ABSTRACT
The objective of the present study was to gain a better understanding of the epidemiology of encephalomyocarditis virus (EMCV) infections in pigs by applying molecular techniques. The diagnostic potential of a reverse transcription-PCR (RT-PCR) targeting 286 nucleotides at the 3′ end of the gene which encodes the viral polymerase was assessed with experimental and field samples. In addition, the use of the amplified sequences for an epidemiological study was evaluated. The heart was clearly shown to be the most suitable organ. The detection limit was determined to be 1 viral particle in 100 mg of heart tissue. The sensitivity and specificity of the assay on the basis of the results obtained in this study were 94 and 100%, respectively. Phylogenetic analysis of the amplified sequences classified EMCVs in two distinct lineages. Group A consists of the reference strain ATCC 129B, all isolates collected between 1991 and 1994 in Belgium in association with reproductive failure, and all Greek isolates. All Belgian isolates collected since the first isolation of EMCV in relation to myocardial failure in fatteners in Belgium group together with the isolates from Cyprus (1996 and 1997), Italy (1986 to 1996), and France (1995) in group B irrespective of their pathogenicity. The analyzed part of the 3D gene differed by 13.0% between Groups A and B. In contrast to the sequence homogeneity of the Belgian isolates collected between 1991 and 1994, molecular diversity, which ranged between 0 and 2%, was observed among the Belgian isolates collected in 1995 and 1996. Among all Greek isolates the diversity ranged between 1 and 8%. However, this diversity does not seem to reflect geographical links between the outbreaks. A RT-PCR for the rapid and specific diagnosis of EMCV in a variety of clinical samples followed by nucleotide sequence analysis proved to be valuable for molecular epidemiological studies.
Encephalomyocarditis virus (EMCV) belongs to the genus Cardiovirus of the familyPicornaviridae (18).
EMCV has been recognized as a pathogen of pigs for many years (19). However, the clinical signs of disease appear to vary. The sudden death of pigs (9, 15) and reproductive failure in sows (6, 11) have both been attributed to EMCV infection. Recently severe outbreaks of disease caused by EMCV have been reported in Belgium (15), Italy (5), and Greece and Cyprus (21a). Rodents have frequently been suggested as the reservoir of EMCV, spreading the disease to a wide variety of animal species (1). On the other hand, pig-to-pig transmission and transplacental transmission of EMCV have also been reported (8,12, 15).
The widespread nature of EMCV, the economic losses caused by an outbreak in pig farms (6), and the marked similarity of the clinical picture caused by EMCV and foot-and-mouth disease virus (FMDV) in very young piglets (1) stress the need for a better understanding of the epidemiology of the disease. Molecular techniques have been used to study the epidemiology of disease caused by other picornaviruses such as FMDV and swine vesicular disease virus (SVDV) (2, 4, 26).
A reverse transcription-PCR (RT-PCR) targeting the 3′ end of the gene coding for the viral polymerase, followed by determination of the sequence of the amplified fragment, is proposed as an alternative to virus isolation (VI) and subsequent virus neutralization (VN). To evaluate this diagnostic assay and to investigate its epidemiological significance, tissue samples from pigs from Belgium, Greece, Cyprus, Italy, and France were analyzed.
MATERIALS AND METHODS
Samples.From 14 animals, which were experimentally infected with a Greek EMCV isolate (13, 14), 53 organ tissue samples (heart, spleen, liver, and lung) were collected. Ninety heart tissue samples were collected from pigs in the field in Belgium, Greece, Cyprus, Italy, and France (Table 1). Furthermore, heart tissue was sampled from 43 EMCV-negative pigs.
As the “gold standard,” all samples were examined for EMCV by VI and subsequent VN as described previously (14).
RNA extraction procedure.Approximately 100 mg of tissue was homogenized in 1 ml of TRIzol reagent with an Ultra-Turrax homogenizer (24,000 rpm). After 5 min, 0.2 ml of chloroform was added and the solution was vortexed and incubated at room temperature for 2 to 3 min. Following centrifugation (12,000 × g for 20 min at 4°C) the upper aqueous phase was transferred to a fresh tube and 0.5 ml of ice-cold isopropanol was added. This mixture was stored at −20°C for 1 h. Finally, the RNA was completely precipitated by centrifugation at 12,000 × g for 30 min at 4°C. The resulting pellet was washed with 1 ml of 75% ethanol and air dried after centrifugation (12,000 × g for 5 min at 4°C). Finally, RNA was dissolved in 50 μl of RNase-free water by incubation for 10 min at 56°C.
RT-PCR.The general guidelines for PCR manipulations described by Kwok and Higuchi (16) were observed.
For each RT reaction, 3.8 μl of RNA extraction product was added to 6.2 μl of a RT premixture to obtain a final concentration of 1× first-strand buffer (Gibco BRL), a 0.5 mM each deoxynucleoside triphosphate (Eurogentec), 10 mM dithiothreitol (Gibco BRL), 100 U of Moloney murine leukemia virus reverse transcriptase (Gibco BRL), and 3.5 pmol of random hexamers (Pharmacia Biotech) per μl. Subsequently, RT was carried out at 37°C for 15 min, followed by heating for 5 min at 95°C.
A downstream primer, primer P1 (5′-CCCTACCTCACGGAATGGGGCAAAG-3′; nucleotides [nt] 7655 to 7631), and an upstream primer, primer P2 (5′-GGTGAGAGCAAGCCTCGCAAAGACAG-3′; nt 7370 to 7395) were designed as described previously (25).
The RT product was diluted with 90 μl of diethyl pyrocarbonate-treated H2O. Each PCR mixture consisted of the following reagents in a total volume of 10 μl: 2 μl of the diluted RT product, 1 μl of 10× PCR buffer (Eurogentec), 1 mM MgCl2 (Eurogentec), 0.2 mM each deoxynucleoside triphosphate (Eurogentec), 0.15 U of GoldStar Polymerase (Eurogentec), diethyl pyrocarbonate-treated H2O, and 15 pmol of primers P1 and P2. After an initial denaturation at 95°C for 10 s, 25 cycles of ramping up and down from 94 to 60°C followed by a final extension at 60°C for 10 s were performed. A second amplification with the same primers and same amplification protocol was performed with 0.1 μl of the first amplification product.
PCR products were visualized on agarose gels by UV fluorescence following ethidium bromide staining.
Detection limit.An EMCV-positive heart tissue sample with a titer of 108 median tissue culture infective doses/100 mg was processed as described above for RNA extraction. The resulting RNA extraction product contains 2 × 106 viral particles/μl. Subsequently, this extraction product was titrated, and each dilution was treated as described above to perform the double RT-PCR.
Sequencing and phylogenetic analysis.For sequencing the described PCR was scaled up to a volume of 50 μl. Ten microliters of the respective reaction mixtures was loaded onto an agarose gel to control the amplification, and the amplicons were purified from the resulting 40 μl with Microcon-100 microconcentrators (Amicon). Sequencing was performed with the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit with AmpliTaq DNA polymerase (FS) according to the manufacturer’s protocol (Perkin-Elmer). Primer P2 was used as the sequencing primer. An ABI PRISM 310 Genetic Analyzer was used to analyze the sequences.
The determined sequences were aligned by using the clustal alignment algorithm of the software package Sequence Navigator (Perkin-Elmer). A phylogenetic tree was constructed according to the level of sequence relatedness across the interval from nt 7401 to 7631 of the polymerase gene of the EMCV genome by the NEIGHBOR-JOINING method as implemented in the computer program NEIGHBOR, and a dendrogram was plotted by using the program DRAWGRAM. All programs are from the PHYLIP 3.5c phylogeny package (7).
RESULTS
The sensitivities of the proposed diagnostic procedure, in reference to the results of VI, applied to heart, spleen, lung, and liver tissue samples collected from 14 experimentally infected pigs were determined to be 100, 90, 69, and 55%, respectively.
No EMCV-specific nucleic acids were detected in the heart tissue samples taken from 43 EMCV-negative pigs. The results of VI and double RT-PCR for heart tissue samples collected from pigs in the field in Belgium, Greece, Cyprus, Italy, and France are presented in Table1. Of the 90 samples which were EMCV positive by VI and subsequent VN, 5 were found to be EMCV negative by the double RT-PCR developed in the present study. The present results indicate that the sensitivity [number of true positives/(number of true positives + number of false negatives)] and specificity [number of true negatives/(number of false positives + number of true negatives)] of the proposed assay are 94 and 100%, respectively, under these experimental conditions.
Designation and origin of the heart tissue samples analyzed by virus isolation and RT-PCR
The expected amplicon was revealed up to a dilution of the extraction product of 2 × 106, indicating that the detection limit of the proposed method is 1 viral particle per 100 mg of heart tissue.
The relationships between the different EMCV isolates revealed after analyzing 230 nt of the amplified region at the 3′ end of the viral polymerase gene are shown in Fig. 1.
Phylogenetic tree derived from sequences at the 3′ end of the 3D gene (230 nt) of the EMCVs listed in Table 1. The tree was constructed as described in Materials and Methods. The same superscript suffix indicates EMCVs from the same farm. Isolates Belgium/91, Belgium/92, Belgium/93, and Belgium/94 are Belgian EMCVs, listed in Table 1, which were collected in 1991, 1992, 1993, and 1994, respectively.
DISCUSSION
The PCR technique (23) has increased the sensitivity of detection of viral nucleic acid sequences in clinical specimens. In the present study, RT-PCR followed by nucleotide sequence analysis resulted in the identification and characterization of EMCVs in clinical samples. In reference to the previously described cell culture-based diagnostic technique (14), this approach offers considerable improvements in terms of both the manipulations involved and the duration of the assay. In addition, due to the primer design (25), a distinction between EMCV and FMDV, which causes a clinical picture similar to that caused by EMCV in young piglets (1), can readily be made. This feature is critical because a false-positive diagnosis of FMDV infection can result in considerable economic losses due to trade losses and the need for increased infection control and eradication measures (24).
The heart was clearly shown to be the most suitable organ for the detection of EMCV. The detection limit of 1 viral particle in 100 mg of heart tissue is equivalent to the levels obtained by similar diagnostic RT-PCRs for the detection of FMDV (10, 17). In particular, the ability to detect EMCV in heart tissues collected in Belgium from 1991 to 1994, which required several passages in vitro to isolate EMCV (11), confirms the sensitivity of the assay.
In addition to the diagnostic potential of the RT-PCR, the use of the sequences of the amplified regions for epidemiological study was evaluated. Villaverde et al. (26) and Rodrigo and Dopazo (22) proposed the use of the 3D gene, in addition to the VP1-coding gene, as a molecular marker for FMDV and other picornaviruses. Phylogenetic analysis based on the sequences of the genomic region amplified in this study classified the pig and rodent EMCVs into two distinct lineages. Group A consists of the reference strain ATCC 129B (21), all isolates collected between 1991 and 1994 in Belgium in association with reproductive failure, and all Greek isolates. Remarkably, all Belgian isolates collected since the first isolation of EMCV in relation to myocardial failure in fatteners in Belgium (15) group together with the isolates from Cyprus (1996 and 1997), Italy (1986 to 1996), and France (1995) in group B, irrespective of their pathogenicity. The analyzed part of the 3D gene differed by 13.0% between groups A and B. Villaverde et al. (26) showed that the sequences of 3D genes of FMDV isolates, which are epidemiologically related to each other and which are derived from the same ancestor, varied by 1.5% over a period of 12 years. These data suggest that the recent EMCV outbreaks in Belgium were due to the introduction of an EMCV isolate which is genetically related to EMCVs isolated in other European regions since 1986.
More detailed analysis of the dendrogram (Fig. 1) reveals that the sequences of isolates collected over 1 month from pigs on affected farms in Belgium are identical, indicating that during an EMCV outbreak the proposed target region of the virus does not mutate. On the other hand, when a farm experiences EMCV-related problems for the second time more than 1 year after the first outbreak, molecular characterization shows that both outbreaks are caused by viruses whose sequences differ by more than 2%.
In contrast to the sequence homogeneity of the Belgian isolates collected between 1991 and 1994, molecular diversity, ranging between 0 and 2%, was observed among the Belgian isolates collected in 1995 and 1996. Among all Greek isolates the diversity ranged between 1 and 8%. However, this diversity does not seem to reflect geographical links between the outbreaks. Although the Belgian isolates obtained from 1991 to 1994 appeared to be a homogeneous group of viruses in this study, antigenic differences between some of these isolates have been observed by Brocchi et al. (3). The monoclonal antibodies applied by Brocchi et al. (3) were directed against neutralizing external epitopes of EMCV, while the amplified region encodes part of an internal nonstructural protein, the viral polymerase.
The observed clustering is not indicative of the pathogenicity of the different isolates. This is not surprising since several reports (20, 27) indicate that the virulence markers of EMCV are situated in the capsid-coding region and are restricted to only a few amino acids. Koenen et al. (15) already demonstrated a correlation between the virulence properties of different EMCV isolates for pigs and the genetic makeup of the gene which encodes capsid protein VP1, while differences in virulence are not reflected by the sequence analyzed in the present study.
ACKNOWLEDGMENTS
This work was supported by grants from the European Union, Brussels, Belgium, in the framework of an AIR project (grant AIR3 CT 93-1465).
We are grateful to N. J. Knowles (IAH, Pirbright, United Kingdom) for critical review of the manuscript and E. Paschaleri (Institute of Infectious and Parasitic Diseases, Thessaloniki, Greece) and E. Brocchi (Instituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia, Brescia, Italy) for providing samples. The technical assistance of R. Debaugnies and E. Vermassen is gratefully acknowledged.
FOOTNOTES
- Received 20 May 1998.
- Returned for modification 30 June 1998.
- Accepted 1 September 1998.
- Copyright © 1998 American Society for Microbiology
