Double Gene Targeting Multiplex Polymerase Chain Reaction−Restriction Fragment Length Polymorphism Assay DiscriminatesBeef, Buffalo, and Pork Substitution in Frankfurter ProductsM. A. Motalib Hossain,† Md. Eaqub Ali,*,†,‡ Sharifah Bee Abd Hamid,† Asing,† Shuhaimi Mustafa,§

Mohd Nasir Mohd Desa,§ and I. S. M. Zaidul∥

†Nanotechnology and Catalysis Research Centre (NANOCAT) and ‡Centre for Research in Biotechnology for Agriculture(CEBAR), University of Malaya, Kuala Lumpur 50603, Malaysia§Institute of Halal Products Research, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia∥Deparment of Pharmaceutical Technology, Faculty of Pharmacy, International Islamic University, Kuantan 25200, Pahang, Malaysia

*S Supporting Information

ABSTRACT: Beef, buffalo, and pork adulteration in the food chain is an emerging and sensitive issue. Current moleculartechniques to authenticate these species depend on polymerase chain reaction (PCR) assays involving long and single targetswhich break down under natural decomposition and/or processing treatments. This novel multiplex polymerase chain reaction−restriction fragment length polymorphism assay targeted two different gene sites for each of the bovine, buffalo, and porcinematerials. This authentication ensured better security, first through a complementation approach because it is highly unlikely thatboth sites will be missing under compromised states, and second through molecular fingerprints. Mitochondrial cytochrome band ND5 genes were targeted, and all targets (73, 90, 106, 120, 138, and 146 bp) were stable under extreme boiling andautoclaving treatments. Target specificity and authenticity were ensured through cross-amplification reaction and restrictiondigestion of PCR products with AluI, EciI, FatI, and CviKI-1 enzymes. A survey of Malaysian frankfurter products revealedrampant substitution of beef with buffalo but purity in porcine materials.

KEYWORDS: double gene-targets, multiplex PCR-RFLP, molecular fingerprints, species-specific identification

■ INTRODUCTION

Authentication of animal species in food products is anemerging issue having implications for health, religion, culture,and fair economic practices.1,2 The recent food scandalsinvolving animal species is not limited to the inclusion ofhorse meat and pork in beef products;3,4 pork and rat meat inlamb products;5 and monkey, dog, and cat meat in exoticdishes.6 These issues are highly alarming because most of theseanimal adulterants are potential carriers of infecting zoonosesand strictly prohibited by several religious, cultural, andregulatory laws.Beef and buffalo are economically and culturally important

meat having the top rate of consumption in most parts of theworld. Religious, cultural, and geographical restrictions andpreferences over the consumption of beef, buffalo, and pork arehuge, and social outcry over their adulteration andconsumption have taken place from time to time.7,8 WhileEgyptians prefer buffalo because of their cultural preferences,some Europeans and Indians avoid beef because of the fear ofbovine spongiform encephalopathy (BSE) and religiousrequirements.9 On the other hand, pork is totally unacceptableto the Muslim, Jewish and select Christian dominations despiteits popularity in Western countries.10,11 Thus, the social,religious, health, and business interests in beef, buffalo, andpork are enormous, and there should be a trustworthy but low-cost method for their discrimination in the food chain.

Authentication of food components using physical attributesis either impossible or extremely difficult because of the loss ofmorphological biomarkers during processing and packag-ing.12,13 In this regard, molecular techniques have showngreat success.1 Although the protein- and lipid-basedbiomarkers are less effective because of their susceptibility todenaturation and/or modification under processing condi-tions,14 DNA biomarkers involving mitochondrial DNA haveshown great success because of their abundant presence inmultiple copies in most cells along with intraspecies conservedand interspecies polymorphic fingerprints.13,15,16 Polymerasechain reaction (PCR)-based detection schemes are amazingbecause they amplify specific DNA targets even from single or afew copies to easily detectable quantities under complexmatrices, eliminating sample scarcity and saving purificationcost and time.6,17 The species-specific PCR restriction fragmentlength polymorphism (PCR-RFLP) assays are especiallyinteresting because they offer the opportunity to authenticatea product by restrictive digestion of the amplified PCRproducts using one or more restriction enzymes (REs).6

Using the sequence variation that exists within a defined regionof DNA, the differentiation of even closely related species is

Received: May 17, 2016Revised: July 15, 2016Accepted: July 18, 2016Published: July 18, 2016

Article

pubs.acs.org/JAFC

© 2016 American Chemical Society 6343 DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

possible using a PCR-RFLP assay.18 Such assays have beensuccessfully applied to discriminate closely related species suchas cattle, yak, and buffalo;19 cattle−buffalo and sheep−goat;20swine and wild boar;21 and various fish species.22 However,these methods are mostly based on single and long-length DNAtargets which break down under natural or environmentaldecomposition and food processing treatments, making themless trustworthy and inconclusive for forensic investigation.23,24

To the best of our knowledge, no RFLP authentication hasbeen reported for mPCR products wherein multiple amplifiedproducts do exist. In this regard, multiplex PCR-RFLP (mPCR-RFLP) assay, especially the double gene targeting one withshort amplicon targets, would be especially useful andtrustworthy for the simultaneous detection of beef, buffalo,and pork products in various food products. Because of thepresence of more than one target for the same species, thedetection of the missing target would be complemented by asecond target because it is highly unlikely that both targetswould be broken down under the state of decomposition. Toaddress this issue, we report for the first time a double genetargeting PCR-RFLP assay with short amplicon targets for thediscriminatory authentication of bovine, buffalo, and porcinematerials in frankfurter formulation, a popular food item widelyconsumed across the globe.

■ MATERIALS AND METHODSMeat Sample Collection. Authentic muscle tissues of cow (Bos

taurus), buffalo (Bubalus bubalis), goat (Capra hiscus), lamb (Ovisaries), chicken (Gallus gallus), duck (Anas platyrhychos), pigeon(Columba livia), quail (Coturnix coturnix), cod (Gadus morhua),salmon (Salmo salar), pangas (Pangasius pangasius), tuna (Thunnusorientalis), tilapia (Oreochromis niloticus), rohu (Labeo rohita), frog(Rana kunyuensis), and turtle (Cuora amboinensis) were purchased intriplicate on three different days from various wet markets andsupermarkets (Pasar Borong Selangor, Serdang, Pudu Wet Market,Kuala Lumpur and Tesco, Petaling Jaya, Selangor). Pork (Sus scrofa)was purchased in triplicate from three different vendors from aChinese wet market located at Seri Kambangan in the Selangor state ofMalaysia. Three different animals of dog (Canis lupus familiariz), cat(Felis catus), rat (Rattus rattus) and monkey (Macaca fascicularis)species were killed by Dewan Bandaraya Kuala Lumpur (DBKL) aspart of population control and Wildlife Malaysia for other purposes,and muscle tissues were collected following institutional and nationallaws.24 All samples were transported under ice chilling and stored at−20 °C prior to DNA extraction.25

Preparation of Frankfurter. Model beef, buffalo, and porkfrankfurters were made in the laboratory following Razzak et al.26

(Table 1). The prepared frankfurters were deliberately contaminatedby spiking of 1%, 0.5%, and 0.1% of buffalo and pork, beef and pork,and beef and buffalo meat, respectively. As-prepared 0.1% contami-nated frankfurters were autoclaved at 121 °C under 15 psi pressure for2.5 h.27

To authenticate the four PCR products of beef and buffalo (Cocytb,CoND5, Bucytb, and BuND5) by RFLP analysis, beef and buffalofrankfurters were adulterated by spiking buffalo and beef, respectively,and were heat-treated by boiling at 98 °C for 90 min and autoclavingat 121 °C under 15 psi pressure for 2.5 h. Porcine frankfurters werealso boiled at 98 °C for 90 min and autoclaved at 121 °C under 15 psipressure for 2.5 h, and RFLP analysis was performed in a separateassay. All samples were stored at −20 °C until DNA extraction.DNA Extraction. Total DNA from meat and fish samples was

extracted using Yeastern Genomic DNA Mini Kit (Yeastern BiotechCo., Ltd., Taipei, Taiwan).6 Briefly, 20 mg of muscle tissue was groundand homogenized with a micro pestle followed by the addition of lysisbuffer and proteinase K. The mixture was incubated at 60 °C for celllysis and protein degradation. The spin column technique was used forthe attachment of DNA to the glass fiber matrix during centrifugation.

Ethanolic wash buffer was used to remove the contaminants, and thepurified DNA was eluted in an elution buffer. A 100 mg sample wastaken for the extraction of DNA from wheat (Triticum aestivum),onion (Allium cepa), garlic (Allium sativum), ginger (Zingiberofficinale), and pepper (Capsicum annuum) using DNeasy Plant MiniKit (QIAGEN GmgH, Hilden, Germany) following the manufacturer’sinstruction. NucleoSpin Food DNA kit (Macherey-Nagel GmbH &Co. KG, Duren, Germany) was used to extract DNA from foodproducts (frankfurters) (200 mg). The concentration and purity of allextracted DNA were determined using an ultraviolet−visilbe (UV−VIS) spectrophotometer instrument (NanoPhotometer Pearl, ImplenGmbH, Germany) based on the absorbance value at 260 nm andabsorbance ratio at A260/A280, respectively.

28

Design of Species-Specific Primers. Six sets of oligonucleotideprimers specific to the cytb and ND5 genes of cow, buffalo, and pigspecies were designed following a standardized procedure published inour earlier report.25 The designed primers and target ampliconsequences are presented in Table 2. The specificity of the designedprimers was ensured by three different testing systems. First, the basiclocal alignment algorithm search tool (BLAST) against nonredundantnucleotide sequences identified the target species as well as thedissimilarity index value with other species in the NCBI database.Second, the primers were aligned against 29 different nontargetspecies, of which 17 were land animal, 8 were fish, and 4 were plantspecies. All sequences were aligned using the MEGA 5 software toidentify the conserved and variable sequences to evaluate the presenceof mismatched bases. The designed primers were purchased from theFirst Base Laboratories Sdn. Bhd. The final species-specificity wasconfirmed in a practical PCR experiment through a cross-amplificationreaction in the presence of the target and 27 different nontargetspecies.

Multiplex PCR Assay. Prior to the optimization of the mPCRassay, simplex PCR assay was performed for each of the target specieswith individual set of primers as described in our earlier report.25 Thesimplex PCR was performed in a 25 μL reaction mixture comprising 5μL of 5X GoTaq Flexi Buffer, 0.2 mM each of dNTP, 2.5 mM MgCl2,0.625 U GoTaq Flexi DNA Polymerase (Promega, Madison, WI,United States), 0.4 μM of each primer and 2 μL (20 ng/μL) of DNAtemplate. For negative control, template DNA was replaced by distilledwater. In the simplex PCR, 0.4 μL universal eukaryotic primers(forward primer: AGGATCCATTGGAGGGCAAGT and reverseprimer: TCCAACTACGAGCTTTTTAACTGCA) that targeted 99bp site of eukaryotic 18S rRNA gene29 was added as positive controlfor eukaryotes. All PCRs were performed in an ABI 96 Well VerityThermal Cycler (Applied Biosystems, Foster City, CA, United States)with an initial denaturation at 95 °C for 3 min followed by 35 cycles ofdenaturation at 95 °C for 30 s, annealing at 60 °C for 30−35 s,extension at 72 °C for 40 s and the final extension at 72 °C for 5 min.

Table 1. Formulation of Model Frankfurter

frankfurter (≥70 g/piece)

ingredients beef buffalo pork

minced meat 45a 45a 45a

soy protein 7.5 7.5 7.5starch/breadcrumb 6.5 6.5 6.5chopped onion 2.5 2.5 2.5chopped ginger 0.15 0.15 0.15cumin powder 0.75 0.75 0.75garlic power 0.5 0.5 0.5black pepper 0.23 0.23 0.23tomato paste 2.0 2.0 2.0butter 2.5 2.5 2.5saltc SA SA SAothersb,c SA SA SA

aTo prepare ≥70 g frankfurter specimens, 10%, 1%, and 0.1% of beef,buffalo, and pork were mixed with a balanced amount of respectiveminced meat. bFlavoring agents and enhancers. cSA, suitable amounts.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6344

PCR products were visualized in 2% agarose gel stained with FlorosafeDNA stain (First Base Laboratories Sdn. Bhd., Selangor, Malaysia)under a gel documentation system (AlphaImager HP, Alpha InnotechCorp., California, United States) (data not shown). After optimizingsimplex PCR assays for individual species, duplex, triplex, tetraplex andfinally hexaplex PCR systems were developed as given in Tables 3 and

4. Because of the poor resolution of agarose gel, multiplex PCRproducts were separated and visualized in an automated QIAxcelAdvanced Capillary Electrophoresis System (QIAGEN GmbH,Hilden, Germany).Enzymatic Digestion and RFLP Analysis. The sequences of the

amplified PCR products were retrieved from NCBI and a publiclyavailable NEBcutter version 2.0 software (http://tools.neb.com/NEBcutter) was used to select the specific and appropriate restrictionendonucleases for all the PCR amplicons prior to test the mPCR-RFLP assay to ensure distinctive RFLP patterns for all targets. Therestriction patterns of the PCR amplicons of beef, buffalo, and porkmitochondrial cytb and ND5 genes are given in Table 5.RFLP Analysis of Beef and Buffalo PCR Products. The simplex

PCR products of beef cytb and buffalo ND5 genes were digested withEciI and AluI restriction endonucleases (New England Biolab, Ipswich,MA, United States), respectively. On the other hand, beef ND5 andbuffalo cytb products were digested with FatI. The total volume ofeach digestion reaction was 25 μL, which was composed of 1 μg ofunpurified PCR product, 1× digestion buffer (supplied with theenzyme), 1U of each enzyme, and a balanced amount of sterilizeddistilled water. The reaction mixtures were gently mixed and spun

down and incubated at 37 °C with EciI and AluI and 55 °C with FatIin a shaking water bath for 60 min. Finally, the digestion reaction wasstopped by heating the reaction mixtures at 65 °C for EciI and 80 °Cfor AluI and FatI for 20 min. The tetraplex PCR products of Cocytb,CoND5, Bucytb, and BuND5 were digested simultaneously in a 25 μLreaction mixture containing 16 μL of unpurified PCR product, 2.5 μLof digestion buffer, 1.5 μL of AluI, 2.5 μL of EciI, and 2.5 μL of FatI.The reaction was mixed by gentle shaking, spun down, and incubatedin a shaking water bath first at 37 °C for 60 min and then at 55 °C for60 min. Enzymatic digestion was stopped by heating the mixture at 80°C for 20 min in a water bath. The digests were separated in anautomated QIAxcel Advanced Capillary Electrophoresis System(QIAGEN GmbH, Hilden, Germany) using a QIAxel DNA High-Resolution Kit (QIAGEN GmbH, Hilden, Germany).

RFLP Analysis of Pork PCR Products. Pork Pocytb and PoND5PCR products were digested with CviKI-1 and FatI restrictionendonucleases (New England Biolab, Ipswich, MA, United States) in aseparate reaction tube of 25 μL reaction volume comprising 1 μg ofunpurified PCR product, 1× digestion buffer supplied with theenzyme, 1U of each enzyme, and a required amount of sterilizeddistilled water. The reaction mixtures were mixed gently and spundown followed by incubation at 37 °C for CviKI-1 and 55 °C for FatIin a shaking water bath for 60 min to digest the targets properly.Postdigested reaction was inactivated by heating the mixtures for 20min at 80 °C for FatI while no inactivation was required for CviKI-1enzyme.

■ RESULTS AND DISCUSSIONQuality and Quantity of Extracted DNA. Total genomic

DNA was extracted from pure, admixed, and meat products(beef, buffalo, and pork frankfurters) under raw and processed(boiled and autoclaved) states. All specimens were made onthree different dates by three independent analysts asdocumented in our earlier report.13 The concentration andpurity of the extracted DNA were determined at ≥100 ng/μLbased on absorbance at 260 nm and absorbance ratio at 260/280 nm, and the lower concentrations were prepared by serialdilution using deionized and nuclease free distilled waterbecause spectrophotometric measurements at low concen-trations were not reproducible. The 260/280 nm absorbance of

Table 2. Sequences of Primers Used in This Study

name species target gene sequence (5′−3′) amplicon size (bp)

Cocytb cow (Bos taurus) Cytb forward: CGGCACAAATTTAGTCGAAT 120reverse: TGGACTATGGCAATTGCTATG

CoND5 cow (Bos taurus) ND5 forward: GGTTTCATTTTAGCAATAGCATGG 106reverse: GTCCAATCAAGGGTATGTTTGAG

Bucytb buffalo (Bubalus bubalis) Cytb forward: GGGTTCTAGCCCTAGTTCTCTCT 90reverse: ATGGCCGGAACATCATACTT

BuND5 buffalo (Bubalus bubalis) ND5 forward: TCGCCTAGCTTCTTACACAAAC 138reverse: TGGTTTGTGACTGTGATGGAT

Pocytb pork (Sus scrofa) Cytb forward: TATCCCTTATATCGGAACAGACCTC 146reverse: GCAGGAATAGGAGATGTACGG

PoND5 pork (Sus scrofa) ND5 forward: GATTCCTAACCCACTCAAACG 73reverse: GGTATGTTTGGGCATTCATTG

Table 3. Concentration of PCR Componentsa

PCRdNTP(mM)

MgCl2(mM)

tag pol(unit)

primer(μM)

duplex and triplex 0.2 2.5 0.94 0.2−0.4tetraplex 0.25 3.5 1.0 0.16−0.4multiplex 0.25 4.0 1.25 0.12−0.6aIn all PCR experiments, 5 μL of 5X GoTaq Flexi Buffer was used.

Table 4. Cycling Parameters of PCR Reactions

35 cycles (40 cycles for multiplex)

PCRreaction

initialdenaturation denaturation annealing extension

finalextension

duplex andtriplex

95 °C for3 min

95 °C for30 s

60 °C for45 s

72 °C for45 s

72 °C for5 min

tetraplex 95 °C for3 min

95 °C for40 s

60 °C for60 s

72 °C for50 s

72 °C for5 min

multiplex 95 °C for5 min

95 °C for50 s

60 °C for90 s

72 °C for50 s

72 °C for7 min

Table 5. Restriction Digests of the PCR Products

target restriction enzyme amplicon size (bp) fragment size (bp)

Cocytb EciI 120 75, 45CoND5 FatI 106 87, 19Bucytb FatI 90 50, 40BuND5 AluI 138 130, 8Pocytb CviKI-1 146 80, 45, 21PoND5 FatI 73 52, 21

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6345

all of the samples was 1.7−2.0, which indicated good qualityDNA in all specimens.28 The amount of DNA extracted fromanimal and fish muscle tissue (20 mg) was 74−152 ng/μL,from plant species (100 mg) was 46−134 ng/μL, and fromfrankfurter (200 mg) was 33−57 ng/μL.PCR Specificity. Species-specific PCR assay is a simple and

low-cost technique that could be performed in mostlaboratories; it is often conclusive, and has been widely usedfor meat speciation. Currently, simplex19,23 and multiplex PCRassays24,29 have been proposed for the authentication ofcommon meat in the food chain, and the development ofeffective primers always plays a key role in the successfulidentification of authentic species.25 Studies demonstrate that

even a single base mismatch at the 3′ end often interferes withPCR efficiency and/or results in amplification failure.6

Considering this pitfall, this report critically evaluated themismatched bases in the primer annealing regions.In this study, six pairs of primers (two pairs of each species)

were designed targeting cytb and ND5 genes of cow, buffalo,and pig species to develop a double gene targeted mPCR assaywith short length of amplicons (Table 2). The designed primersequences were aligned in silico against the similar regions of 29nontarget species including 17 terrestrial animal, 8 fish, and 4plant species, as cited in Design of Species-Specific Primers.Complete sequence matching was found only with cow, buffalo,and pig species, and 3−20 nucleotide (12.5−80%) mismatches

Figure 1. Specificity test of the developed mPCR assay. In the gel images of panels a and b, lane M, DNA ladder; lane N, negative template control;and lane 1, mPCR products of cytb and ND5 of beef, buffalo, and pork. In panel a, lanes 2, 3 and 4, PCR products of cytb and ND5 of beef, buffalo,and pork, respectively; lanes 5−15, PCR products from goat, lamb, dog, cat, rabbit, monkey, donkey, chicken, duck, pigeon, and quail, respectively. Inimage b, lanes 2−15, PCR products from rat, salmon, tuna, cod, tilapia, rohu, pangas, frog, turtle, wheat, onion, garlic, ginger, and pepper,respectively.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6346

were found with other species. The pairwise distance was alsocomputed using the neighbor-joining method;30 the lowestdistance (0.144) was observed between the cow and goatspecies, and the highest (1.993) was found between the cowand wheat species (data not shown). These indicated adequategenetic distances among the studied species, eliminating theprobability of any cross-target detection.12 Moreover, theanalysis of phylogenetic tree demonstrated similar findings,supporting the results of other in silico tests (data not shown).Finally, the theoretical results were experimentally validated

by an authentic PCR test against the target and 27 differentnontarget species using 20 ng of DNA extracted from all of thetested samples. Specific PCR products [106, 138, and 73 bp(ND5 of beef, buffalo, and pork) and 120, 90, and 146 bp (cytbof beef, buffalo, and pork)] were found only from beef, buffalo,and pork, and such a product was absent from the othersamples (goat, lamb, dog, cat, rabbit, monkey, donkey, chicken,duck, pigeon, quail, rat, salmon, tuna, cod, tilapia, rohu, pangas,frog, turtle, wheat, onion, garlic, ginger, and pepper). On theother hand, the use of the universal eukaryotic primers whichamplified 99 bp product from all species reflected the presence

of good quality DNA in all tubes, eliminating the possibility ofany false-negative detection (Figures 1SM−6SM). Afterconfirmation of the simplex PCR, the mPCR system wasdeveloped step by step through the duplex, triplex, andtetraplex and hexaplex (multiplex) PCR systems. The novelmPCR system clearly amplified targeted products (73, 90, 106,120, 138, and 146 bp) from beef, buffalo, and pork samples, andno cross-amplifications were observed in any nontarget species(Figure 1), confirming that the developed mPCR assay washighly specific for the discriminatory detection of beef, buffalo,and pork.Earlier, several simplex7,8,14,31 and multiplex PCR32−34and

PCR-RFLP18,35,20assays were proposed for the authetication ofbeef, buffalo, and pork. However, all of those were based onsingle gene target and longer amplicons (>150 bp) which arebroken down under harsh food processing treatments, leadingto the amplification failure or truncated PCR products,incurring extra cost and compromising reliability.36 In thisstudy, we have developed double gene targeted mPCR assayinvolving short length of the targets (73−146 bp) which arethermodynamically more stable than those of the longer targets.

Figure 2. Gel image (a) and the electropherograms (b−d) of mPCR for the detection of double gene-targeted beef, buffalo, and pork in deliberatelyadulterated model beef, buffalo, and pork frankfurters under raw and processed states. In the gel image, M, Ladder; lanes 1−3, m-PCR of beeffrankfurter spiked with 1%, 0.5%, and 0.1% of buffalo and pork, respectively, under raw state; lanes 5−7, mPCR of buffalo frankfurter spiked with 1%,0.5%, and 0.1% of beef and pork, respectively, under raw state; lanes 9−11, mPCR of pork frankfurter spiked with 1%, 0.5%, and 0.1% of beef andbuffalo, respectively, under raw state; lanes 4, 8, and 12, mPCR of heat-treated (autoclaved for 2.5 h) 0.1% adulterated beef, buffalo, and porkfrankfurter, respectively; lane N, negative control. The corresponding electroferograms of lane 4, 8, and 12 are shown labeled as b, c, and d,respectively.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6347

Figure 3. RFLP analysis of simplex and mPCR products before (lanes 1, 3, 5, 7, and 9) and after (lanes 2, 4, 6, 8, and 10) restriction digestion. In thegel image, lanes 1 and 2, cytb of buffalo; lanes 3 and 4, ND5 of beef; lanes 5 and 6, cytb of beef; lanes 7 and 8, ND5 of buffalo; and lanes 9 and 10,mPCR of cytb and ND5 of beef and buffalo. Corresponding electropherograms are shown with labels.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6348

Because of the presence of double targets for each species, thisnovel assay could complement the detection of a missing targetbecause it is highly unlikely that both gene sites would be lostunder the states of decomposition. To the best of ourknowledge, this is the first report of a double gene targetedmPCR assay for the differential identification of beef, buffalo,and pork.Specificity and Sensitivity under Complex Food

Matrices. Common forms of meat adulteration take place inminced meat products such as frankfurters, meatballs, andburgers.25,37 Therefore, we evaluated the performance of thedeveloped mPCR assay for the screening of beef, buffalo, andpork materials in commercial frankfurters which are verypopular all over the world.25 To simulate the commercialmatrices, beef, buffalo, and pork frankfurters were made in thelaboratory and were deliberately adulterated with 10%, 1%, and0.1% raw meat of two other target species as described inMaterials and Methods. The 0.1% spiked frankfurters of threespecies were autoclaved at 121 °C and 15 psi for 2.5 h tosimulate extensive cooking effect.36 The model beef, buffalo,and pork frankfurters, adulterated with 1%, 0.5%, and 0.1% ofbuffalo and pork, beef and pork, and beef and buffalo, amplifiedall the six targets (Figure 2; lanes 1−3, 5−7, and 9−11)representing all three target species. The 0.1% adulteratedautoclaved frankfurters also positively amplified six targets forbeef, buffalo, and pork (lanes 4, 8, and 12), reflecting thesensitivity and discriminatory attributes of the novel PCR assay.Previously, Razzak et al.26 detected 0.1% porcine, canine,

feline, monkey, and rat meat under mixed food matrices using apentaplex PCR assay where the amplicon size ranged from 108to 172 bp. Safdar et al.29 also reported a 0.1% limit of detection(LOD) for the identification of ovine, caprine, fish, and bovinematerial using a tetraplex PCR assay involving 119−271 bpamplicons in heat-treated (133 °C at 300 kPa for 20 min)mixed meat. In another report, Safdar et al.38 documented0.01% LOD for the identification of horse, soybean, poultry andpork with 85−212 bp amplicon targets. However, instead ofusing processed samples, they used raw meat. Earlier, we havescientifically proven that the stability of the PCR assay underextensive processing atmosphere largely depends on theamplicon sizes; longer targets break down before the shorterones.6,23 This study has carefully addressed this point and kept

amplicon lengths between 73 and 146 bp; additionally, doublegene sites were used as targets for each species to complementa potential missing target. Therefore, this novel mPCR assayoffered better reliability but equivalent sensitivity compared tothose of other published reports.

Authentication by RFLP. Species-specific PCR assay isoften conclusive,24 but it has yet to be considered a definitiveanalytical method because of certain “hard-to-control” featuresof the amplification process.23,39 For example, it sometimesproduces artifacts due to contamination by alien DNA at aminute scale,39,40 but these ambiguities or doubts could beeliminated by the verification of the amplified product throughat least one of three different methods, namely, PCR-RFLPassay, probe hybridization, and target product sequencing.41

Probe hybridization is an attractive technique because it candetect multiple species in a single experimental run through theuse of multiple labeled probes,42 but this procedure requirespurified DNA and is also laborious, expensive, and time-consuming.6 In contrast, DNA sequencing is a more efficientand reliable tool, but it requires an expensive laboratory setupand is often not suitable for the analysis of processed foodunder complex matrices43,44 because of the coextraction of thefood ingredients that often bring errors into the final results.45

In contrast, the PCR-RFLP assay can overcome all of theselimitations and has been widely used to authenticate theoriginal PCR product amplified from a particular genefragment.46,47 It comprises the generations of a specificfragment profile through restriction digestion with one ortwo endonucleases. A carefully selected restriction endonu-clease cleaves the PCR product at specific recognition sites,producing a set of DNA fragments of different lengths thatcould be separated and visualized by gel electrophoresis;48 thus,it distinguishes the artificial PCR product from the originalthrough the analysis of the restriction fingerprints.40,49

In this study, the tetraplex PCR products of beef and buffalowere digested simultaneously with three restriction enzymes ascited in Materials and Methods, and clear fingerprints wereobtained for each of the four different targets (Figure 3 andTable 5). First, each target was digested separately with anappropriate RE (Table 5) to study its individual restrictionpatterns in order to eliminate any ambiguities that may arisefrom the final tetraplex PCR products that were the mixture of

Figure 4. PCR-RFLP analysis of mPCR products of deliberately adulterated raw and heat-treated (boiled and autoclaved) beef (lanes 1−6) andbuffalo (lanes 7−12) frankfurters. In gel image, lanes 1 and 2, buffalo-adulterated raw beef frankfurter before and after digestion, respectively; lanes 3and 4, buffalo-adulterated boiled (98 °C for 90 min) beef frankfurter before and after digestion, respectively; lanes 5 and 6, buffalo-adulteratedautoclaved (121 °C and 15 psi pressure for 2.5 h) beef frankfurter before and after digestion, respectively; lanes 7 and 8, beef-adulterated raw buffalofrankfurter before and after digestion, respectively; lanes 9 and 10, beef-adulterated boiled (98 °C for 90 min) buffalo frankfurter before and afterdigestion, respectively; lanes 11 and 12, beef-adulterated autoclaved (121 °C and 15 psi pressure for 2.5 h) buffalo frankfurter before and afterdigestion, respectively.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6349

four different amplicons (Figure 3). Both buffalo cytb (90 bp)(Figure 3, lane 1) and beef ND5 (106 bp) (Figure 3, lane 3)products were digested by FatI RE, which generated twofragments for each target (50 and 40 bp for buffalo cytb (lane2) and 87 and 19 bp for beef ND5 (lane 4)). On the otherhand, beef cytb (120 bp) (lane 5) was digested by EciI thatproduced two fragments (75 and 45 bp) (lane 6). In contrast,buffalo ND5 product (lane 7) was digested with AluI, whichresulted in another two fragments (130 and 8 bp) (lane 8).

However, 8 bp fragment was not detected because it wentbeyond the lower limit of instrumental resolution, which was≤15 bp. Finally, the mPCR products (lane 9) were subjected toRE digestion with the three enzymes (FatI, EciI, and AluI) in asingle tube, and this generated molecular fingerprints whichwere composed of a total of seven fragments (19, 40, 45, 50, 75,87, and 130) (lane 10). The origins of these products (lane 9)were confirmed by the separate digests of the four targets (lanes1−8).

Figure 5. PCR-RFLP analysis of simplex PCR products of pork PoND5 and Pocytb before and after restriction endonuclease digestion. In the gelimage, lanes 1 and 2, PCR products of PoND5 before and after digestion; lanes 3 and 4, products of Pocytb before and after digestion, respectively.Corresponding electropherograms are indicated by corresponding labels.

Figure 6. RFLP analysis of pork PoND5 (lanes 1−6) and Pocytb (lanes 7−12) PCR products before (lanes 1, 3, 5, 7, 9, and 11) and after (lanes 2, 4,6, 8, 10, and 12) restriction digestion. In the gel view, PCR products from raw (lanes 1, 2, 7, and 8), boiled (lanes 3, 4, 9, and 10), and autoclaved(lanes 5, 6, 11, and 12) pork frankfurter; lane M, DNA ladder.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6350

After the mPCR-RFLP assay under pure states wasoptimized, it was subsequently optimized and evaluated forthe screening of commercial beef and buffalo frankfurters underraw, boiled, and autoclaved states.12 Dummy frankfurters weredeliberately adulterated, and their restriction digestion patternswere studied (Figure 4). The digest of all samples (lanes 1, 3, 5,7, 9, and 11) clearly presented the signature fingerprints of 7fragments (lanes 2, 4, 6, 8, 10, and 12), reflecting that variationsin food processing treatments cannot affect the stability of anyof the four biomarkers developed in this study; in other words,this novel mPCR-RFLP assay was sensitive, reliable, and robustfor the discriminatory detection of beef and buffalo inprocessed foods.Simplex PCR products of pork Pocytb and PoND5 were

digested individually with CviKI-1 and FatI RE, respectively,because in silico studies demonstrated overlapping fragmentswith beef and buffalo. Postdigested PoND5 PCR product (73bp) (Figure 5, lane 1) produced 2 fragments of 52 and 21 bp(Figure 5, lane 2), and Pocytb PCR product (146 bp) (lane 3)generated 3 fragments of 80, 45, and 21 bp (lane 4). Similarproducts were found from boiled (98 °C for 90 min) andautoclaved (121 °C at 45 psi for 2.5 h) pork frankfurters. Therestriction digestion maps of different heat-treated (boiled andautoclaved) samples were similar to those from the raw sample(Figure 6).Previously, Haider et al.1 reported a PCR-RFLP assay with a

710 bp amplicon that was amplified using common primer pairsfor the cow, chicken, turkey, sheep, pig, buffalo, camel, anddonkey. Girish et al.20 also documented a PCR-RFLP assaywith 456 bp amplicon length for the detection of Goat, Sheep,Cattle and Buffalo. Recently, Kumar et al.50 proposed a RFLPpattern with a 609 bp target to discriminate cattle, buffalo, goat,sheep and pig. In addition, Erwanto et al.51 demonstrated aPCR-RFLP technique for a 359 bp product. However, suchlong targets (359−710 bp) are more prone to break down andthus would definitely lose their applicability for the analysis ofprocessed foods. In contrast, here we reported a double genesite and short amplicon length (≤146 bp) mPCR-RFLP andsystematically proved its reliability and sensitivity under raw,

boiled (98 °C for 90 min), and autoclaved (121 °C and 15 psipressure for 2.5 h) atmospheres for differential identification ofbeef, buffalo, and pork in pure, admixed, and frankfurterformulation.

Analysis of Commercial Frankfurters. The motivation ofthe substitution of an expensive meat with its cheapercounterpart comes with the inclination of a company to havemore sales and better profit, and instead of raw meat,adulteration could be skillfully manipulated in processed meatproducts.11 Because frankfurter is very popular and consumedwidely all over the world, we have screened 20 halal brandedbeef frankfurters in Malaysian markets (Table 6). It would benoteworthy here that no buffalo frankfurter products werefound in the Malaysian markets; that is, all were labeled as beefproducts. However, all the tested beef frankfurters were foundas both beef and buffalo positive; this indicated that all beeffrankfurter products in Malaysia was buffalo adulterated. Wealso checked chicken and pork frankfurters, but none of themwere beef and buffalo positive; this was probably because theprices of beef and buffalo are higher than those of chicken andpork. Although several PCR assays are proposed for the beefand buffalo differentiation,8,33 none of them were tested undercommercial matrices despite having the risk of PCR inhibitionby multiple ingredients present in commercial products.52,53

Previous reports analyzed only model meat products such askabab, patty, and meat block using simplex PCR systems forbeef and buffalo, which incurs additional cost and time due tothe use of separate assays for each species.15,54 Although severalreports were documented for the analysis of meatball, streakybacon, frankfurter, and burger model products for theidentification of pig species,51,55 all of those were simplexPCR assays. On the other hand, the novel mPCR-RFLP assaywe reported here for frankfurter analysis was more reliable andconfirmatory but less expensive because it discriminated beef,buffalo, and pork with double targets in a single assay platform.

Public Health, Social, and Economic Implications ofthe Study. Beef, buffalo, and pork adulterated meat productshave direct implications for public health, religion, culture, andeconomy. A confirmatory low-cost analytical test involving all

Table 6. Screening of Model and Commercial Frankfurters by the Multiplex PCR Assaya

adulteration detected species

sample (frankfurter) species % state beef buffalo pork PCR accuracy (%)

Model Frankfurterbeef buffalo and pork 1.0 raw 9/9 9/9 9/9 100beef buffalo and pork 0.5 raw 9/9 9/9 9/9 100beef buffalo and pork 0.1 raw 9/9 9/9 9/9 100beef buffalo and pork 0.1 autoclaved for 2.5 h 9/9 9/9 9/9 100buffalo beef and pork 1.0 raw 9/9 9/9 9/9 100buffalo beef and pork 0.5 raw 9/9 9/9 9/9 100buffalo beef and pig 0.1 raw 9/9 9/9 9/9 100buffalo beef and pork 0.1 autoclaved for 2.5 h 9/9 9/9 9/9 100pork beef and buffalo 1.0 raw 9/9 9/9 9/9 100pork beef and buffalo 0.5 raw 9/9 9/9 9/9 100pork beef and buffalo 0.1 raw 9/9 9/9 9/9 100pork beef and buffalo 0.1 autoclaved for 2.5 h 9/9 9/9 9/9 100

Commercial Frankfurterbeef − − raw 20/20 20/20 0/20 100pork − − raw 0/10 0/10 10/10 100chicken − − raw 0/10 0/10 0/10 100

aThe numerator and denominator of each fraction denote the number of positive detection and total number of samples analyzed using the mPCRassay.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6351

three species definitely can help market regulation, preventingor at least reducing adulteration events to a great extent. Theshort amplicon length and double-gene targeting mPCR-RFLPsystem that is documented here is greatly reliable for beef,buffalo, and pork identification in a single assay platformbecause of (1) the alternative targets which can complementthe detection of a missing target for each species, (2) theshorter length of the targets which offer better stability evenunder the state of decomposition, and (3) the opportunity ofrechecking the product authenticity by distinctive molecularfingerprints of the RFLP digests using four different restrictionenzymes. The use of an internal positive control just eliminatedthe chances of any false negative detection. Species specificity ofall targets was confirmed by cross-checking all the primersagainst 27 nontarget species. The stability of the assay wasfurther qualified under various cooking treatments, includingextensive autoclaving (121 °C and 15 psi pressure for 2.5 h)that breaks down DNA. Finally, it was found sensitive enoughto detect all the beef, buffalo, and pork targets in raw andprocessed frankfurter products with as low as 0.1% adulteration.Thus, the novel assay demonstrated sufficient merits to be usedby regulatory bodies for beef, buffalo, and pork authenticationeven in degraded specimens.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.jafc.6b02224.

Figure 1SM: Specificity test of the simplex PCR of beefcytb (120 bp)-specific primer pair with DNA of differentspecies (PDF)Figure 2SM: Specificity test of the simplex PCR of beefND5 (106 bp)-specific primer pair with DNA of differentspecies (PDF)Figure 3SM: Specificity test of the simplex PCR ofbuffalo cytb (90 bp)-specific primer pair with DNA ofdifferent species (PDF)Figure 4SM: Specificity of the simplex PCR of buffaloND5 (138 bp)-specific primer pair with DNA of differentspecies (PDF)Figure 5SM: Specificity of the simplex PCR of pork cytb(146 bp)-specific primer pair with DNA of differentspecies (PDF)Figure 6SM: Specificity of the simplex PCR of pork ND5(73 bp)-specific primer pair with DNA of differentspecies (PDF)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: eaqubali@gmail.com; eaqubali@um.edu.my. Tel:+603-7967-6959. Fax: +603-7967-6956.FundingThis work was supported by University of Malaya ResearchGrant No. GC001-14SBS to M.E.A.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors acknowledge the kind gift of monkey, dog, cat, andrat meat samples from the wildlife Malaysia and DewanBandaraya Kuala Lumpur (DBKL).

■ ABBREVIATIONS USED

Cytb, cytochrome b; ND5, NADH dehydrogenase sub unit 5;DNA, deoxyribonucleic acid; RE, restriction enzyme; PCR,polymerase chain reaction; RFLP, restriction fragment lengthpolymorphism; mPCR, multiplex PCR; BLAST, basic localalignment algorithm search tool; NCBI, National Center forBiotechnology Information; bp, base pair

■ REFERENCES(1) Haider, N.; Nabulsi, I.; Al-Safadi, B. Identification of meat speciesby PCR-RFLP of the mitochondrial COI gene. Meat Sci. 2012, 90,490−3.(2) Bottero, M. T.; Dalmasso, A. Animal species identification in foodproducts: Evolution of biomolecular methods. Vet. J. 2011, 190, 34−38.(3) Walker, M. J.; Burns, M.; Burns, D. T. Horse Meat in BeefProducts-Species Substitution 2013. J. Assoc of Public Anal. (Online)2013, 41, 67−106.(4) Russia says wrong animal DNA found in Auchan minced meat.http://www.nst.com.my/news/2015/09/russia-says-wrong-animal-dna-found-auchan-minced-meat (accessed January 1, 2016), NewsStraits Times, 2015.(5) Ali, M. E.; Razzak, M. A.; Hamid, S. B. A. Multiplex PCR inSpecies Authentication: Probability and ProspectsA Review. FoodAnalytical Methods 2014, 7, 1933−1949.(6) Rashid, N. R.; Ali, M. E.; Hamid, S. B.; Rahman, M. M.; Razzak,M. A.; Asing; Amin, M. A. A suitable method for the detection of apotential fraud of bringing macaque monkey meat into the food chain.Food Addit. Contam., Part A 2015, 32, 1013−1022.(7) Girish, P.; Haunshi, S.; Vaithiyanathan, S.; Rajitha, R.;Ramakrishna, C. A rapid method for authentication of Buffalo(Bubalus bubalis) meat by Alkaline Lysis method of DNA extractionand species specific polymerase chain reaction. J. Food Sci. Technol.2013, 50, 141−146.(8) Karabasanavar, N. S.; Singh, S. P.; Umapathi, V.; Girish, P. S.;Shebannavar, S. N.; Kumar, D. Authentication of carabeef (waterbuffalo, Bubalus bubalis) using highly specific polymerase chainreaction. Eur. Food Res. Technol. 2011, 233, 985−989.(9) Sakaridis, I.; Ganopoulos, I.; Argiriou, A.; Tsaftaris, A. A fast andaccurate method for controlling the correct labeling of productscontaining buffalo meat using High Resolution Melting (HRM)analysis. Meat Sci. 2013, 94, 84−8.(10) Ali, M. E.; Hashim, U.; Mustafa, S.; Man, Y. C.; Dhahi, T. S.;Kashif, M.; Uddin, M. K.; Hamid, S. B. A. Analysis of pork adulterationin commercial meatballs targeting porcine-specific mitochondrialcytochrome b gene by TaqMan probe real-time polymerase chainreaction. Meat Sci. 2012, 91, 454−459.(11) von Bargen, C.; Dojahn, J.; Waidelich, D.; Humpf, H. U.;Brockmeyer, J. New sensitive highperformance liquid chromatog-raphy−tandem mass spectrometry method for the detection of horseand pork in halal beef. J. Agric. Food Chem. 2013, 61, 11986−11994.(12) Taboada, L.; Sanchez, A.; Velasco, A.; Santaclara, F. J.; Perez-Martín, R. I.; Sotelo, C. G. Identification of Atlantic Cod (Gadusmorhua), Ling (Molva molva), and Alaska Pollock (Gaduschalcogrammus) by PCR−ELISA Using Duplex PCR. J. Agric. FoodChem. 2014, 62, 5699−5706.(13) Ali, M. E.; Al Amin, M.; Hamid, S. B. A.; Hossain, M. M.;Mustafa, S. Lab-on-a-chip-based PCR-RFLP assay for the confirmeddetection of short-length feline DNA in food. Food Addit. Contam.,Part A 2015, 32, 1373−1383.(14) Lopez, I.; Pardo, M. A. Application of relative quantificationTaqMan real-time polymerase chain reaction technology for theidentification and quantification of Thunnus alalunga and Thunnusalbacares. J. Agric. Food Chem. 2005, 53, 4554−4560.(15) Mane, B. G.; Mendiratta, S. K.; Tiwari, A. K. Beef specificpolymerase chain reaction assay for authentication of meat and meatproducts. Food Control 2012, 28, 246−249.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6352

(16) Giusti, A.; Castigliego, L.; Rubino, R.; Gianfaldoni, D.; Guidi, A.;Armani, A. A Conventional Multiplex PCR assay for the detection oftoxic Gemfish species (Ruvettus pretiosus and Lepidocybiumflavobrunneum): a simple method to combat health frauds. J. Agric.Food Chem. 2016, 64 (4), 960−968.(17) Cook, D.; Pfister, J. A.; Constantino, J. R.; Roper, J. M.;Gardner, D. R.; Welch, K. D.; Hammond, Z. J.; Green, B. T.Development of a PCR-based method for detection of Delphiniumspecies in poisoned cattle. J. Agric. Food Chem. 2015, 63, 1220−1225.(18) Hsieh, Y.-W.; Hwang, D.-F. Molecular phylogenetic relation-ships of puffer fish inferred from partial sequences of cytochrome bgene and restriction fragment length polymorphism analysis. J. Agric.Food Chem. 2004, 52, 4159−4165.(19) Chen, S.-Y.; Liu, Y.-P.; Yao, Y.-G. Species authentication ofcommercial beef jerky based on PCR-RFLP analysis of themitochondrial 12S rRNA gene. J. Genet. Genomics 2010, 37, 763−769.(20) Girish, P. S.; Anjaneyulu, A. S.; Viswas, K. N.; Shivakumar, B.M.; Anand, M.; Patel, M.; Sharma, B. Meat species identification bypolymerase chain reaction-restriction fragment length polymorphism(PCR-RFLP) of mitochondrial 12S rRNA gene. Meat Sci. 2005, 70,107−12.(21) Mutalib, S. A.; Nazri, W. S. W.; Shahimi, S.; Yaakob, N.; Sani, N.A.; Abdullah, A.; Babji, A. l. S.; Ghani, M. A. Comparison betweenpork and wild boar meat (Sus scrofa) by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). Sains Malays.2012, 41, 199−204.(22) Nebola, M.; Borilova, G.; Kasalova, J. PCR-RFLP analysis ofDNA for the differentiation of fish species in seafood samples. Bull.Vet. Inst. Pulawy 2010, 54, 49−53.(23) Focke, F.; Haase, I.; Fischer, M. DNA-based identification ofspices: DNA isolation, whole genome amplification, and polymerasechain reaction. J. Agric. Food Chem. 2011, 59, 513−520.(24) Ali, M. E.; Asing; Hamid, S. B. A.; Razzak, M. A.; Rashid, N. R.A.; Amin, M. A.; Mustafa, S. A suitable method to detect potentialfraud of bringing Malayan box turtle (Cuora amboinensis) meat intothe food chain. Food Addit. Contam., Part A 2015, 32, 1223−1233.(25) Ali, M. E.; Razzak, M. A.; Hamid, S. B. A.; Rahman, M. M.;Amin, M. A.; Rashid, N. R. A. Asing, Multiplex PCR assay for thedetection of five meat species forbidden in Islamic foods. Food Chem.2015, 177, 214−224.(26) Razzak, M. A.; Hamid, S. B. A.; Ali, M. E. A lab-on-a-chip-basedmultiplex platform to detect potential fraud of introducing pig, dog,cat, rat and monkey meat into the food chain. Food Addit. Contam.,Part A 2015, 32, 1902−1913.(27) Rahman, M. M.; Ali, M. E.; Abd Hamid, S. B.; Mustafa, S.;Hashim, U.; Hanapi, U. K. Polymerase chain reaction assay targetingcytochrome b gene for the detection of dog meat adulteration inmeatball formulation. Meat Sci. 2014, 97, 404−409.(28) Herrero, B.; Vieites, J. M.; Espineira, M. Fast real-time PCR forthe detection of crustacean allergen in foods. J. Agric. Food Chem.2012, 60, 1893−1897.(29) Safdar, M.; Junejo, Y. A multiplex-conventional PCR assay forbovine, ovine, caprine and fish species identification in feedstuffs:Highly sensitive and specific. Food Control 2015, 50, 190−194.(30) Tamura, K.; Peterson, D.; Peterson, N.; Stecher, G.; Nei, M.;Kumar, S. MEGA5: molecular evolutionary genetics analysis usingmaximum likelihood, evolutionary distance, and maximum parsimonymethods. Mol. Biol. Evol. 2011, 28, 2731−2739.(31) Che Man, Y.; Aida, A.; Raha, A.; Son, R. Identification of porkderivatives in food products by species-specific polymerase chainreaction (PCR) for halal verification. Food Control 2007, 18, 885−889.(32) Rea, S.; Chikuni, K.; Branciari, R.; Sangamayya, R. S.; Ranucci,D.; Avellini, P. Use of duplex polymerase chain reaction (duplex-PCR)technique to identify bovine and water buffalo milk used in makingmozzarella cheese. J. Dairy Res. 2001, 68, 689−698.(33) Gupta, R.; Rank, D. N.; Joshi, C. G. Duplex-PCR forIdentification and Differentiation of Cattle and Buffalo ProcessedMeat. J. Adv. Vet. Res. 2011, 1, 13−16.

(34) He, H.; Hong, X.; Feng, Y.; Wang, Y.; Ying, J.; Liu, Q.; Qian, Y.;Zhou, X.; Wang, D. Application of Quadruple Multiplex PCRDetection for Beef, Duck, Mutton and Pork in Mixed Meat. Journalof Food and Nutrition Research 2015, 3, 392−398.(35) Verkaar, E. L. C.; Nijman, I. J.; Boutaga, K.; Lenstra, J. A.Differentiation of cattle species in beef by PCR-RFLP of mitochondrialand satellite DNA. Meat Sci. 2002, 60, 365−369.(36) Ali, M. E.; Al Amin, M.; Razzak, M. A.; Hamid, S. B. A.;Rahman, M. M.; Rashid, N. A. Short Amplicon-Length PCR AssayTargeting Mitochondrial Cytochrome b Gene for the Detection ofFeline Meats in Burger Formulation. Food Anal. Methods 2016, 9,571−581.(37) Rahman, M. M.; Ali, M. E.; Hamid, S. B. A.; Bhassu, S.; Mustafa,S.; Al Amin, M.; Razzak, M. A. Lab-on-a-Chip PCR-RFLP Assay forthe Detection of Canine DNA in Burger Formulations. Food AnalyticalMethods 2015, 8, 1598−1606.(38) Safdar, M.; Junejo, Y.; Arman, K.; Abasiyanik, M. F. A highlysensitive and specific tetraplex PCR assay for soybean, poultry, horseand pork species identification in sausages: development andvalidation. Meat Sci. 2014, 98, 296−300.(39) Yang, I.; Kim, Y.-H.; Byun, J.-Y.; Park, S.-R. Use of multiplexpolymerase chain reactions to indicate the accuracy of the annealingtemperature of thermal cycling. Anal. Biochem. 2005, 338, 192−200.(40) Doosti, A.; Dehkordi, P. G.; Rahimi, E. Molecular assay to fraudidentification of meat products. J. Food Sci. Technol. 2014, 51, 148−152.(41) Maede, D. A strategy for molecular species detection in meatand meat products by PCR-RFLP and DNA sequencing usingmitochondrial and chromosomal genetic sequences. Eur. Food Res.Technol. 2006, 224, 209−217.(42) do Nascimento, C.; de Albuquerque, R. F.; Monesi, N.;Candido-Silva, J. A. Alternative method for direct DNA probe labelingand detection using the checkerboard hybridization format. J. ClinMicrobiol. 2010, 48, 3039−3040.(43) Girish, P.; Anjaneyulu, A.; Viswas, K.; Anand, M.; Rajkumar, N.;Shivakumar, B.; Bhaskar, S. Sequence analysis of mitochondrial 12SrRNA gene can identify meat species. Meat Sci. 2004, 66, 551−556.(44) Mafra, I.; Ferreira, I. M. P. L. V. O.; Oliveira, M. B. P. P. Foodauthentication by PCR-based methods. Eur. Food Res. Technol. 2008,227, 649−665.(45) Albers, C. N.; Jensen, A.; Bælum, J.; Jacobsen, C. S. Inhibition ofDNA polymerases used in Q-PCR by structurally different soil-derivedhumic substances. Geomicrobiol. J. 2013, 30, 675−681.(46) Park, J.-K.; Shin, K.-H.; Shin, S.-C.; Chung, K.-Y.; Chung, E.-R.Identification of meat species using species-specific PCR-RFLPfingerprint of mitochondrial 12S rRNA gene. Korean Journal forFood Science of Animal Resources 2007, 27, 209−215.(47) Sharma, N.; Thind, S.; Girish, P.; Sharma, D. PCR-RFLP of 12SrRNA gene for meat speciation. J. Food Sci. Tech. Mys. 2008, 45, 353−355.(48) Ballin, N. Z.; Vogensen, F. K.; Karlsson, A. H. Speciesdetermination−Can we detect and quantify meat adulteration? MeatSci. 2009, 83, 165−174.(49) Murugaiah, C.; Noor, Z. M.; Mastakim, M.; Bilung, L. M.;Selamat, J.; Radu, S. Meat species identification and Halalauthentication analysis using mitochondrial DNA. Meat Sci. 2009,83, 57−61.(50) Kumar, D.; Singh, S.; Karabasanavar, N. S.; Singh, R.; Umapathi,V. Authentication of beef, carabeef, chevon, mutton and pork by aPCR-RFLP assay of mitochondrial cytb gene. J. Food Sci. Technol.2014, 51, 3458−3463.(51) Erwanto, Y.; Abidin, M. Z.; S; Rohman, A. Pig speciesidentification in meatballs using polymerase chain reaction- restrictionfragment length polymorphism for Halal authentication. Int. Food Res.J. 2012, 19, 901−906.(52) Bottero, M.; Civera, T.; Anastasio, A.; Turi, R.; Rosati, S.Identification of cow’s milk in “Buffalo” cheese by Duplex PolymersaseChain Reaction. J. Food Prot. 2002, 65, 362−366.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6353

(53) Di Pinto, A.; Forte, V.; Conversano, M.; Tantillo, G. Duplexpolymerase chain reaction for detection of pork meat in horse meatfresh sausages from Italian retail sources. Food Control 2005, 16, 391−394.(54) Mane, S. K.; Mendiratta, A. K.; Tiwari; Bhilegaokar, K. N.Detection of Adulteration of Meat and Meat Products with BuffaloMeat Employing Polymerase Chain Reaction Assay. Food Anal.Methods 2012, 5, 296−300.(55) Ali, M. E.; Hashim, U.; Mustafa, S.; Man, Y. B. C. Swine-SpecificPCR-RFLP Assay Targeting Mitochondrial Cytochrome B Gene forSemiquantitative Detection of Pork in Commercial Meat Products.Food Anal. Methods 2012, 5, 613−623.

Journal of Agricultural and Food Chemistry Article

DOI: 10.1021/acs.jafc.6b02224J. Agric. Food Chem. 2016, 64, 6343−6354

6354