Recognition of coxsackievirus A by Enterovirus genus-specific immune and molecular markers in experimentally infected suckling mice

Pathologie Biologie 53 (2005) 318–323 http://france.elsevier.com/direct/PATBIO/

Original article

Recognition of coxsackievirus A by Enterovirus genus-specific immune and molecular markers in experimentally infected suckling mice Infection expérimentale du souriceau nouveau-né par les coxsackievirus A : détection antigénique et moléculaire et étude cinétique Siwar Nsaibia a, Samia Ben Othman a, Abdelhalim Trabelsi a, Thomas Bourlet b, Mahjoub Aouni a, Bruno Pozzetto b,* a

Laboratory of Transmissible Diseases and of Biologically Active Substances, MDT01, Faculty of Pharmacy, University of Center, Monastir, Tunisia b Laboratory of Bacteriology–Virology, GIMAP, Faculty of Medicine Jacques Lisfranc, Hôpital Nord, University of Saint-Etienne, 42055 Saint-Etienne cedex 2, France Received 29 July 2004; accepted 7 December 2004 Available online 20 January 2005

Abstract Objective. – Most of coxsackieviruses A (CV-A) are difficult to isolate in cell culture and are responsible for flask paralysis in suckling mice. The aim of the present work was to analyze the ability of immune and RT-PCR techniques to detect viral components of three different serotypes, CV-A6, CV-A13, and CV-A14, in skeletal muscles of experimentally infected suckling mice. Material and methods. – The antigen detection was done by immunofluorescence technique on trypsinized muscular cells and by immunoperoxidase assay on frozen sections of skeletal muscle, using a monoclonal antibody directed towards a conserved epitope of the VP1 capsid protein among enteroviruses. The nested RT-PCR technique used primers located in the 5′ non coding region of viral RNA. Results. – The group antigen was present in muscle cells of suckling mice infected by the three serotypes of CV-A which were assayed. Similarly, the muscle specimens were positive by nested RT-PCR. A kinetic study performed with CV-A13 and CV-A14 showed that the RT-PCR assay was positive as soon as 24 h after infection whereas the detection of VP1 antigen and symptoms of flask paralysis were observed only 48 and 72 h after infection, respectively. Conclusion. – These results show that the tested serotypes of CV-A can be easily detected in muscle specimens of suckling mice by using antigenic and molecular techniques currently available for the diagnosis of enterovirus infections. © 2005 Elsevier SAS. All rights reserved. Résumé But de l’étude. – La plupart des coxsackievirus A (CV-A) sont difficilement cultivables sur systèmes cellulaires mais provoquent des paralysies flasques chez les souriceaux âgés de 48 heures. Ce travail analyse la capacité de techniques immunologiques et de RT-PCR à détecter les composants viraux dans les cellules du muscle squelettique de souriceaux nouveau-nés infectés expérimentalement par 3 sérotypes différents, CV-A6, CV-A13, et CV-A14. Matériel et méthodes. – La détection d’antigène a été effectuée par technique d’immunoperoxydase sur coupes musculaires congelées, en utilisant un anticorps monoclonal dirigé contre un épitope de groupe de la protéine de capside VP1 des entérovirus. Résultats. – L’antigène de groupe a été détecté dans les cellules musculaires de souriceaux infectées par les 3 sérotypes de CV-A. De même, les prélèvements musculaires étaient positifs par une technique de RT-PCR nichée utilisant des amorces communes au genre Enterovirus situées dans la région 5′ non codante du génome viral. Une étude cinétique réalisée avec CV-A13 et CV-A14 a montré que la détection par RT-PCR est positive dès 24 heures post infection alors que la détection de l’antigène VP1 et l’apparition des symptômes de paralysie flasque surviennent seulement à 48 et 72 heures post infection, respectivement. * Corresponding author. E-mail address: [email protected] (B. Pozzetto). 0369-8114/$ - see front matter © 2005 Elsevier SAS. All rights reserved. doi:10.1016/j.patbio.2004.12.007

S. Nsaibia et al. / Pathologie Biologie 53 (2005) 318–323

319

Conclusion. – Ces résultats montrent que les sérotypes examinés de CV-A peuvent être facilement détectés dans des échantillons musculaires de souriceaux par des techniques antigéniques et moléculaires couramment utilisées pour le diagnostic des infections à entérovirus. © 2005 Elsevier SAS. All rights reserved. Keywords: Coxsackievirus A; Enterovirus; Suckling mice; RT-PCR; Antigen detection; Monoclonal antibody; Experimental infection Mots clés : Coxsackievirus A ; Entérovirus ; Souriceau nouveau-né ; RT-PCR ; Anticorps monoclonal ; Infection expérimentale ; Détection d’antigène

1. Introduction

2. Materials and methods

The coxsackievirus (CV) A subgroup comprises 23 serotypes belonging to the Enterovirus genus of the Picornaviridae family. These viruses are characterized by their ability to cause flask paralysis when inoculated to newborn mice under 24 h of life, a feature which allows their differentiation from CV-B responsible for lesions involving the central nervous system, pancreas, liver and brown fat in suckling mice. Despite this common characteristic, CV-A constitutes a very heterogeneous group of viruses. Most of them are fastidious to cultivate on standard cell lines with the exception of some serotypes (CV-A7, -A9 and -A16) [1]. Different cell receptors have been characterized for entry of CV-A into susceptible cells [2]. The diseases caused by CV-A in humans range from very specific ones (e.g. hand-foot-and-mouth disease due to CV-A16 or epidemic conjunctivitis due to CV-A24) to uncharacteristic symptoms (fever, upper respiratory infections, skin rashes...) [1]. From a taxonomic point of view, CV-A viruses have been classified by molecular analysis into three of the four Human enterovirus species, A (CV-A2-8, 10, 12, 14 and 16), B (CV-A9) and C (CV-A1, 11, 13, 15, 17-22 and 24) [3,4]. Recently, the complete genome of all members of the species Human enterovirus A [5] and Human enterovirus C [6] has been sequenced.

2.1. Virus strains, inoculum preparation and titration on suckling mice

In clinical virology, one of the main difficulties with most CV-A is their fastidious growth in standard cell lines and the need to use suckling mice for their isolation from clinical specimens. Therefore, the prevalence of human infections with these viruses is largely underestimated. The diagnosis of enterovirus infection has been considerably improved by the use of group-specific immunological and molecular tools allowing the rapid identification of these viruses at the genus level. These techniques include RT-PCR using primers designed from 5′ non coding region (NCR) of enterovirus RNA [7,8] and monoclonal antibodies (Mab) directed towards a group-specific epitope of the structural protein VP1 [9]. The aim of this work was to study the ability of the tools mentioned above to recognize CV-A isolated from muscle tissue of suckling mouse, and to further improve the routine diagnosis of this subgroup of viruses. Mab 5D8/1 (Dako) and RT-PCR performed in the 5′ NCR were shown to detect successfully the tested CV-A serotypes. A kinetic study performed in experimentally infected newborn mice demonstrated that the PCR technique was the first assay to give a positive signal over time.

Strains of CV-A used in this study were chosen according to the recent classification of the Enterovirus genus [2]: CV-A14 as a member of Human enterovirus A, CV-A13 as a member of Human enterovirus C, and CV-A6 as unclassified serotype. The latter serotype was the only one from the three tested to be able to grow currently in cell culture. The three strains of CV-A consisted in clinical isolates recovered from stool samples at least 10 years ago and kept frozen at –30 °C. The identity of the serotypes was controlled by neutralization assay with monovalent polyclonal antisera in suckling mice. Experiments in mice were performed according to the International rules of Bioethics related to animals used for biological research [10]. The viral inoculum was made of eviscerated carcasses of infected newborn mice that were resuspended in saline buffer at the dilution of 1:10 (w/v) and homogenized by crushing in a frozen mortar. After gentle centrifugation, antibiotics (106 IU/ml of penicillin and 10 mg/ml of streptomycin) were added to the supernatant. The viral inoculum was kept frozen at –30 °C. Twenty-four hours old Swiss Albino newborn mice were infected subcutaneously with 50 µl of virus inoculum or control preparation (saline buffer). In titration experiments approximately eight newborn mice were used for each virus dilution. 2.2. Immunofluorescence assay (IFA) on trypsinized cells The carcasses of infected suckling mice were submitted to trypsinization. The muscle specimen was cut into small pieces that were washed in 10 ml of phosphate buffered saline (PBS) and resuspended in 10 ml of 2.5% trypsin (Gibco BRL) diluted 1:5 in PBS and maintained at 37 °C. After 10 min at 37 °C under gentle shaking, the mixture was centrifuged and the first supernatant was discarded. The pellet was mixed again with diluted trypsin and incubated for 10 min as described above. After centrifugation, the previous step was repeated four times. The five supernatants were mixed together and submitted to a last washing step in PBS. The cells were deposited onto slides for IFA, air-dried and fixed in cold acetone. The IFA was done as previously described [11]; Mab 5-D8/1 diluted 1:40 in PBS was used as first antibody and a

320

S. Nsaibia et al. / Pathologie Biologie 53 (2005) 318–323

Table 1 Primers used in this study Oligonucleotide First procedure (RT-PCR) –Primer 006 (sense) –Primer 007 (antisense) Second procedure (nested RT-PCR) Outer primers –Primer K1 (sense) –Primer K2 (antisense) Inter primers –Primer K3 (sense) –Primer K4 (antisense) a

Positiona

Sequence (5′–3′)

445–464 577–596

TCC TCC GGC CCC TGA ATG CG ACC GAC GAA TAC CAC TGT TA

References [6]

[11] 577-596 67–83

ATT GTC ACC ATA AGC AGC CA ACC TTT GTA CGC CTG TT

166–182 447–463

AAG CAC TTC TGT TTC CC ATT CAG GGG CCG GAG GA

With reference to the sequence of CV-B5.

fluorescein-conjugated anti-mouse globulin (Argene Biosoft) diluted 1:100 in PBS containing 0.01% Evans blue as second antibody. Non-inoculated mouse muscular cells were tested in parallel as negative controls. 2.3. Immunoperoxidase assay (IPA) on frozen sections Frozen sections (3 µm-thin) of muscle specimens taken from the hind limbs of suckling mice were fixed in cold acetone, incubated in methanol containing 0.6% H2O2 (v/v) for 20 min, rinsed in distilled water and in PBS and pretreated for 20 min with porcine serum (Dako) diluted 1:5 in PBS at 37 °C. The slides were then submitted to the following treatments, separated by washing steps using PBS: Mab 5-D8/1 diluted 1:40 in PBS for 30 min at 37 °C, peroxidaseconjugated rabbit anti-mouse globulin (Dako) diluted 1:100 in PBS, peroxidase-conjugated porcine anti-rabbit globulin (Dako) diluted 1:100 in PBS and peroxidase substrate [PBS containing 0.06% diaminobenzidine (w/v) and 0.3% H2O2 (v/v)]. 2.4. RNA extraction from muscle specimens and RT-PCR assay The RNA extraction from muscle specimens from the hind limbs of suckling mice was performed according to the onestep procedure of Chomcynski and Sacchi [12] using TRIreagent® (Sigma) as described previously [13]. The reverse transcription step used the reverse transcriptase of murine moloney leukemia virus (MMLV; Q-Biogen) as enzyme and oligonucleotides 007 or K1 as primers (Table 1) in a 20 µl reaction mixture containing 75 mM KCl, 50 mM Tris–HCl, pH 8.3, 3 mM MgCl2, 10 mM dithiotreitol, 0.2 mM (each) deoxynucleoside triphosphate, 50 pM of primers and 200 IU of enzyme. Two PCR assays were performed on muscle specimens. Both of them used primers located in the 5′ NCR of enterovirus RNA (Table 1). The first procedure was as reported by Trabelsi et al. [11], using primers 006 and 007 described by Zoll et al. [8]. The second procedure was a nested RT-PCR performed according to the technique described by Kämmerer et al. [14]. The latter technique, followed by digestion

of PCR products by restriction enzymes BglI, XmnI and StyI, was previously described for partial identification of enteroviruses at the serotype level [14,15]. In both assays, the PCR mixture contained 50 mM KCl, 10 mM Tris–HCl, pH 8.9, 3.6 mM MgCl2, 0.2 mM (each) deoxynucleoside triphosphate, 80 pM of primers described in Table 1 and 1.25 IU of Taq DNA polymerase (Q-Biogen). In both assays, the amplification step consisted of 35 PCR cycles (denaturation for 1 min at 94 °C, primer annealing for 1 min at 42 °C for the one-step PCR and 56 °C for the nested PCR, and elongation for 2 min at 72 °C) in a Hybaid SPR T001 thermocycler. In each experiment, a water sample and muscle tissue from mock-infected mice was used as negative control for extraction and RT-PCR steps.

3. Results 3.1. CV-A titration The three strains of CV-A, titrated on newborn mice by using 10-fold dilutions, gave the following results expressed as lethal doses 50% (LD50): 104.2, 104.8 and 104.3 for CV-A6, -A13, and -A14, respectively. In further experiments, newborn mice were inoculated with approximately 100 LD50 of stock suspensions of each serotype kept at –30 °C. 3.2. Immune detection By IFA, Mab 5-D8/1 was found able to recognize successfully the three serotypes of CV-A in trypsinized carcasses of experimentally infected suckling mice, demonstrating cytoplasmic inclusions typical of enterovirus cytopathic effect; similar observations were performed by IPA on frozen sections of muscle specimens taken from the hind limbs of suckling mice infected by the same viruses (data not shown). 3.3. Detection by RT-PCR assays As shown in Fig. 1, the two PCR assays using primers located in the 5′ NCR of enteroviral RNA gave positive results with muscle specimens from the hind limbs of newborn mice

S. Nsaibia et al. / Pathologie Biologie 53 (2005) 318–323

321

tive as soon as 24 h after infection, versus 48 h for immune techniques (Fig. 2) and 72 h for the appearance of flask paralysis.

4. Discussion

Fig. 1. RT-PCR results of muscle specimens taken from hind limbs of suckling mice infected subcutaneously by 100 LD50 of three CV-A serotypes. In panel A, the procedure described by Zoll et al. [8] was used; lane 1:100 bpladder DNA marker (BioLabs); lane 2: mock-infected mouse; lanes 3–5: suckling mice infected by CV-A6, -A13 and -A14, respectively; lane 6: water control. In panel B, RT-PCR was followed by an enzymatic digestion using BglI (B), XmnI (X) and StyI (S), according to Kämmerer et al. [14]; lanes 1 and 11: 100 bp-ladder DNA marker; lanes 2–4: mouse infected by CV-A6; lanes 5–7: mouse infected by CV-A13; lanes 8–10: mouse infected by CV-A14.

infected by the three serotypes of CV-A. Nested PCR products obtained after RNA amplification of the 3 CV-A strains were digested by three restriction enzymes: BglI, XmnI, and StyI. BglI was not found able to cut the PCR products of any of the three CV-A. CV-A6 and CV-A14 were cut by XmnI, and CV-A6 by StyI only (Fig. 1B). 3.4. Kinetic analysis of CV-A infection in suckling mice The kinetic studies were performed using CV-A13 and -A14 serotypes. After inoculation as described above, a couple of suckling mice was killed 6, 12, 24, 48 and 72 h after infection. For each time, muscle samples from the hind limbs were tested either by IPA (one animal) or by nested RT-PCR (other animal) assay according to the technique of Kämmerer [14]. Fig. 2 illustrates the kinetic of detection of CV-A14 in muscle specimens by IPA. As shown in Table 2, RT-PCR was posi-

Despite frequent involvement in pathologic process, CV-A stains are difficult to isolate in clinical virology because most serotypes are fastidious to cultivate in cell culture. In this work, we studied the ability of immunological and molecular tools currently used for the diagnosis of enterovirus infection in cell culture or directly on clinical samples, to identify CV-A from suckling mouse muscle tissue. Mab 5-D8/1 has been used successfully as a groupspecific reagent for the detection of enterovirus in cell culture [11,16]. When used in suckling mice, preliminary experiments showed a high level of non-specific staining. To reduce this background, the mouse tissues were digested with trypsin before staining; actually, no cross-reactivity was observed on detached cells from mock-infected mice. For immunoenzymatic staining of tissue sections, the successive use on a same slide of two different peroxidase-labeled globulins was shown to reduce significantly the level of non-specific staining, as recommended by the conjugate manufacturer. Using these procedures, we demonstrated that the group-specific Mab 5-D8/1 was able to recognize three serotypes of CV-A belonging to different species of Enterovirus (Human enterovirus A and C) or remaining still unclassified (CV-A6) [2]. These results were predictable because of the conservation of the PALTAVETG motif in the capsid protein VP1 recognized by Mab 5-D8/1 [17,18], among the Enterovirus genus and especially in CV-A serotypes [19–21]. The muscle specimens were then studied by RT-PCR using primers located in the 5′ NCR that exhibits a high level of conservation in the Enterovirus genus. With different sets of primers [8,14,15], it was possible to amplify the three serotypes of CV-A (Fig. 1). Previous studies using similar primers showed that different serotypes of CV-A could be amplified by RT-PCR [7,8,14,15]. Using the restriction enzymes proposed by Kämmerer et al. [14] and Kuan [15] to characterize other serotypes of enterovirus, we showed that the three serotypes of CV-A could be distinguished: CV-6 was cut by XmnI and StyI, CV-A14 by XmnI only, and CV-A13 was not cut by any of these enzymes. These results have potential implications in the clinical diagnosis of infections by CV-A. Although PCR tests using primers located in the 5′ NCR are

Table 2 Kinetic study of the markers of CV-A infection in suckling mice infected with 100 LD50 of CV-A13 and -A14 by subcutaneous route Time post infection (h) 6 12 24 48 72

Presence of flaccid paralysis CV-13 CV-14 – – – – – – – – + +

Detection by IPA CV-13 – – – + +

CV-14 – – – + +

CV-13 – – + + +

Detection by RT-PCR CV-14 – – + + +

322

S. Nsaibia et al. / Pathologie Biologie 53 (2005) 318–323

Fig. 2. Kinetic study of the cytopathic effect of CV-A14 on frozen sections of muscle specimens taken from hind limbs of suckling mice infected with 100 LD50 by subcutaneous route, by IPA using Mab 5-D8/1. Panels A to D correspond to staining observed 12, 24, 48 and 72 h after infection, respectively, at 400× magnification. Note the presence of viral cytoplasmic inclusions at 48 and 72 h after infection.

able to detect CV-A in different clinical samples such as cerebrospinal fluids, stools or nasopharyngeal specimens, it is not possible to know which serotype is involved [22]. To circumvent this difficulty, Gjoen and Brunn [23] proposed a PCR assay using primers in the VP1/VP2 junction sequence of enterovirus, that was specific for most of CV-A serotypes. Alternatively, the technique used in this study combine a nested PCR of the 5′ NCR and an enzymatic restriction analysis. Recently, approaches using the direct sequencing of the VP1 [24,25] or the VP1/VP2 junction [26] were also shown to allow the identification of CV-A at the serotype level. Another useful application of these findings for routine diagnosis of enterovirus infection is the ability of groupspecific techniques to confirm rapidly that a suckling mouse is infected by an enterovirus following the inoculation of a clinical specimen. Classical testing needs to neutralize the clinical specimen with different polyclonal antisera in order to confirm that the disease observed in suckling mouse is thus due to an enterovirus. By using Mab 5-D8/1 on trypsinized muscle cells and/or by doing a standard RT-PCR test on a muscle sample, it is possible to confirm rapidly the enterovirus infection. PCR was positive 1 day before the detection of antigens by IPA and 2 days before the appearance of flask paralysis. The early detection of the virus by PCR cannot be

due to a contamination of the muscle sample by the inoculum since the viral extract was injected in the dorsal region of the suckling mice and the detection was performed in the hind limbs of the animals. This early detection of virus in muscle specimens may be useful to shorten the recognition of EV infection in presumably infected suckling mice but also to avoid that the mouse mother kills and eats the ill pups (personal observation). In conclusion, our results confirm that fastidious serotypes of CV-A can be detected by using group-specific techniques commonly applied to the diagnosis of enterovirus infection in cell culture or directly on clinical specimens. In laboratories still using suckling mice for isolation of CV-A, these techniques can be useful to confirm the enteroviral origin of typical or less typical symptoms observed in animals possibly infected by such viruses.

Acknowledgements This work was supported in part by grants from the RhôneAlpes Region and the French Agency of Technical Cooperation via the CMCU (Comité mixte de coopération universitaire) franco-tunisien.

S. Nsaibia et al. / Pathologie Biologie 53 (2005) 318–323

References [1]

Hyypiä T, Stanway G. Biology of coxsackie virus A. Adv Virus Res 1993;42:343–73. [2] David E, Almond JW. Cell receptors for picornaviruses as determinants cell tropism and pathogenesis. Trends Microbiol 1998;6:1998– 2002. [3] Pöyry T, Kinnunen L, Hyypiä T, Brown B, Horsnell C, Hovi T, et al. Genetic and phylogenetic clustering of enteroviruses. J Gen Virol 1996;77:1699–717. [4] King AM, Brown QF, Christian P, Hovi T, Hyypiä T, Knowles NJ, et al. Picornaviridae. In: Van Regenmortel MHV, Fauquet CM, Bishop DHL, Carstens EB, Estes MK, Lemon SM, et al., editors. Virus taxonomy. Seventh Report of the International Committee on taxonomy of viruses. New York: Academic Press; 2002. p. 657–78. [5] Oberste MS, Penaranda S, Maher K, Pallansch MA. Complete genome sequences of all members of the species Human enterovirus A. J Gen Virol 2004;85:1597–607. [6] Brown B, Oberste MS, Maher K, Pallansch MA. Complete genomic sequencing shows that polioviruses and members of human enterovirus species C are closely related in the noncapsid coding region. J Virol 2003;77:8973–84. [7] Rotbart HA. Enzymatic RNA amplification of the enteroviruses. J Clin Microbiol 1990;28:438–42. [8] Zoll GJ, Melchers WJG, Kopecka H, Jambroes G, Van Der Poel HJA, Galama JMD. General primer-mediated polymerase chain reaction for detection on enteroviruses: application for diagnostic routine and persistent infections. J Clin Microbiol 1992;30:160–5. [9] Yousef GE, Brown IN, Mowbray JF. Clinical and research application of an enterovirus group-reactive monoclonal antibody. Intervirology 1987;28:199–205. [10] Anonymous. Guide for the care and use of laboratory animals. Institute of Laboratory animal resources. Commission on life sciences. National research council. Washington: National Academy Press; 1996. [11] Trabelsi A, Grattard F, Nejmeddine M, Aouni M, Pozzetto B. Evaluation of an enterovirus group-specific and anti-VP1 monoclonal antibody, 5-D8/1, in comparison with neutralization and PCR for rapid identification of enteroviruses in cell culture. J Clin Microbiol 1995; 33:2454–7. [12] Chomcynski P, Sacchi N. Single step method of isolation of RNA by guanidium thiocyanate–phenol–chlorophorm extraction. Anal Biochem 1987;162:156–9.

323

[13] Bourlet T, Omar S, Grattard F, Pozzetto B. Detection of coxsackievirus B3 in intestinal tissue of orally-infected mice by a standardized RT-PCR assay. Clin Diagn Virol 1997;8:143–50. [14] Kämmerer U, Kunkel B, Kun K. Nested PRC for specific detection and rapid identification of human picornaviruses. J Clin Microbiol 1994;32:285–91. [15] Kuan MM. Detection and rapid differentiation of human enteroviruses following genomic amplification. J Clin Microbiol 1997;35: 2598–601. [16] Bourlet T, Gharbi J, Omar S, Aouni M, Pozzetto B. Comparison of a rapid culture method combining an immunoperoxidase test and a group specific anti-VP1 monoclonal antibody with conventional virus isolation techniques for routine detection of enteroviruses in stools. J Med Virol 1998;54:204–9. [17] Samuelson A, Forsgren M, Johansson BO, Wahren B, Sällberg M. Molecular basis for serological cross-reactivity between enteroviruses. Clin Diagn Lab Immunol 1994;1:336–41. [18] Samuelson A, Forsgren M, Sällberg M. Characterization of the recognition site and diagnostic potential of an enterovirus group-reactive monoclonal antibody. Clin Diagn Lab Immunol 1995;2:385–6. [19] Roivainen M, Nävänen A, Korkolainen M, Huhtala M-L, Hovi T. Antigenic regions of poliovirus type 3/Sabin capsid proteins recognized by human sera in the peptide scanning technique. Virology 1991;180:99–107. [20] Hovi T, Roivainen M. Peptide antisera targeted to a conserved sequence in poliovirus capsid protein VP1 cross-react widely with members of the genus Enterovirus. J Clin Microbiol 1993;31:1083–7. [21] Airaksinen A, Roivainen M, Stanway G, Hovi T. Site-saturation of the PALTAVETG motif in coxsackievirus A9 capsid protein VP1 reveals evidence of conservation of a periodic hydrophobicity profile. J Gen Virol 1999;80:1919–27. [22] Muir P, Kämmerer U, Korn K, Mulders MN, Poyri T, Weissbrich B, et al. Molecular typing of enteroviruses: current status and future requirements. The European Union Concerted Action on Virus Meningitis and Encephalitis. Clin Microbiol Rev 1998;11:202–7. [23] Gjoen KV, Brunn N. Specific detection of coxsackie virus A by the polymerase chain reaction. Clin Diagn Virol 1997;8:183–8. [24] Oberste MS, Maher K, Kilpatrick DR, Flemister MR, Brown BA, Pallansch MA. Typing of human enteroviruses by partial sequencing of VP1. J Clin Microbiol 1999;37:1288–93. [25] Oberste MS, Maher K, Kilpatrick DR, Pallansch MA. Molecular evolution of the human enteroviruses: correlation of serotype with VP1 sequence and application to picornavirus classification. J Virol 1999;37:1941–8. [26] Caro V, Guillot S, Delpeyroux F, Crainic R. Molecular strategy for serotyping of human enteroviruses. J Gen Virol 2001;82:79–91.