Arch Virol (2008) 153: 205–209

DOI 10.1007/s00705-007-1082-y
Printed in The Netherlands

Brief Report
A highly sensitive and specific multiplex RT-PCR to detect foot-and-mouth
disease virus in tissue and food samples

H.-F. Bao, D. Li, J.-H. Guo, Z.-J. Lu, Y.-L. Chen, Z.-X. Liu, X.-T. Liu, Q.-G. Xie

Key Laboratory of Animal Virology of Ministry of Agriculture, State Key Laboratory of Veterinary Etiologic
Biology, Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Lanzhou, Gansu, China

Received 28 July 2007; Accepted 18 September 2007; Published online 8 November 2007
# Springer-Verlag 2007

Summary (FMDV), is a positive-sense single-stranded RNA

Three sets of primers to detect foot-and-mouth dis- virus that is a member of the genus Aphthovirus in
ease virus (FMDV) using multiplex RT-PCR were the family Picornaviridae [7, 11] and occurs as
designed based on several reference nucleotide seven distinct serotypes throughout the world: A,
sequences, and their reaction conditions were de- O, C, Asia1 and South African Territories (SAT)
termined. By testing ten-fold serial dilutions of 1–3. Several global outbreaks have occurred in his-
FMDV, the sensitivity of multiplex RT-PCR is 100 tory, resulting in severe loss of stockbreeding and
times higher than conventional RT-PCR. Mean- damage to international trade of animal products.
while, its specificity was confirmed compared with Therefore, it is very important to diagnose FMD
other related vesicular disease viruses. Furthermore, rapidly and accurately for control and eradicate it.
30 field samples from different animals were tested, RT-PCR techniques have provided a simple and
and the results supported the method’s potential ap- rapid way to detect clinically suspicious samples
plications in routine veterinary quarantine and epi- and differentiate serotypes of FMD [1, 2, 5, 9, 10,
demic surveillance of FMDV. 12]. However, these reports did not show how to
detect virus in oesophageal-pharyngeal (OP) fluid

samples and quarantined meat in which the amount
Foot-and-mouth disease (FMD) is an extremely of FMDV were very low. In the present study, three
contagious viral disease of cloven-hoofed domestic primer sets were designed and used in a multiplex
and wild animals. The causal agent, FMD virus RT-PCR for rapid detection of virus in tissue sam-
ples, OP fluids (probang) and carriers of FMD. The
goal of the study was to optimise sensitivity to en-
Correspondence: Zai-Xin Liu, Key Laboratory of Animal able a rapid detection without producing informa-
Virology of Ministry of Agriculture, State Key Laboratory of
tion on the serotype involved.
Veterinary Etiologic Biology, Lanzhou Veterinary Research
Institute, Chinese Academy of Agricultural Sciences, The National Foot-and-Mouth Disease Reference
Lanzhou, Gansu 730046, China Laboratory of China (NFMDRLC) at Lanzhou
e-mail: [email protected] Veterinary Research Institute (LVRI) provided O=
206 H.-F. Bao et al.

CHA=99 a suckling-mouse-adapted strain, Asia 1= Due to the prevalence of FMDV type Asia 1 in
JS=CHA=2005 (1107.2TCID50=50 mL), a cell cul- China from 2005 to 2006, the Asia 1=JS=CHA=
ture-adapted strain, A=GS=LX=62 of FMDVs, swine 2005 BHK21 cell-adapted isolate was chosen to
vesicular disease virus (SVDV), vesicular stoma- determine the sensitivity of the multiplex RT-
titis virus (VSV), bovine viral diarrhea virus PCR. The 109 dilution was detected using three
(BVDV) and vesicular exanthema of swine virus primer sets in one tube (Fig. 1c), but only a 107
(VESV). Eighteen tissue samples obtained from 3 dilution was detected using a single primer set for
artificially FMDV-infected cattle on the second, conventional RT-PCR (data not shown). Mean-
seventh, fourteenth, seventeenth, twenty-first and while, 18 tissue samples of lymph nodes from 3
twenty-seventh day after challenge. One tissue sam- artificially FMDV-infected cattle were detected.
ple was obtained from negative cattle control. Thirty All of the samples were positive on the second,
field samples were obtained from an endemic area seventh, and fourteenth day after challenge when
of FMD type Asia 1 in 2005. using both multiplex RT-PCR and conventional
Three pairs of primers were designed based on RT-PCR. One sample was positive on the seven-
reference nucleotide sequences in VP1(50 -TGCGG teenth and twenty-first days, but the rest were neg-
CGGCCACCACTACTTC-30 (þ), 50 -GACTCGAC ative when using multiplex-RT-PCR. All of the
GTCTCCCGCCAACT-30 ()), 3A (50 -AGCTCCA samples were negative on the seventeenth and twen-
CGAAAATGTCGAG-30 (þ), 50 -ACGACGGGGGC ty-first days when using conventional RT-PCR. Dur-
TTTTGCTTTCAC-30 ()) and 3C (50 -CGTGATGT ing all of the test process, one tissue sample from
GGCGAGAATGAAGAA-30 (þ), 50 -CGGAAACG one healthy cow used as a control was negative
CACGAGCAGTAT C-30 ()) conserved regions. (Table 1).
FMDV RNA was extracted from pre-treated sam- Thirty field samples from different animals in an
ples using trizol reagent(invitrogen) according to epidemic area were detected with two RT-PCRs.
the manufacturer’s instructions. Multiplex RT-PCR Fifteen samples were positive with multiplex RT-
and conventional RT-PCR were compared at the PCR, but only 3 samples were positive with con-
same time for the same samples. The reverse tran- ventional RT-PCR. Some strains were isolated from
scription reaction for cDNA was carried out at 42  C the positive samples, and the complete VP1 nucle-
for 1 h in a 20-mL volume containing 1RT buffer otide sequences were obtained and compared with
(50 mM Tris–HCl, pH 8.3, 75 mM KCl, 8 mM those from vesicular fluids of ill cattle from the
MgCl2, 10 mM DTT), 10 mL of RNA, 10 U AMV same region. It revealed that they were the same
reverse transcriptase (Promega), 200 mM dNTP FMD (data not show), which confirmed our detec-
mix, 10U of RNAsin inhibitor (Promega), single or tion result (Table 2).
tri-mixed primers for RT. The PCR was carried out Multiplex PCR [3] is gaining popularity because
in a 50-mL volume containing 5 mL of cDNA, 4 mL of its experimental simplicity and greater effective-
of 25 mM MgCl2 (2 mM), 4 mL of 2.5 mM dNTPs ness as well as decreased effort and shorter time
(0.2 mM), 0.25 mL of Taq DNA polymerase (1.5 U), required. In general, the development of a multi-
0.5 mL of each primer (12.5 pmol=mL). Amplifica- plex PCR assay is not an easy task. The most com-
tion conditions were 94  C, 5 min; 35 cycles of mon problem is that quite a number of primers have
94  C, 50 sec; 58  C, 50 sec; 72  C, 1 min; and 72  C, to be used in the same reaction tube, and these
8 min. molecules may interact with each other, which
Three different fragments were amplified sepa- may block the reaction [4, 6, 8]. A further problem
rately, which were 634 bp (in the 3A region), 483 bp may be the reliable identification of the various
(in the VP1 region) and 278 bp (in the 3C region), PCR products. To overcome these problems, a very
when three serotypes FMDV samples were used in careful primer selection and sequence analysis
multiplex RT-PCR (Fig. 1a). Besides, four related was performed here, which yielded clear reactions.
viruses (BVDV, SVDV, VSV, VESV) were assayed All of the primer sequences were aligned with
and all of them were negative (Fig. 1b). the nucleotide sequences database at the National
A multiplex RT-PCR to detect FMDV 207

Fig. 1. (a) Electrophoresis map of a 1.5% agarose gel to show three different sizes of fragments (278, 483, 634 bp) for
three types of FMDV strains (O=CHA=99, Asia 1=JS=CHA=05,A=GS=LX=62) after amplification by multiplex-RT-PCR.
1 is O=CHA=99, 2 is Asia 1=JS=CHA=05, 3 is A=GS=LX=62, 4 is an OP fluid sample, 5 is the cell line BHK21, 6 is water
control, M is DNA ladder marker. (b) Shows the result of the specificity test for multiplex RT-PCR to detect FMDV. Four
related viruses (BVDV, SVDV, VSV, VESV) were assayed, and all of them were negative. 1 is BVDV, 2 is SVDV, 3 is VSV,
4 is VESV, 5 is BHK21 cells, 6 is healthy mouse tissue, 7 is Asia 1=JS=CHA=2005, used as a positive control, 8 is a water
control, 9 is DNA marker (250–2000 bp). (c) Shows the result of the sensitivity test for multiplex RT-PCR to detect FMDV.
The Asia1=JS=CHA=2005 cell-culture-adapted strain was used in ten-fold serial dilutions. 1–8 are 103 –1010 dilutions,
respectively, M is DNA molecular marker (from 100 to 2000 bp)

Table 1. The result of two kinds of RT-PCR assay for 18 samples at different times from three artificially FMDV-infected
cattle

Days Multiplex RT-PCR Conventional RT-PCR

nd th th th st th
Samples 2 7 14 17 21 27 2nd 7th 14th 17th 21st 27th
No. 1 þ þ þ þ þ  þ þ þ   
No. 2 þ þ þ    þ þ þ   
No. 3 þ þ þ    þ þ þ   
Control            
208 H.-F. Bao et al.

Table 2. The result of using multiplex RT-PCR or conven- This requirement was confirmed by testing them in
tional RT-PCR to assay 30 samples from animals from the study. Although FMDV has seven serotypes and
different epidemic areas
is highly variable, conserved regions can still be
No. and Samples Assay result found in its genome. For any diagnostic method
Multi RT-PCR Virus
to be effective, sensitivity and specificity are two
RT-PCR isolation very important factors. In order to detect virus in
samples with a much lower quantity of FMDV, such
1#tongue tissue þ   as OP fluid samples, quarantined meat and carriers
2#tongue tissue þ  þ
3#tongue tissue    of FMD to exclude positive ones, three primer sets
4#tongue tissue    from different genome regions were added in one
5#tongue tissue    PCR tube to improve its sensitivity. Three different
6#tongue tissue    serotypes (O, A, Asia 1) of FMDV strains were am-
1#vesicular tissue    plifiable using these primer pairs, which did not
2#vesicular tissue   
3#vesicular tissue   
amplify other vesicular disease viral pathogens
4#vesicular tissue    such as BVDV, SVDV, VSV, VESV, as shown in
1#OP fluid þ þ þ Fig. 1b. The result that the multiplex RT-PCR was
2#OP fluid þ   100 times more sensitive than conventional RT-
3#OP fluid þ   PCR and its specificity confirmed the primer sets’
4#OP fluid þ   rationality and accuracy. Four other serotypes (C,
5#OP fluid þ  
6#OP fluid    SAT 1–3) were not considered when these three
7#OP fluid    primer sets were designed because these types are
8#OP fluid    not found at present in the country and surround-
9#OP fluid    ings and these four types of reference strains are
10#OP fluid    not stored in our laboratory. However, it would be
1#spinal cord þ þ 
2#spinal cord þ  
better to consider them in the future.
3#spinal cord þ   Regarding the amplification conditions, several
4#spinal cord þ   different annealing temperatures and the concen-
5#spinal cord    tration of Mg2þ were tested for suitability. A 58  C
1#lymph node þ þ þ annealing temperature and 2.5 mM Mg2þ concen-
2#lymph node þ   tration were determined (data not shown). Every
3#lymph node þ  
4#lymph node þ   primer set was tested in conventional RT-PCR.
5#lymph node    Although the reaction conditions in the study were
the most suitable ones for all three primer sets, the
primer set in 3C region (278 bp) was the most sen-
Center for Biotechnology Information (NCBI) site sitive one (see Fig. 1c). Therefore, this primer set
using the Basic Local Alignment Search Tool was used to test sensitivity in this study.
(BLAST) programs for identifying possible repeti- We applied the multiplex RT-PCR method to
tive sequences and also to confirm virus specificity. investigate meat products and OP fluids of possi-
Specific primer pairs were selected so that ampli- ble persistently infected carriers that contained
fied DNA fragments could be separated distinctly low amounts of FMDV, which standard serotyping
by size using agarose gel electrophoresis. multiplex cannot detect. Our method is designed
An additional requirement for the FMDV primer specially for routine quarantine examination to
design was that these sequences had to be selected quickly and accurately inspect suspicious meat
from a highly conserved region in the FMDV ge- products and live animals. The results document
nome, in order to provide a general amplification of the sensitivity and applicability of the method. Our
all possible variants of FMDV. At the same time, next step is to develop it into a diagnostic kit for
the amplification of other viruses had to be voided. quarantine.
A multiplex RT-PCR to detect FMDV 209

Acknowledgements tion of a multiplex PCR for differentiation of foot-and-

mouth disease virus strains native to India. J Virol
We are thankful to National Foot-and-Mouth Disease Methods 126: 1–11
Reference Laboratory of China (NFMDRLC) at Lanzhou 6. Henegariu O, Heerema NA, Dlouhy SR, Vance GH,
Veterinary Research Institute (LVRI) for providing the Vogt PH (1997) Multiplex PCR: critical parameters and
necessary facilities to carry out this work. This work has step-by-step protocol. Biotechniques 23: 504–511
been supported by ‘‘Chinese national 973 project, No. 7. Kitching RP (1999) Foot-and-mouth disease: current
2005CB523201.’’ and ‘‘National key technology R&D world situation. Vaccine 26: 1772–1774
program of China, No. 2006BAD06A03.’’ 8. Markoulatos P, Siafakas N, Moncany, M (2002) Multi-
plex polymerase chain reaction: a practical approach. J
References Clin Lab Anal 16: 47–51
9. Nunez JI, Blanco E, Hernandez T (1998) A RT PCR
1. Amaral-Doel CMF, Owen NE, Kitching RP, Doel TR assay for the differential diagnosis of swine. J Virol
(1993) Detection of foot and mouth disease viral Methods 72: 227–235
sequences in clinical specimens and ethyleneimine- 10. Reid SM, Hutching GH, De Clercq K (1997) Diag-
inactivated preparations by the polymerase chain reac- nosis of foot-and-mouth disease by RT-PCR: evalu-
tion. Vaccine 11: 415–421 ation of primers for the serotypic characterization of
2. Callens M, De Clercq K (1997) Differentiation of the viral RNA in clinical samples. J Virol Methods 83:
seven serotypes of foot-and-mouth disease viral reverse 113–123
transcriptase polymerase chain reaction. J Virol Meth- 11. Sobrino F, Sáiz M, Jim
enez-Clavero MA, N
u~
nez JI,
ods 67: 35–44 Rosas MF, Baranowski E, Ley V (2001) Foot-and-
3. Chamblain JS, Gibbs RA, Renier IE, Nguyen PN, mouth disease virus: an old, well know virus, and still
Caskey CT (1988) Deletion screening of the Duchenne a current threat. Vet Res 32: 1–30
muscular dystrophy locus via Multiplex DNA amplifi- 12. Vangrysperre W, De Clercq K (1996) Rapid and sen-
cation. Nucl Acids Res 16: 11141–11156 sitive polymerase chain reaction based detection
4. Elnifro EM, Ashshi AM, Cooper RJ, Klapper PE (2000) and typing of foot-and-mouth disease virus clinical
Multiplex PCR: optimization and application in diag- samples and cell culture isolates, combined with a
nostic virology. Clin Microbiol Rev 13: 559–570 simultaneous differentiation with other gnomically
5. Giridharan P, Hemadri D, Tosh C, Sanyal A, and=or symptomatically related viruses. Arch Virol
Bandyopadhyay SK (2005) Development and evalua- 141: 331–344