Meat Science 51 (1999) 143±148

A quick and simple method for the identi®cation of meat species

and meat products by PCR assay
T. Matsunaga a, K. Chikuni b*, R. Tanabe b, S. Muroya b, K. Shibata a, J. Yamada a,
Y. Shinmura a
a
Japan Meat Processors Association, Ebisu 1-5-6 Shibuya-ku, Tokyo 150, Japan
b
National Institute of Animal Industry, Tsukuba Norindanchi, PO Box 5, Ibaraki 305, Japan

Received 28 August 1997; received in revised form 30 June 1998; accepted 2 July 1998

Abstract
The polymerase chain reaction (PCR) was applied to identify six meats (cattle, pig, chicken, sheep, goat and horse) as raw
materials for products. By mixing seven primers in appropriate ratios, species-speci®c DNA fragments could be identi®ed by only
one multiplex PCR. A forward primer was designed on a conserved DNA sequence in the mitochondrial cytochrome b gene, and
reverse primers on species-speci®c DNA sequences for each species. PCR primers were designed to give di€erent length fragments
from the six meats. The products showed species-speci®c DNA fragments of 157, 227, 274, 331, 398 and 439 bp from goat, chicken,
cattle, sheep, pig and horse meats, respectively. Identi®cation is possible by electrophoresis of PCR products. Cattle, pig, chicken,
sheep and goat fragments were ampli®ed from cooked meat heated at 100 or 120 C for 30 min, but horse DNA fragments could not
be detected from the 120 C sample. Detection limits of the DNA samples were 0.25 ng for all meats. # 1998 Elsevier Science Ltd.
All rights reserved.
Keywords: PCR; Meat species identi®cation

1. Introduction DNA sequences and identi®ed cattle, pig and chicken

meats. In the present study, the authors developed a
Accurate analytical methods are indispensable for the simple method using multiplex PCR for simultaneous
labeling of meat products, and that requires simple and identi®cation of six meats.
fast procedures. DNA hybridization (Baur et al., 1987;
Chikuni et al., 1990; Winterù et al., 1990; Ebbehùj and
Thomsen, 1991a, b) and PCR methods (Chikuni et al., 2. Methods
1994a, b; Fei et al., 1996) have been used for identi®ca-
tion of meats and meat products. DNA hybridization 2.1. DNA extraction
methods are complicated and generally inadequate, but
PCR easily ampli®es target regions of template DNA in All meat samples were obtained from commercial
a much shorter time (Saiki et al., 1985), and thus is sui- sources. DNA was prepared from cattle, pig, chicken,
table for meat identi®cation. Chikuni et al. (1994a, b) sheep, goat and horse meats as described by Sambrook
distinguished sheep, goat and cattle meats using a satel- et al. (1989). DNA/RNA mixtures were extracted in 20
lite DNA sequence and eight mammals and ®ve birds vol of 100 mM Tris±HCl, (pH 9.0) containing 100 mM
using the cytochrome b sequence. That method con- NaCl, 5 mM EDTA and 1% SDS, for 30 min at the
sisted of PCR ampli®cation followed by restriction room temperature. The DNA/RNA solutions were
digestions, thus the procedures for mixed meats or meat extracted with an equal volume of phenol/chloroform/
products were complicated. Fei et al. (1996) designed isoamyl-alcohol (25:24:1), and then with an equal
multiplex PCR primers based on mitochondrial D-loop volume of chloroform. RNA in the DNA/RNA solu-
tion was degraded by 100 mg mlÿ1 Ribonuclease A
(Sigma) for 1 h at 37 C. DNA was extracted twice with
* Corresponding author. E-mail: chikunc@niai.a€rc.go.jp an equal volume of phenol/chloroform/isoamyl-alcohol

0309-1740/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0309 -1 740(98)00112 -0
 144
T. Matsunaga et al./Meat Science 51 (1999) 143±148
Fig. 1. Nucleotide sequences of the primers and target region on cytochrome b gene. Open boxes indicate the common forward primer SIM and complementary sequences of species-speci®c reverse
primer. Dots and closed boxes indicate identical and di€erent nucleotides to the primer sequences, respectively.
 T. Matsunaga et al./Meat Science 51 (1999) 143±148 145

and once with an equal volume of chloroform. DNA on 38 bp sequence. Because each 1% mismatching of
concentrated by ethanol precipitation was dissolved in bases in a double-stranded (ds) DNA reduces melting
10 mM Tris±HCl, (pH 7.4) 0.1 mM EDTA for use as temperature (Tm) by 1±1.5 C (Sambrook et al., 1989),
the PCR template. DNA concentration was determined the mismatches of SIM would decrease the Tm of pri-
based on absorbance at 260 nm. Each piece of 25 g from mer-template dsDNA about 10±15 C. Thus, SIM was
cattle, pig, chicken, sheep, goat and horse meats was designed as longer than species-speci®c primers that
heated for 30 min at 100 or 120 C. The DNA/RNA were 26±29 nucleotides long. The reverse primers were
mixture was extracted from 500 mg of heated meat with designed on species-speci®c regions. The primers were
10ml of 100 mM Tris±HCl, (pH 9.0) containing 1% expected to amplify target sequences at the same e-
SDS and 5 mM DTT for 5 min at 95 C. The DNA ciency in multiplex PCR. Fig. 2 shows 4% agarose gel
solution was puri®ed by phenol/chloroform/isoamyl- electrophoresis of PCR products ampli®ed from the six
alcohol extraction. species. PCR products from goat, chicken, cattle, sheep,
pig and horse DNAs were single DNA fragments of
2.2. Polymerase chain reaction (PCR) 157, 227, 274, 331, 398 and 439 bp, respectively. The six
meats could thus be identi®ed based on the length of
PCR ampli®cation was conducted in 50 ml of 10 mM PCR products with no cross reaction.
Tris±HCl, (pH 8.3) containing 50 mM KCl, 1.5 mM
MgCl2, 200 mM dNTP mix, primer mix (4±60 pmol 3.2. Identi®cation of cooked meat
each), 250 ng template DNA and 1.25 unit Taq DNA
polymerase (Perkin±Elmer). Oligonucleotide primers Cooked samples were obtained from the six meats
were the common forward primer SIM (50 -GACCTC- heated for 30 min at 100 or 120 C, and DNA extracted
CCAGCTCCATCAAACATCTCATCTTGATGAAA- from 500 mg of the meats was used as a template for
30 ) and reverse primers [goat primer G (50 -CTCGA- PCR. The PCR products were analyzed by 4% agarose
CAAATGTGAGTTACAGAGGGA-30 ), chicken primer gel electophoresis. Fig. 3 shows the species-speci®c
C (50 -AAGATACAGATGAAGAAGAATGAGGCG- DNA fragments to have been ampli®ed from the
30 ), cattle primer B (50 -CTAGAAAAGTGTAAGA- cooked meat heated at 100 and 120 C except for horse
CCCGTAATATAAG-30 ), sheep primer S (50 -CTATG- meat cooked at 120 C. The cattle fragment was ampli-
AATGCTGTGGCTATTGTCGCA-30 ), pig primer P ®ed from 120 C sample, but the signal was weaker than
(50 -GCTGATAGTAGATTTGTGATGACCGTA-30 ), the other species.
and horse primer H (50 -CTCAGATTCACTCGACG-
AGGGTAGTA-30 )] (Fig. 1) that were designed from
published sequences of cattle, pig, chicken, sheep, goat
and horse mitochondial cytochrome b genes (Anderson
et al., 1982; Irwin et al., 1991; Desjardins and Morais,
1990). These primers were mixed in the ratio of
1:0.2:3:0.6:3:0.6:2 for SIM:G:C:B:S:P:H, and used toge-
ther for the multiplex PCRs of this study (the ratio 1
means 20 pmol primer/50 ml PCR solution). Thirty-®ve
cycles of ampli®cation were run using a GeneAmp PCR
System 2400 (Perkin±Elmer) as follows: denaturation at
94 C for 0.5 min, annealing at 60 C for 0.5 min, and
extension at 72 C for 0.5 min. Following ampli®cation,
8 ml PCR solution was electrophoresed on 4% NuSieve
GTG agarose gel (FMC) for 30 min at 100 V in TAE
bu€er (40 mM Tris±acetate, 1 mM EDTA, pH 8.0) and
then stained with ethidium bromide (0.5 mg mlÿ1) for
1 h.

3. Results

3.1. Meat identi®cation by species-speci®c primers

Fig. 1 shows primer regions on cytochrome b Fig. 2. Agarose gel electrophoresis of PCR product ampli®ed from the
sequences of the six species. The common forward pri- six meats. G, goat; C, chicken; B, cattle; S, sheep; P, pig; H, horse. M
mer SIM mismatches with the six species at 3±5 bases is a molecular marker, f174/Hincdigest.
146 T. Matsunaga et al./Meat Science 51 (1999) 143±148

Fig. 3. Agarose gel electrophoresis of PCR products ampli®ed from cooked meat. G, goat; C, chicken; B, cattle; S, sheep; P, pig; H, horse. Lane (±)
is a PCR product ampli®ed from a reaction solution without a template DNA. M is a molecular marker, f174/Hincdigest.

3.3. Semi-quanti®cation of mixed DNA ducts. When pig DNA increased, from lanes 2±6, the
band of 398 bp fragment became intense and that of 274
DNAs extracted from cattle and pig meats were bp fragment faint.
mixed for use as templates in the ratios of 88:12, 75:25,
50:50, 25:75 and 12:88. Fig. 4 shows the bands of cattle 3.4. Detection limits of DNA samples
and pig-speci®c fragments of 274 and 398 bp, along with
the relationships between template DNA amounts and Fig. 5 shows the results of PCR ampli®cation from
band intensity. In lane 4 (50:50), two bands from cattle mixed DNA templates of 25, 2.5, 0.25, 0.025 and 0.0025
and pig DNAs indicated similar amounts of PCR pro- ng each. Low molecular bands in all lanes in many cases
were probably primer±dimers produced from seven pri-
mers during PCR. Lanes 1±3 show six bands corre-
sponding to the six species, thus the detection limits was
0.25 ng for all meat species.

4. Discussion

The aim of this study was to develop a simple method

for simultaneous identi®cation of multiple meat species.
Multiplex PCR, in which many primers are used toge-
ther for ampli®cation of multiple target regions, is a
hopeful technique for this purpose. The design of pri-
mers is very important on multiplex PCR techniques,
because primer speci®city and Tm are more critical than
conventional PCRs. Ratios of mismatching in this study
were more than 15% between a species-speci®c primer
and the other species sequences except a pair of sheep
primer and goat template. The mismatches more than
15% decrease Tm more than 15 C, and that make
reverse primers anneal only to the species-speci®c
sequence in multiplex PCR. The sheep primer S mis-
Fig. 4. Agarose gel electrophoresis of PCR products ampli®ed from
cattle and pig DNA mixture. Lanes 1,(100:0); 2,(88:12); 3,(75:25); matched with goat DNA at only two nucleotides, how-
4,(50:50); 5,(25:75); 6,(12:88); 7,(0:100). M is molecular marker, ever, 30 end mismatching was fatal for PCR
f174/Hincdigest. ampli®cation and resulted in no sheep band from a goat
 T. Matsunaga et al./Meat Science 51 (1999) 143±148 147

Fig. 6. Agarose gel electrophoresis of PCR products ampli®ed from

Fig. 5. Agarose gel electrophoresis of PCR products ampli®ed from DNA of the six meats. Lanes 1, 12.5 ng; 2, 25 ng; 3, 12.5 ng; 4, 25 ng.
0.0025, 0.025, 0.25, 2.5 and 25 ng from DNA of the six meats. Lanes 1, Lanes 1 and 2 are the products by mixed primers of the equal con-
2, 3, 4 and 5 are 25, 2.5, 0.25, 0.025 and 0.0025 ng of six meats DNA, centration. Lanes 3 and 4 are of the changed concentration. M is
respectively. Lane 6 is a PCR product ampli®ed from the reaction molecular marker, f174/Hincdigest.
solution without template DNA. M is molecular marker, f174/
Hincdigest.
reverse primers. A little di€erence of Tm among the
reverse primers would a€ect the eciency. In order to
template (Fig. 2). The other primers were also designed control the eciency, the ratio of the primers was
to mismatch with di€erent species at 30 end or next changed according to the results of preliminary experi-
nucleotides (Fig. 1). ments. Lanes 3 and 4 in Fig. 6 showed that all bands
Primer speci®city in the entire DNA of a target spe- from the six species were detectable by using an appro-
cies was examined by conventional PCR using a pair of priate ratio of primer mixture and Fig. 4 showed a pos-
SIM and an each species-speci®c primer. The size of sibility of semi-quantitative analysis for beef and pork
PCR products was as expected with no additional frag- mixture. However, quantitative ampli®cation was found
ment from a target species (data not shown). This result to be unsuccessful as for the other species. Fig. 5
showed that the species-speci®c primers ampli®ed only showed di€erent signal intensity from the same con-
one size fragment from a target species. Primer speci®- centration of DNA of the six species.
city to the other species was examined by multiplex Multiplex PCRs detected meat species prepared at
PCR using the same primer mixture in the method. Fig. high temperature (Fig. 3). Cooked meat DNAs (except
2 showed that multiplex PCR resulted in a single band horse) were ampli®ed from all samples heated at all
of target size from one meat species and no fragment temperatures, but no horse DNA fragment could be
produced by non-speci®c ampli®cation. detected at 120 C. Because 439 bp horse DNA was the
The primers were designed to amplify target sequen- longest fragment, targeted horse DNA fragment would
ces of the six species at similar eciency. Fig. 6 shows, be a€ected more by heat than other DNA fragments at
however, multiplex PCRs using equal amount mixture high temperature. The primer for horse should thus be
of the primers did not result in equal signals from the six designed to amplify a shorter fragment.
species (lanes 1,2). In general, quantitative PCR is di- PCR is quite useful for routine analysis of meat spe-
cult because of unequal eciency of ampli®cation. cies identi®cation, being quick and sensitive. By the
Ampli®cation eciency is a€ected by the di€erence of present method, the six meats could all be identi®ed at
primer sequences. The common primer SIM was the same time more easily and sensitive than usual
designed to be shared by the six species, therefore, methods. The analytical conditions must be improved
ampli®cation eciency of PCR was a€ected by only for quantitative di€erentiation.
148 T. Matsunaga et al./Meat Science 51 (1999) 143±148

References Ebbehùj, K.F., Thomsen, P.D., 1991a. Species di€erentiation of heated

meat products by DNA hybridization. Meat Science 30, 221±234.
Anderson, S., de Bruijin, M.H.L., Coulson, A.R., Eperon, I.C., San- Ebbehùj, K.F., Thomsen, P.D., 1991b. Di€erentiation of closely rela-
ger, F., Young, I.G., 1982. Complete sequence of cattle mitochon- ted species by DNA hybridization. Meat Science 30, 359±366.
drial DNA. Conserved features of the mammalian mitochondrial Fei, S., Okayama, T., Yamanoue, M., Nishikawa, I., Mannen, H.,
genome. Journal of Molecular Biology 156, 683±717. Tsuji, S., 1996. Species identi®cation of meats and meat products by
Baur, C., Teifel-Greiding, J., Leibhardt, E., 1987. Spezi®zierung hit- PCR. Animal Science and Technology, (Jpn) 67, 900±905.
zedenaturierter Fleischproben durch DNA-Analyse. Archiv fur Irwin, D.M., Kocher, T.D., Wilson, A.C., 1991. Evolution of the
Lebensmittelhygiene 38, 172±174. cytochrome b gene of mammals. Journal of Molecular Evolution 32,
Chikuni, K., Ozutsumi, K., Koishikawa, T., Kato, S., 1990. Species 128±144.
identi®cation of cooked meats by DNA hybridization assay. Meat Saiki, R.K., Schart, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich,
Science 27, 119±128. H.A., Arnheim, N., 1985. Enzymatic ampli®cation of b-globin
Chikuni, K., Tabata, T., Kosugiyama, M., Monma, M., Saito, M., genomic sequences and restriction site analysis for diagnosis of
1994a. Polymerase chain reaction assay for detection of sheep and sickle cell anemia. Science 230, 1350±1354.
goat meats. Meat Science 37, 337±345. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A
Chikuni, K., Tabata, T., Saito, M., Monma, M., 1994b. Sequencing of Laboratory Manual, 2nd Edition. Cold Spring Harbour Laboratory
mitochondrial cytochrome b genes for the identi®cation of meat Press, New York.
species. Animal Science and Technology (Jpn) 65, 571±579. Winterù, A.K., Thomsen, P.D., Davies, W., 1990. A comparison of
Desjardins, P., Morais, R., 1990. Sequence and gene organization of DNA-hybridization, immunodi€usion, countercurrent immunoelec-
the chicken mitochondrial genome: A novel gene order in higher trophoresis and isoelectric focusing for detecting the admixture of
vertebrates. Journal of Molecular Biology 212, 599±634. pork to beef. Meat Science 27, 75±85.