Specific detection of Coxsackie viruses A by the polymerase chain reaction

_~ IClinical and [Diagnostic • ]Virology ELSEVIER

Clinical and Diagnostic Virology 8 (!997) 183-188

Specific detection of Coxsackie viruses A by the polymerase chain reaction K.V. Gjoen *, A.-L. Bruu Department of Virology, National Institute of Public Health, Postboks 4404 Torshov, 0403 Oslo, Norway

Received 20 November 1996; accepted 9 May 1997

Abstract Background: Most of the Coxsackie virus A strains are difficult to identify using traditional diagnostic methods such as virus isolation followed by neutralization with type-specific antisera. For the laboratory diagnoses of infections with Coxsackie viruses A, inoculation into newborn mice has traditionally been the method of choice. However, such investigations are complicated and time-consuming. Objectives: To develop a reverse transcriptase (RT) and polymerase chain reaction (PCR) assay for specific detection of Coxsackie viruses A. Study design: A total of 43 clinical specimens containing Coxsackie viruses A, B or echoviruses were investigated retrospectively. Nineteen samples were Coxsackie virus A positive, whereas 24 samples were positive for Coxsackie viruses B or echoviruses. Thirteen non-typable specimens from eight patients were also included, since they were characterized as enterovirus-like by electron microscopy. Results: All the specimens containing Coxsackie virus A were positive with the Coxsackie virus A PCR assay. In addition, five out of eight samples characterized as enterovirus-like by electron microscopy were PCR positive. The PCR assay did not amplify Coxsackie viruses B or echoviruses identified in our laboratory. Conclusion: The RT-PCR protocol established here should provide a useful alternative to the complicated and time-consuming diagnostic method based on live animals. © 1997 Elsevier Science B.V. Keywords: Enteroviruses; Coxsackie viruses A; RT-PCR

1. Introduction

* Corresponding author. Tel.: +47 22042200; fax: +47 22042447.

T h e C o x s a c k i e viruses A b e l o n g to the Enterovirus genus o f the Picornavirulae f a m i l y a n d are f r e q u e n t l y i s o l a t e d h u m a n p a t h o g e n s , implic a t e d in the e t i o l o g y o f a n u m b e r o f diseases. This

0928-0197/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S0928-0197(97)00017-2

184

K.V. Gjoen, A.-L. Bruu / Clinical and Diagnostic Virology 8 (1997) 183 188

group consists of 23 different serotypes. Coxsackie viruses A may cause sporadic infections or epidemic outbreaks, and are most commonly associated with herpangina, hand-foot-and-mouth disease, aseptic meningitis, febrile illness, and conjunctivitis (Melnick et al., 1996). Unlike the other enteroviruses, most of the Coxsackie virus A strains are difficult to identify using traditional diagnostic methods, such as virus isolation followed by neutralization with type-specific antiserum. The identification of these viruses is mainly based on their pathogenicity in newborn mice (Melnick et al., 1996). Taking advantage of the conserved sequences in the 5'-nontranslated region of the enteroviral genome, polymerase chain reaction tests have been developed for detection of the majority of human enteroviruses in a single PCR assay (Rotbart, 1990; Zoll et al., 1992; Muir et al., 1993; Halonen et al., 1995). However, inoculation of newborn mice is often necessary for specific identification of Coxsackie virus A infections. To avoid the use of live animals for diagnostic purposes, we have developed a reverse transcriptase (RT)-PCR assay, which specifically amplifies Coxsackie viruses A. Our results indicate that RTPCR can provide a rapid diagnostic tool for identifying strains of Coxsackie viruses A responsible for infections.

2. Materials and methods 2. I. Clinical specimens and virus isolation

A total of 43 clinical specimens containing Coxsackie viruses or echoviruses were investigated retrospectively. The specimens had been stored at - 70°C for 1-6 years at the Department of Virology, National Institute of Public Health. Seven samples from six cases were Coxsackie virus A9 culture-positive (human embryonic fibroblasts), and 24 samples were Coxsackie virus B or echovirus culture-positive (monkey kidney cell lines). Twelve samples were identified as Coxsackie virus A based on inoculation into newborn mice. The viruses had originally been isolated from appropriate cell cultures, and the viral

serotypes had been determined using LBM antiserum pools (WHO standard serum pools, Enterovirus Reference Laboratorium, Copenhagen) (Lira and Benyesh-Melnick, 1960) or by inoculation of culture supernatants (human embryonic fibroblasts) with cytopathic effect (CPE) into newborn mice. In addition to the use of WHO standard serum pools, the identification of the enterovirus strains was confirmed by neutralization assay with monovalent antiserum (ATCC, American Type Culture Collection (Rockville, MD)) against the appropriate serotype and some other related enteroviruses. The final Coxsackie virus A diagnosis in infected mice was based on histological examination of striated muscles. Additionally, 13 non-typable culture-positive samples (from eight cases) were included, since enterovirus characteristic CPE (in human fibroblasts) was observed. These samples were characterized as enterovirus-like because of their morphology and size in electron microscopy. The material used in the PCR amplification assays were enterovirus-positive cell cultures identified by WHO serum pools, by inoculation into newborn mice, or by electron microscopy. The 18 (n = 12 + 6) Coxsackie virus A/A9-positive patients, and the eight enterovirus-like-positive patients were divided into three groups according to their clinical symptoms. Group A (n = 15) represents those with symptoms consistent with hand-foot-and-mouth diseases (exanthema of the buccal mucosa accompanied by mild fever and vesicular lesions on the hands and feet). Group B (n = 4) consisted of cases with symptoms of a CNS-affection (meningitis, encephalitis, and neck stiffness). Group C (n = 7) consisted of cases with uncharacteristic symptoms such as fever, diarrhoea, and influenza. Prototype strains of Coxsackie virus A9 and A16 (ATCC), as well as Coxsackie viruses B1 5, and echoviruses 3, 7, 9, 11, 21, 30, and 31 were included in the study for comparison. The viruses were propagated in appropriate cell cultures for viral RNA extraction in the present study. When extensive cytopathic effect was observed, the virus-infected cells were harvested and collected by centrifugation.

185

K.V. Gioen, A.-L. Bruu / Clinical and Diagnostic Virology 8 (1997) 183 188 Table 1 Sequence, melting temperature, genomic location, and nucleotide position in Coxsackie virus A9 of oligonucleotide primers Primer

Sequence 5' 3'

Position

Localization

Tm 1°C) (%GC)

Upper Lower

ATGGCAACAGGAAAA(A/T)T(T/G)(T/C)TA ~ ACAAAAGT(A/T)AACTCTA(A/T/C)ATCAA a

2106 2127 b 2790 2812 b

VP2 VP1

41.4 38.5

~ The two primers contain following 'wobble' bases, giving specificity for Coxsackie viruses A9, 16, and 24. b The corresponding positions in Coxsackie virus A16 are: upper primer 2095 2115, lower primer 2885~ 2906:and in Coxsackie virus A24: upper primer 2140-2160, lower primer 2837 2858.

2.2. R N A extraction

3. Results

RNA was prepared from 0.25 ml of virus infected cell culture using T R I z o F M LS Reagent as described by the manufacturer (Life Technologies TM, Gibco-BRL, USA). The isolated R N A was subsequently reverse transcripted to cDNA.

The Coxsackie virus A RT-PCR assay presently described only detected enteroviruses belonging to Coxsackie viruses A. Results of Coxsackie virus A PCR are summarized in Table 3 together with nested enterovirus PCR. The two PCR assays were used to investigate a panel of 19 samples with Coxsackie viruses A. All saml~les from which Coxsackie viruses A, Coxsackie virus A9, and 16 prototype viral stocks from ATCC were isolated proved positive in both PCR assays. In comparison, all samples containing Coxsackie viruses B or echoviruses were PCR negative with Coxsackie virus A PCR, though positive with enterovirus PCR. These enterovirus serotypes were those most commonly detected in clinical specimens in our laboratory during 1990 1995. These results demonstrate the specificity of the Coxsackie virus A PCR assay. The sensitivity of detection was assessed using serial dilution on isolated R N A from the prototype Coxsackie virus A9. The sensitivity of the Coxsackie virus A PCR was compared to the enterovirus PCR assay (Glim~tker et al., 1992), and proved to be between 1 x 10 4 a n d 1 × 10 5 dilution of Coxsackie virus A9 RNA. Serial dilution experiments demonstrated that the nested enterovirus PCR assay was 2 log more sensitive than the Coxsackie virus A PCR assay (Fig. 1). In 13 enterovirus PCR-positive samples (eight patients) identified as enterovirus-like by electron microscopy, Coxsackie virus A R N A was demonstrated in seven samples from five cases (Table 3). The discrepancy observed between the two PCR assays may be due to the higher s.:nsitivity of the nested enterovirus PCR assay. On the other hand,

2.3. R T and P C R

Antisense primer was used for the c D N A synthesis according to the manufacturer's protocol (Perkin-Elmer Cetus, USA). The PCR is based on the VP2/VPI junction sequence in Coxsackie viruses A9, 16, and 24 (Chang et al., 1989; Supanaranond et al., 1992; P6yry et al., 1994). The two PCR primers (R&D Systems, UK) were designed using O L I G O 4.0 software (National Biosciences, Inc., USA). The primer sequences, positions, localization, and melting temperatures (Tm) are shown in Table 1, whereas comparison of the primer sequences with corresponding region in other enteroviruses are shown in Table 2. The PCR was run in MicroAmp tubes (Perkin-Elmer) in a Perkin-Elmer 9600 thermal cycler. Reagents (90 /tl) for the PCR were added to the 1 0 / t l of cDNA, and the PCR was run for 35 cycles of 94°C 30 s, 50°C 30 s, and 72°C 60 s. The final concentrations during PCR were: 50 m M KC1, 10 mM Tris HCI pH 8.3, 2 m M MgC12, 0.2 m M of each dNTP, 1 /~M of each primer and 1 U Taq D N A polymerase (Perkin-Elmer). The PCR products were visualized by agarose gel electrophoresis and ethidium bromide staining. A amplification product of the expected size was taken as a positive result. All samples were also tested with a RT-PCR detecting the majority of human enteroviruses (Glimfiker et al., 1992).

K.V. Gjoen, A.-L. Bruu / Clinical and Diagnostic Virology 8 (1997) 183-188

186

Table 2 Comparison of the primer sequences with corresponding region in enteroviruses: only mismatches are marked Virus type

Upper primer sequence 5 ' - 3 '

Lower primer sequence 5' 3'

ATGGCAACAGGAAAA(A/T)T(T/G)(T/C)TA

ACAAAAGT(A/T)AACTCTA(A/T/C)ATCAA

CAV9 (D00627) CAV16 (U05876) CAV24 (D90457) CAV21 (D00538) CAV2 (L28146) CBVI (M16560) CBV3 (M16572) CBV4 (X05690) CBV5 (X67706) E C H O 9 (X84981) E C H O l l (X80059) ECHO12 (X77708) poliol (V01148) polio2 (X00595) polio3 (X01076)

G

C G

C C T T C Y T T T T T C T G G T C C T G

G T G

T C

G G C

C T C T CC C

C C

G

C G G T T

Y C G G T

G C C

G

A C

G

G G G

C C G C C G G C

G G

G

G G

T TT T Y G G G CT C CT GT T

C G A AG C C C

A C C C C C

G G G

G

G G

TG G G GC

G

G G

Enterovirus sequences were obtained from the GenBank, and the accession numbers are given in parenthesis.

enterovirus-like positive patients may also be infected with enteroviruses other than Coxsackie viruses A. As shown in Table 4, three out of five Coxsackie virus A PCR-positive patients with an enterovirus-like diagnosis belonged to group A. Only one of the three Coxsackie virus A PCRnegative patients belonged to group A. In patients with uncharacteristic clinical symptoms belonging to group C, five out of seven were Coxsackie virus A PCR positive. In comparison,

1

7

9

18

123 bp-

Fig. 1. Comparison of sensitivity between two RT-PCR assays on serial dilutions of RNA isolated from Coxsackie virus A9 prototype strain. In lanes 1 7 are 10° to 106 ten-fold dilutions of Coxsackie virus A9 R N A amplified by the Coxsackie virus A PCR protocol, and in lanes 9-18 are 10° to 109 ten-fold dilutions of Coxsackie virus A9 R N A amplified by the nested enterovirus PCR protocol. The size marker in lane 8 is a 123-bp D N A ladder.

three out of these five Coxsackie virus A PCRpositive patients were identified as Coxsackie virus A positive by inoculation into newborn mice or cell culture.

4. Discussion We have developed a specific and sensitive PCR assay for detection of infections caused by Coxsackie viruses A. Although infections caused by Coxsackie viruses A are usually mild, they are epidemiologically important to diagnose in order to confirm outbreaks of hand-foot-and-mouth diseases. As shown in Table 4, 11 out of 12 Coxsackie virus A-positive patients (coxA) belonged to group A, whereas only one belonged to group C. In contrast, only one patient in group A was Coxsackie virus A PCR negative, which may be a false-negative result. Coxsackie viruses A have also been associated with more severe diseases like aseptic meningitis and encephalitis. The four patients in group B were Coxsackie virus A PCR positive. Finally, all the 18 patients identified as Coxsackie virus A/A9 positive by inoculation into newborn mice or cell culture, proved positive in the Coxsackie virus A PCR. Our re-

K.V. Gjoen, A.-L. Bruu / Clinical and Diagnostic Virology 8 (1997) 183 188

187

Table 3 Comparison of Coxsackie virus A PCR and enterovirus P C R on cell culture materials grouped by results in neutralization typing and inoculation into newborn mice Type of virus

No. of virus isolates

Coxsackie virus A, P C R positive

Enterovirus, PCR positive

Coxsackie virus type A Coxsackie virus type A9 CA9 (prototype) C A I 6 (prototype) CBV 1 - 5 ~ Echoviruses b Enterovirus-like"

12 7 1 1 10 14 13

12 7 1 1 0 0 7

12 7 1 1 10 14 13

Two of each serotype of Coxsackie virus B1-5 (CBV1-5). b Two of each of following seven serotypes; echoviruses 3, 7, 9, 11, 21, 30, and 31. Enterovirus-like diagnosis was based on electron microscopy.

sults clearly show the ability of this assay to detect infections caused by Coxsackie viruses A. Four Coxsackie virus A serotypes sequenced to date (Coxsackie virus A9, A16, A21, and A24) represent three different genetic clusters, indicating a high degree of diversity within the Coxsackie virus A group (P6yry et al., 1994; Pulli et al., 1995). Coxsackie virus A9 is genetically related to Coxsackie viruses B and echoviruses, and the disease pattern often resembles that of Coxsackie viruses B (meningitis and carditis). Coxsackie virus A24 and some other serotypes are genetically related to polioviruses. Hand-foot-andmouth disease is usually caused by Coxsackie virus A16, which together with related Coxsackie viruses A form a cluster of their own. The disease pattern of Coxsackie viruses A is still not completely understood, and direct clinical association of molecular characteristics has not been found in

spite of certain similarities in genetic grouping and disease patterns of some Coxsackie virus A serotypes (Pulli et al., 1995). In this study, some association between genetic grouping and disease pattern may be suspected. All patients in group B were Coxsackie virus A9 positive in cell culture, and the disease pattern also resembles that of Coxsackie viruses B (Table 4). Coxsackie virus A16 is the probable serotype in group A, since the patients had symptoms consistent with hand-footand-mouth disease. In this group only one out of 15 patients were Coxsackie virus A PCR negative (Table 4). Coxsackie virus A24 belongs to a genetic cluster containing viruses with different clinical characteristics (Pulli et al., 1995), and may therefore be the dominant serotype in group C. However, two patients in group C were Coxsackie virus A9 culture positive (Table 4). The Coxsackie virus A PCR assay does not distinguish between

Table 4 Correlation between Coxsackie virus A PCR results and clinical symptoms observed in the patients with the following diagnoses: Coxsackie virus A positive with inoculation into newborn mice (coxA), Coxsackie virus A9 positive in cell cul ure (coxA9), and enterovirus-like (ENL) positive in electron microscopy Coxsackie virus A PCR PCR positive

PCR negative

Group

coxA (n = 12)

coxA9 (n = 6)

ENL (n = 5)

coxA (n = 0)

coxA9 (n = 0)

t ' N L (n = 3)

A (n = 15) B (n = 4) C (n = 7)

11 0 1

0 4 2

3 0 2

0 0 0

0 0 0

1 C 2

188

K.V. Gjoen, A.-L. Bruu / Clinical and Diagnostic Virology 8 (1997) 183 188

Coxsackie virus A9, A16, and A24 serotypes. Further information between disease pattern and genetic grouping may be obtained with sequencing of the amplification products. The Coxsackie virus A PCR assay developed in our laboratory was not as sensitive as the nested enterovirus PCR assay. However, the increased sensitivity observed with the Coxsackie virus A PCR assay, when compared to inoculation into newborn mice or electron microscopy, should make it possible to assess the presence of Coxsackie virus A infections more accurately. The primer pairs constructed for Coxsackie virus A detection specifically amplified Coxsackie viruses A in this study. Coxsackie viruses A2 and A21 are not expected to be amplified due to the high number of mismatches between the PCR primers and the corresponding regions in these two viruses (Table 2). As shown in Table 2, both PCR primers might hybridize with three mismatches to Coxsackie virus B5. Nevertheless, the samples containing Coxsackie virus B5 were Coxsackie virus A PCR negative. These results indicate that under our PCR conditions, these three mismatches in each of the PCR primers are sufficient for discrimination of Coxsackie virus B5. The established Coxsackie virus A PCR protocol is a fast, sensiti6ve, and specific method for detection of some serotypes of Coxsackie viruses A. The method might therefore be a useful alternative to the complicated, and timeconsuming diagnostic method involving living animals.

References Chang K, Auvinen P, Hyypifi T, Stanway G. The nucleotide sequence of coxsackievirus A9: Implications for receptor binding and enterovirus classification. J Gen Virol 1989;70:3269-80. GlimAker M, Johansson B, Olc6n P, Ehrnst A, Forsgren M. Detection of Enteroviral RNA by polymerase chain reaction in cerebrospinal fluid from patients with aseptic meningitis. Scand J Infect Dis 1992;25:547 57. Halonen P, Rocha E, Hierholzer J, Holloway B, Hyypifi T, Hurskainen P, Pallansch M. Detection of enteroviruses and rhinoviruses in clinical specimens by polymerase chain reaction and liquid-phase hybridization. J Clin Microbiol 1995;33:648-53. Lim KA, Benyesh-Melnick M. Typing of viruses by combination of antiserum pools. Application to typing of enteroviruses (coxsackie and ECHO). J Immunol 1960;84:309 17. Melnick JL. Virology. In: Fields BN et al, editors. New York: Raven Press, 1996:655-705. Muir P, Nicholson F, Jetham M, Neogi S, Banatvala JE. Rapid diagnosis of enteroviral infection by magnetic bead extraction and polymerase chain reaction detection of enteroviral RNA in clinical specimens. J Clin Microbiol 1993;31:31--8. Pulli T, Koskimies P, Hyypi/i T. Molecular comparison of Coxsackie A virus serotypes. Virology 1995;212:30 8. P6yry T, Hyypifi T, Horsnell C, Kinnunen L, Hovi T, Startway G. Molecular analysis of coxsackievirus AI6 reveals a new genetic group of enteroviruses. Virology 1994;202:982-7. Rotbart HA. Enzymatic RNA amplification of the enteroviruses. J Clin Microbiol 1990;28:38 442. Supanaranond K, Takeda N, Yamaki S. The complete nucleotide sequence of a variant of coxsackievirus A24, an agent causing acute hemorrhagic conjunctivitis. Virus Genes 1992;6:149-58. Zoll G J, Melchers WJG, Kopecka H, Jambroes G, van der Poel HJA, Galama JMD. General primer-mediated polymerase chain reaction for detection of enteroviruses: application for diagnostic routine and persistent infections. J Clin Microbiol 1992;30:160 5.