

Water 2.3 μL

10X buffer 2 μL

BSA (10 ug/μL) 0.2 μL

PCR product 5 μL

Restriction enzyme 0.5 μL

3) Mix reagents and place tubes in a water bath preheated to 37oC for 2 h.

4) Store tubes on ice or at -20oC until ready to load on agarose gel.

5) Add 3 μL of 6X loading buffer to each tube.

6) Load the entire volume of each sample (23 μL) into a lane of a 2% (w/v) high resolution agarose gel.

7) Load 100 bp DNA molecular weight marker into one or two wells of the gel.

8) Analyse products by electrophoresis at 100 V for 50 min.

87

9) Stain the gel with ethidium bromide, or alternative non-toxic stain.

10) Visualise fragments using a UV transilluminator.

11) Capture gel image using gel documentation system.

Analysis of RFLP products For diagnostic purposes, RFLP bands under 100 bp and over 1500 bp in size are disregarded due to difficulty in accurate sizing. The molecular weights of the restriction fragments are estimated with reference to the DNA molecular weight standard loaded on the same gel.

Tables 4 and 5 summarise the expected fragment lengths for the six restriction enzymes used in this method. All species listed can be differentiated from each other, with the exception of those within the B. tryoni complex. Care should be taken with fragments produced for B. albistrigata and B. kirki, which are very similar for all six restriction enzymes and potentially difficult to confidently distinguish on an electrophoresis gel.

Species

88

Table 5. Analysis of RFLP products from ITS1 fragments from fruit flies

Species ITS1* HinfI VspI Hhal SspI BsrI SnaBI Sau3aI

˂700 700-900 ˃900 DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts

A. ludens X 550 550 X X X X X

A. obliqua X 450, 270 550 X 550, 150 X X 450, 200

A. serpentina X X 420, 250 X X X X 530, 200

B. albistrigata X X X 670, 180 620, 180 X X 450, 400

B. aquilonis X 770 X 640, 190 570, 180 600, 200 X 415

B. bryoniae X 760 X 620, 200 560, 180 600, 230 X 400

B. carambolae X X 480, 350 680, 200 X 650, 250 530, 350 450, 400

B. cucumis X X X 550, 180 X X X X

B. cucurbitae X X X 400, 180 X X X X

B. curvipennis X X X 620, 170 550, 200 570, 250 X 420 B. dorsalis sp. complex

X 770 X 650, 190 X 650, 260 540, 320 X

B. facialis X X X 600, 180 X 600, 200 X 390

B. frauenfeldi X X X 620, 180 X X 450, 400

B. jarvisi X 770 X 640, 180 700 600, 250 X 420

B. kirki X X X 680, 190 620, 180 X X 450, 400

B. latifrons X X X 600, 190 X 600, 200 X X

B. musae X X X 635, 220 X 600, 250 520, 320 X

B. neohumeralis X 770 X 640, 190 570, 180 600, 200 X 420

B. passiflorae X 770 X 650, 190 750 650, 270 X X

B. psidii X X X 640, 190 570, 250 X X X

89

The length of the ITS1 fragment and the response of each to seven restriction enzymes (HinfI, VspI, HhaI, SspI, BsrI, SnaBI, Sau3aI) are indicated for each of the target species. ITS1 fragment length is scored as one of three classes (approximate length in bp). Enzyme responses are measured in two classes - either does not cut (DNC) or cuts. (Cuts – this column shows the length of each fragment in bp). Highlighted boxes denote diagnostic RFLP patterns, which are more than 20 bp different to other fragment produced by that restriction enzyme.

Species ITS1* HinfI VspI Hhal SspI BsrI SnaBI Sau3aI

˂700 700-900 ˃900 DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts DNC Cuts

B. tryoni X 770 X 640, 190 570, 180 600, 200 X 420

B. umbrosa X 730 X 600, 190 680 X X 380

B. xanthodes X 680 X 670, 200 380, 250 X X X

B. zonata X X X 680, 190 750 600, 200 535, 330 X

C. capitata X X 650, 200 X 520, 160 X X X

C. rosa X 800, 200 600, 300 X 570, 480 X X X

D. pornia X X X X 300, 220 X X X

90

7.3.3.2 PCR-RFLP TEST 2

A) PCR

Materials and equipment As for PCR-RFLP Test 1 Section 7.3.3.1

Reagents As for PCR-RFLP Test 1 Section 7.3.3.1 with the exception of primers, which are here:

NS15 5’ CAATTGGGTGTAGCTACTAC 3’ ITS6 5’ AGCCGAGTGATCCACCGCT 3’

Method In a PCR (laminar flow or clean bench) cabinet:

1) label sterile 0.2 ml PCR tubes 2) make a Reagent Master Mix for the total number of samples to be analysed according to the

table below, adding the polymerase last (note, recommend adding an extra volume for more than 10 samples to allow for any loss of solution via adherence to the outside of the pipette tip):

Reagent Final concentration Vol per reaction (μl)

Double-distilled H2O 30.6

Expand High Fidelity polymerase buffer 1 X 5

dNTPs (2.5 μM) 200 μM 4

Forward primer 10 μM (NS16) 0.5 μM 2.5

Reverse primer 10 μM (ITS6) 0.5 μM 2.5

Expand Hi Fidelity Taq Polymerase 2 Units 0.4

Total 45

3) Store Master Mix on ice. 4) Vortex DNA extractions for 5 s. 5) Aliquot 5 μL of each unknown DNA template to labelled 0.2 mL tubes, plus at least one positive

control DNA and one negative control (water or the buffer in which the DNA is suspended). 6) Aliquot 45 μL of master mix to each 0.2 mL tubes (containing 5 μL of template DNA). 7) Mix product and reagents well (or vortex briefly) and centrifuge for 3-5 s. 8) Place samples in PCR machine and program the following temperature profile (based on

Armstrong and Cameron 1998):

• Cycle 1 - denature 94°C 2 min

• Cycles 2 to 35 - denature 94°C 15 s, anneal 60°C 30 s, extend 68°C 2 min

• Cycle - extend 72°C 5 min

9) Add 1 μL of loading dye to 5 μL of PCR product 10) Load onto a 1.5% agarose gel, together with a 100 bp molecular ladder at either side and

electrophorese according to Appendix 2. If product visible at 1.5-1.8 kb then proceed to Section B) – restriction digestion.

91

B) RESTRICTION DIGESTION OF PCR PRODUCT

Equipment See 7.3.3.1 B

Reagents Restriction enzymes, on ice.

o NB. Choose at least four of the enzymes listed in Appendix 3 according to which combination will provide the best discrimination for the potential species; the letters represent RFLP patterns in Appendix 4

o Note: In the specific circumstance of distinguishing only B. tryoni complex and C. capitata the enzyme combination of AluI, DdeI and RsaI (10 U/μl) and SspI (5U/μl) is diagnostic.

Restriction enzyme buffers, 10X Sterile nuclease free H2O

Method 1) Prepare master mix (following recipe below) for each enzyme; multiplying volumes plus one

for the number of reactions required.

Adapted from Armstrong & Cameron (1998)

Final concentration Vol per reaction (μl) Double-distilled H2O 5.6*

10X Buffer 1X 1.0

Enzyme† 4 U 0.4

Total 7.0 * volume of water can be varied to accommodate any change to DNA volumes added, see below

† Standard stock concentrations are usually 10 U/µl; if not, this volume will change accordingly

2) Aliquot 7 µL of each master mix into labelled tubes 3) Add 2-3 μL (100-200 ng) PCR product into each tube depending on how strong or weak the

PCR products are; adjust water in reaction mix accordingly 4) Flick to mix reagents and PCR product, centrifuge briefly for 3-5 s. 5) Place samples in incubator at 37°C 2-3 h unless otherwise recommended by the enzyme

manufacturer 6) Prepare a 2-3% agarose gel according to Appendix 2, at least 10 cm in length, to visualise

fragment pattern and use a 100 bp ladder for determining fragment sizes 7) Compare results with positive controls and fragment patterns in Appendix 4 8) Species diagnosis is considered positive if all four enzyme patterns agree.

92

7.3.4 Real-time PCR detection of Bactrocera tryoni complex

This test was developed and validated by Dhami et al., (2015), further validation was conducted by David Waite and Dongmei Li during 2014-2015 at the Plant Health and Environment Laboratory (PHEL), Ministry for Primary Industries (MPI), New Zealand. This section (7.3.4) was written by Dongmei Li, David Waite and Disna Gunawardana.

INTRODUCTION

Genetically similar but morphologically different species have been reported in several fruit fly species (Clarke et al., 2005; Krosch et al., 2012). Such species are considered as belonging to “species complexes” (Drew, 2004). The commonly called ‘Queensland fruit fly’ (Q-fly) is one such species complex, consisting of four species; B. tryoni, B. aquilonis, B. neohumeralis and B. melas (Drew, 1989). However, while B. melas exists in historic documents, researchers now consider it a cryptic form of B.

tryoni (Clarke et al., 2011).

Analysis of microsatellite markers has shown that there is only intraspecies level of differentiation present between B. tryoni and B. aquilonis (Wang et al., 2003). Those analyses also reflect that B.

neohumeralis and B. tryoni represent separate species (Clarke et al., 2005; Wang et al., 2003). The closely related species B. curvipennis, however, is difficult to distinguish from B. tryoni complex based upon DNA sequencing of the cytochrome c oxidase subunit 1 (COI) gene (Armstrong & Ball, 2005), although it is morphologically distinct. More comprehensive molecular analysis has similarly indicated that B. curvipennis has a paraphyletic relationship with the B. tryoni complex and is currently considered very close to, but not within, the B. tryoni complex sensu Drew 1989 (Blacket et al., 2012; Jiang et al.,

2014).

DNA barcoding relies on PCR of predetermined marker genes, DNA sequencing and comparison of those sequences to a database of reference sequences (Armstrong & Ball, 2005). Currently, DNA barcoding of the COI gene is the most commonly used technique for the molecular identification of fruit flies (Armstrong & Ball, 2005; Blacket et al., 2012). This process is still relatively time consuming, with the quickest possible identification requiring a full day. In contrast, real-time PCR, or quantitative PCR (qPCR), drastically reduces the end-to-end time of analysis. This technique is based on the amplification of DNA monitored in real time and therefore without the need of post-PCR processing (e.g. gel electrophoresis), thus it enables identification within a few hours. The premise for this technology is that highly species-specific PCR primers or probes are designed to give positive reactions only with DNA of the target species. A positive identification is determined by the measurably lower number of PCR cycles taken for the amplicon to reach a given concentration, and for fluorescence to be detected (the quantitation cycle, Cq), than would occur for a negative or suboptimal amplification, as would be expected from DNA of the wrong species. The TaqMan chemistry has the potential for greater specificity through the incorporation of a third oligonucleotide in the reaction. The method has previously been applied to other fruit fly species on the list in this manual (Burgher-MacLellan et al., 2009; Yu et al., 2005; Yu et al., 2004).

Here, a real-time PCR assay for the B. tryoni complex has been developed, validated and applied by PHEL in the routine diagnostics of intercepted fruit fly material at borders, post borders and in recent New Zealand Q-fly responses.

AIM

This assay aims to provide a rapid method for the identification of the B. tryoni complex, from any life stage, using the real-time TaqMan technique.

TARGET

This assay can be used identify B. tryoni complex, (B. tryoni, B. aquilonis, and B. neohumeralis), and the closely related species B. curvipennis.

93

PROCEDURE

This document provides all the supporting information for conducting the real-time PCR assay and analysis of the results, including:

1. DNA extraction and subsequent PCR from a range of life stages of the fruit flies including eggs, part of an adult (one leg is commonly used in diagnostics), part of a larva, part of a pupa, empty pupal case

2. PCR competency test 3. Real-time PCR assay for the B. tryoni complex (FAM probe) 4. Results - analysis of Real-time PCR assays.

7.3.4.1 DNA Extraction protocols

See section 7.3.1.

7.3.4.2 Real-time PCR assay for B. tryoni complex

Set up the PCR assay in a PCR workstation. The assay is described as run on the CFX1000 TM real-time PCR system (BioRad), with the data analysed in the CFX manager 3.0 analysis software (BioRad).

Equipment and material Primers and Probe (Dhami et al., 2015)

o Forward primer: Btry2F, 5’-AATTGTAACAGCCCATGC-3’ o Reverse primer: Btry1R, 5’- GTGGGAATGCTATATCGG-3’ o Probe: Btry2PL: 6FAM-AG[+C]CA[+G]TTTCC[+G]AA[+A]CC-BBQ (or BHQ1)

Real-time PCR mastermix o SsoAdvanced Universal Probe Supermix (Cat#1725280, Biorad), 2x qPCR mix,

containing dNTPs, Sso7d fusion polymerase, MgCl2, stabilizers, ROX normalization dyes.

o Primers/probe, working concentration of 5 µM (primer stock concentration 100 µM, stored at -20°C in dark)

o BSA (Bovine serum albumin, working concentration of 10ng/µl) (Sigma Cat# A788-50g)

DNA samples of unknown Tephritidae species extracted as in section 7.3.1. Internal positive control to test unknown DNA samples for PCR competency to avoid false

negative results o Conventional PCR with Folmer primers LCO1490 and HCO2198 (Folmer et al., 1994) o TaqMan ribosomal RNA control reagents (VIC probe) (Applied Biosystems, cat #

4308329), see Appendix 1 for details. Controls for real-time PCR setup:

o Positive control to monitor the performance of the real-time PCR DNA samples of B. tryoni complex or Plasmid DNA of COI insert of B. tryoni complex (available from MPI PHEL,

NZ on request) o Non-template control

Sterile water or PCR-clean water PCR work station or DNA/RNA-free area for PCR setup Thermal cycler: CFX96 or CFX1000 Touch Real-time PCR (BioRad). Other brands of real-

time thermal cycler can be used. Additional equipment: centrifuge, pipettes, plugged PCR tips, PCR tube, PCR plates, plate

seals.

Method 1. PCR competency test: It is recommended prior to any qPCR analyses to test all the DNA

extractions for PCR competency, using either conventional PCR or 18S internal control TaqMan assay (see Appendix 1 for details)

94

2. Set up the real-time PCR Mastermix in a PCR work station or clean area, see below for the compositions (Table 6). Include enough volume to run the samples in duplicate or triplicate plus positive and non-template controls. It can also be run at 20 µl or 10 µl volume.

a. For 20 µl volume, aliquot 18 µl of master mix to each PCR tube b. For 10 µl volume, aliquot 9 µl of master mix to each PCR tube

Table 6. TaqMan real-time PCR for B. tryoni complex protocol using SsoAdvanced Universal Probe Supermix (BioRad)

Reagents 1 x reaction (µl)a 10x reactions (µl) b

Sterile distilled H2O 2.2 22

2 x Probe Supermix 10.0 100

5 µM Btry2F (400 nM) 1.6 16

5 µM Btry1R (400 nM) 1.6 16

5 µM Btry2PL(250 nM) 1.0 10

BSA (10mg/ml) 1.6 16

DNA template 2.0

Note: aThe compositions for 20 µl are listed in this table, halve the volumes for each reagent if using in 10 µl volume. bUsing 10x reaction as an example, calculate the volumes of each reagents using the number of reactions you are going to test when conducting the assay; note, one or two extra reaction volumes should be included to account for loss due to retention of liquid in pipette tips.

3. Set up the Mastermix, mix well and spin briefly, aliquot 18 µl of the Mastermix to each labelled

PCR tube or well of a plate. 4. Add the appropriate DNA template or control solution to the tubes or the wells of a plate with

the aliquoted Mastermix, close the tubes or seal the plate, mix well and spin the tubes or plate briefly

5. Put the tubes/plate into real-time PCR machine and run the program below:

PCR cycling parameter (CFX 1000TM Thermal cycler)

1 x cycle 95 °C, 2 min

40 x cycles 95 °C, 15 sec

65 °C, 60 sec

Plate read after each cycle

6. Open CFX manager 3.0 program, set up the protocol and run the real-time PCR assay. 7. Name the assay and save the data in a folder. 8. Once the run is finished, the data can be opened in the CFX manager 3.0 and analysed.

95

7.3.4.3 Analysis of Real-time PCR results

The real-time PCR results are analysed here with the CFX Manager 3.0 (BioRad). If different real-time PCR systems are used, analysis of the amplification curves will need to be carried out according to that manufacturers’ manual.

1. Open BioRad CFX Manager 3.0. 2. Click file, open the data file, and choose the real-time PCR data you have saved for the assay. 3. Click on quantification tab, the amplification curve will appear on the screen, see Figure 15 for

an example. 4. Set up the baseline for 50. 5. Report: Click Tools tab and choose reports, an analysis report for the real-time PCR including

amplification curves and Cq values will be generated. It can be saved as pdf format or print it out.

6. Interpretation of the results - 18S internal control: a. If Cq values ≤30 cycles, the DNA extraction is PCR competent and the samples can

be used for real-time PCR assay against B. tryoni complex. b. If Cq values >30 cycles, this indicates that inhibitors are in the DNA sample or there is

insufficient DNA. Re-extraction of the samples will be needed. 7. Interpretation of the results - B. tryoni complex assay (use the PCR competent DNA samples

only): a. For Cq values ≤30 cycles, the sample is considered as positive for B. tryoni complex,

and possible for closely related B. curvipennis (Cq values between 25-30 cycles) i. DNA barcoding needs to be conducted to confirm the species if the Cq values

>25 cycles. ii. The origin of the specimen might assist in distinguishing B. tryoni and B.

curvipennis (e.g. B. curvipennis is endemic to New Caledonia and not present in Australia).

b. Samples with Cq values between 30-35 cycles could be interpreted as either a different species with suboptimal match to the primers or that there is a very low copy number of the target species DNA. These should be considered as questionable and require further investigation by either DNA barcoding or re-extraction of DNA.

c. The negative threshold for the assay is Cq values >35, at which point the samples should be considered as negative for B. tryoni complex, and identification by another method is necessary.

8. Validation of the real-time PCR assay for B. tryoni complex used DNA extracted from fresh and aged (up to 2 year olds) samples. This included B. tryoni from NSW, VIC, QLD (n>80) and B. neohumeralis from QLD (n>20). The following results were observed;

a. Cq values ≤25 cycles were obtained with the DNA extracted from the B. tryoni complex except those with empty pupal cases.

b. Cq values between 25-35 cycles were observed for the DNA extracted from B.

curvipennis samples. DNA barcode sequencing confirmed the identification. c. Negative results (Cq >35) were obtained for all other Tephritidae species tested (see

Appendix 2).

96

Figure 15. Amplification curves of real-time PCR assay for B. tryoni complex. The sample tested positive for B.

tryoni complex, DNA sequences has further confirmed the results.

B. tryoniPositive control

Sample tested

Water control

97

7.3.5 Allozyme electrophoresis

This section was written by Mark Adams.

AIM

Allozyme electrophoresis provides a method for the rapid molecular identification of various species of fruit fly.

TARGETS

Routinely targets Bactrocera tryoni, Ceratitis capitata, Dirioxia pornia, Bactrocera dorsalis s.s. and Bactrocera jarvisi (Table 3). Additional species can be incorporated where suitable reference material is provided as freshly-frozen specimens (either as larvae or adults).

SUITABILITY

Suitable for the comparison of genetic profiles, as expressed in a range of soluble proteins, from live, recently-dead, or freshly-frozen larvae or adults. The service is currently routinely provided by the South Australia Museum's Evolutionary Biology Unit laboratory. The procedures take 2-3 hours to complete for a single screen of up to 20 specimens for 10 different genes.

Given the comparative nature of the technique and its continued reliance on reference samples, it is important to note that additional species can only be identified as "new" (i.e. not one of the five reference species) unless suitable, known-identity samples can also be provided for a putative match. Moreover, the incorporation of additional species into the routine screening procedure may also require a re-evaluation of which enzyme markers are diagnostic for these additional species, in order to satisfy the minimum requirement of three diagnostic genetic differences between every pair of species.

PROCEDURE OVERVIEW

Crude extracts of soluble protein from live, recently-dead, or freshly-frozen larvae or adults are compared electrophoretically against known-identity extracts representing these five species.

Test samples are readily identifiable by their comparative allozyme profile (i.e. relative band mobility) at a suite of at least six enzyme markers, encoded by a minimum of 10 independent genes, and together able to unambiguously diagnose the five reference species from one another at a minimum of three genes.

Neither B. aquilonis nor B. neohumeralis are listed in the five target species, so this method is not designed to differentiate between B. tryoni and these other two, genetically very similar species.

7.3.5.1 Specimen preparation

Test specimens

Need to be supplied either (a) alive, (b) freshly dead and kept cool and moist, or (c) frozen when alive and not allowed to thaw until tested (dry ice required for transport; ice is not suitable)

Can represent any life history stage

98

Reference specimens

Frozen specimens representing the species requiring discrimination must be available. When kept at -70oC, these remain suitable for use as controls for at least 10 years.

A single homogenized larva provides enough homogenate to act as a reference specimen on up to eight separate occasions. Once prepared, these reference homogenates can be stored at -20 oC as separate ~5 µL aliquots inside glass capillary tubes. Thus reference specimens for any test run are usually available as pre-prepared homogenates straight from the freezer.

Specimen Preparation (ideally in cold room at 4oC)

Specimens are hand-homogenized in an equal volume of a simple homogenizing solution (0.02 M Tris-HCl pH 7.4 containing 2 g PVP-40, 0.5 mL 2-mercaptoethanol and 20 mg NADP per 100 mL).

~0.5 µL of homogenate loaded directly onto each gel.

The remaining homogenate is transferred as a series of ~5 µL aliquots into individual glass capillary tubes and stored at -20oC. These samples can either be subjected to further allozyme analysis if doubt remains as to species identity, or used as fresh reference material for the species thus identified (activity declines over a 12 month period at -20oC).

7.3.5.2 Electrophoresis (ideally in cold room at 4oC)

Allozyme analyses are conducted on cellulose acetate gels (CellogelTM) according to the principles and procedures of Richardson et al. (1986). Table 7 indicates the suite of enzymes most commonly used for fruit-fly genetic identifications and details the electrophoretic conditions employed for each.

7.3.5.3 Gel interpretation

The interpretation of allozyme gels requires some expertise; the intensity of allozyme bands changes over time after a gel is stained, plus banding patterns can be affected by the “freshness” of the specimen and by what type of gut contents are present (some plants contain compounds which affect fruit fly enzymes once the sample is homogenised). Richardson et al. (1986) devote an entire section to the interpretation of allozyme gels, but there is no substitute for experience.

7.3.5.4 Recording of results

All gels are routinely scanned several times over the time course of stain incubation and the resultant JPG files archived as a permanent record.

99

Table 7. Enzymes most commonly used for fruit-fly genetic identifications using allozyme electrophoresis.

Enzyme Abbr E.C. No.

No. of genes

Buffer1 Run time Stain Species delineated

Aconitase hydratase

ACON 4.2.1.3 2 I17 1.2 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni vs B. jarvisi vs B. dorsalis vs D. pornia

Aminoacylase ACYC 3.5.1.14 1 C 1.5 h @ 250 V

Manchenko (1994)

C. capitata vs B. tryoni vs B. jarvisi / B. dorsalis vs D. pornia

Alcohol dehydrogenase

ADH 1.1.1.1 2 C 1.5 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni vs B. jarvisi vs D. pornia

Aspartate aminotransferase

GOT 2.6.1.1 2 B 1.5 h @ 250 V

method 3; Manchenko (1994)

C. capitata vs B. tryoni / B. dorsalis vs B. jarvisi vs D. pornia

Glycerol-3-phosphate dehyrogenase

GPD 1.1.1.8 1 C 1.5 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni / B. jarvisi / B. dorsalis vs D. pornia

Glucose-6-phosphate isomerase

GPI 5.3.1.9 1 B 1.5 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni / B. dorsalis vs D. pornia

3-Hydroxybutyrate dehydrogenase

HBDH 1.1.1.30 1 B 1.5 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni / B. jarvisi vs B. papaya vs D. pornia

Isocitrate dehydrogenase

IDH 1.1.1.42 2 B 1.5 h @ 250 V

Richardson et al. (1986)

B. tryoni vs B. dorsalis

Malate dehydrogenase

MDH 1.1.1.37 2 C 1.5 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni / B. dorsalis / B. jarvisi vs D. pornia

Dipeptidase PEPA 3.4.13 2 B 1.3 h @ 250 V

Richardson et al. (1986)

C. capitata vs B. tryoni / B. jarvisi / B. dorsalis vs D. pornia

1Code for buffers follows Richardson et al. (1986).

17 Sample origin is placed in the centre of the gel