- Email: [email protected]
Identification of rhinoviruses by cDNA probes
Journal of Virological Methods, 27 (1990) 61-68
Elsevier VIRMET 00966
Identification of rhinoviruses by cDNA probes Petri Auvinenl,
Thedi Ziegler’, Tim Skern2, Ernst Kuechleti, Glyn Stanway and Timo Hyypiti’
‘Department of Virology, University of Turku, Turku, Finland, 21nstitutefor Biochemistry, University of Vienna, Vienna, Austria and ‘Department of Biology, University of Essex, Colchester, U.K.
(Accepted 12 September 1989)
Summary We have used nucleic acid hybridization for the detection and grouping of human rhinoviruses (HRV) according to their genetic relationships. Fifteen rhinovirus reference strains, seventy-one clinical isolates and four enteroviruses were propagated in cell cultures, spotted onto membrane filters and hybridized with radioactively labelled cDNA probes covering different parts of the genomes of HRVlB, HRV-2, HRV-14, HRV-85 and HRV-89. When the rhinovirus and enterovirus reference strains were tested, the 5’ probe of HRV-2 hybridized with thirteen of the fifteen HRV reference strains, with poliovirus type 3 and with ECHO virus 11. The HRV-14 5’ probe reacted with eleven HRV reference strains and with all the enteroviruses studied. Sixty-nine of the 71 clinical isolates were recognised by the HRV-2 5’ probe, whereas the HRV-14 probe from the same part of the genome hybridized with 54 field isolates. One of the two isolates that remained negative with the HRV-2 5’ probe was detected with the HRV-2 probe that derived from the P2 region of the genome, and the other isolate was not detected by any of the probes. Probes from other parts than the 5’ end of the genome were generally more specific, and clusters could be formed based on the reactivity of the HRV strains with these probes. Rhinovirus; Nucleic acid hybridization
Correspondence to: P. Auvinen, University of Turku, Department SF-20520 Turku, Finland.
of Virology, Kiinamyllynkatu 13,
0166~0934/90/$03.50@ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
62
Introduction More than one hundred serologically distinct viruses have been classified as human rhinoviruses (HRV), the most frequent agent causing common cold. The viruses have an icosahedral capsid consisting of 60 copies of each of the four structural proteins which enclose a monocistronic, positive strand RNA genome with an approximate length of 7200 nucleotides. With accumulating knowledge about the genomic composition, it has become evident that the different serotypes share much homology including some common sequences, although HRV-14 is less closely related to the others studied (Stanway et al., 1984; Skern et al., 1985; Callahan et al., 1985; Duechler et al., 1987; Hughes et al., 1988). The laboratory diagnosis of HRV infections is currently based on virus isolation; HRVs are identified by their typical cytopathogenic effect and by the lability to acid treatment. Further typing can be done by neutralization tests. These tests, however, are laborious and time-consuming and this has prompted efforts to develop more rapid diagnostic techniques based on nucleic acid hybridization (AlNakib et al., 1986; Bruce et al., 1989). Detection of HRV using enzymatic amplification by the polymerase chain reaction has been reported recently (Gama et al., 1988, 1989). We describe studies on the applicability of cDNA probes representing HRV-lB, HRV-2, HRV-14, HRV-85 and HRV-89 genomes for the detection and identification of HRV reference strains and clinical isolates in cell culture.
Materials and Methods Cells, reference
viruses, clinical samples and antisera
HRV serotypes lB, 2, 3, 9, 11, 12, 13, 14, 22, 29, 36, 38, 39, 48 and 89 were grown in HeLa-Ohio cells (a kind gift of Dr. D.A.J. Tyrrell, Salisbury, U.K.). Coxsackieviruses A9 and B3, ECHO virus 11 and poliovirus type 3 were propagated in LLC-MK2 cells. All reference strains and specific antisera were obtained from the American Type Culture Collection (Rockville, MD). Antisera against HRV 2, 3, 11, 12, 13, 14, 36, 38, 48 and 89 were used for typing of selected isolates by standard neutralization assay. Specimens were collected from adult and paediatric patients with upper respiratory tract infection during a 15 month period between September 1986 and November 1987. HRVs were isolated in HeLa-Ohio cells as described elsewhere (Arola et al., 1988). Preparation of specimens
Cells were inoculated for nucleic acid hybridization with reference viruses and with first passage material of the clinical isolates at a high multiplicity of infection. The cells were scraped into the supernatant when the cytopathic changes were advanced, pelleted by low speed centrifugation and stored at -70°C until used. Un-
63
-
PlYlI
Fig. 1. Localization and the size of the HRV-2 and HRV-14 probes. The map of rhinovirus genome is shown in the upper part of the figure. The codes of the probes are presented on the right.
infected HeLa and LLC cells were used as controls. Prior to hybridization, cells and clinical samples were treated with proteinase K (100 lqjrnl) and 0.5% SDS at 37°C for 1 h. Nucleic acids were extracted with phenol and chloroform, and material corresponding to 2 X 105 cells was spotted, after denaturation by formamide-formaldehyde (WC for 1 h), to a filter membrane (GeneScreen PlusTM, New England Nuclear, Boston, MA) in a manifold apparatus (Schleicher and Schuell, Dassel, F.R.G.). Hybridization procedure
cDNA probes covering different regions of the genomes of HRV-1B (Hughes et al., 1988), HRV-2 (Skem et al., 1985; Fig. l), HRV-14 (Stanway et al., 1984; Fig. l), HRV-85 (Stanway et al., unpublished data), HRV-89 (Duechler et al., 1987) and poliovirus 3 (Cann et al., 1983) were used for the hybridization analysis. The probes contained both the insert and the vector with the exception that the HRV-14 probes 752, 2817 and 3643 were restriction enzyme fragments separated by agarose gel electrophoresis. DNA was eluted from gel slices by isotachophoresis (Gfverstedt et al., 1984). Probes were labelled by nick translation with CX-~~P dCTP (Amersham International plc, Buckinghamshire, U.K.). Hybridization was carried out at 42°C overnight in a solution containing 50% formamide (Auvinen et al., 1989). After hybridization, filters were washed and exposed to X-ray films (Trimax XD, 3M, Milan, Italy).
64
Results Reactivity of the reference
viruses with the cDNA probes
The reactivity of the HRV-2 and HRV-14 probes with the reference viruses is summarized in Table 1. Clone 109 from the 5’ end of HRV-2 gave a positive signal with 13 of the 15 HRV reference strains and also with poliovirus type 3 and ECHO virus 11 (Fig. 2b). After an extended exposure, a weak positive signal was observed also with HRV-29 and HRV-48, and with all the enteroviruses. HRV-11 hybridized only with the probe localized at the 5’ end of HRV-2. Probe 752 from the 5’ non-coding region of HRV-14 reacted with eleven HRV reference strains and with all the enteroviruses. The probes covering the 5’ end of HRV-lB, HRV85 and HRV-89 genomes also gave a positive signal with a majority of the reference strains. The HRV-2 and HRV-14 probes covering the capsid regions distinguished between two groups of reference viruses. The first group contained eight HRV serotypes (2, 9, 12, 13, 22, 36, 38 and 39) while four HRV strains (3, 14, 48 and 89) were in the other group. HRV-2 probe 96, representing the middle part of the TABLE 1 Reactivity of the HRV-2 and HRV-14 probes with the reference virus strains in the spot hybridization test virus strain virus titre
1B 2 3 9 11 12 13 14 22 29 36 38 39 48 89 CA9” CB3b Echo11 Polio3
10-7 lo-’
HRV-14 probes
HRV-2 probes 109
107 132 899
96
24 44
752
2817
3643
PAM2
+ +
+
+
+ +
+ +
_ -
_
+ +
10-7
+
-
-
-
+
+
+
+
lo-’ lo-’ lo-* lo-’ 10-7 lo-’ 10-s 10-s lo-’ lo-’ 10-6 10-7
+ + + + + + + + + +
+ _
-
+ _ _
-
+ + + -
+ + -
+ _ _ + +
+ -
+ -
+
+ + + + + + + +
_ + +
-
_
+ + + +
+ + +
+ + + _
“Coxsackievirus A9; %Zoxsackievirus p3.
+ + -
-
f +
+ + + + + + + +
+
+ + + +
a 18 2 3 9 11 12 13 14
b
c
22 CA9 29cB3 36 El1 38 P3 ,39 LLC 48 89 HeLa
Fig. 2. Examples of hybridization patterns with reference strains using HRV-2 and poliovirus probes. (a) Picornavirus serotypes used in the study. (b) HRV-2 probe covering nucleotides l-685. (c) Poliovirus 3 probe covering nucleotides l-784. lB-89 are rhinovirus serotypes, CA9 = Coxsackievims A9, CB3 = Coxsackievims B3, El1 = ECHO virus 11, P3 = poliovirus 3, LLC and HeLa are control cells.
nonstructural region, reacted more specifically with a restricted number of strains than did the probe deriving from the part coding for the capsid proteins. The HRV14 3’ probe was broadly reactive and recognized 80% of the rhinovirus reference strains and all the enteroviruses studied (Table 1). A probe mapping to position 2.G3.8 kb of the HRV-89 genome reacted with HRV-36, and a probe covering region 6.0-7.2 kb reacted also with HRV-14 (data not shown). Reactivity of the clinical isolates with the probes
Every probe reacted with some of the clinical isolates (Table 2). Probes deriving from the 5’ end showed the widest reactivity. The HRV-14 5’ probe 752 detected 54 of the 71 clinical isolates (76%), and the HRV-2 probe 109 hybridized with all except two (97%). One of these two negative specimens, however, reacted with the HRV-2 probe 96 deriving from the P2 region. The other one remained negative with all the probes. The probes originating from the capsid and the P2 region reacted more selectively with the clinical isolates. The group of clones from the capsid region of HRV2 gave a positive signal with 21 specimens, and the HRV-14 fragment detected 18 isolates. The highest specificity was observed with the clone 96 from the P2 region which detected only four of the clinical isolates. The broadly reacting HRV-14 3’ probe PAM2 recognized 40 isolates (56%). Twenty of the clinical samples were also tested by direct hybridization. Only one of them was positive when the combination of both HRV-2 and HRV-14 5’ probes were used.
66 TABLE 2 Reactivity patterns of the HRV-2 and HRV-14 probes with clinical isolates Reactivity pattern
HRV-14 probes
HRV-2 probes 109
107 132
96
24 44
752
2817
+
+ +
+ + + -
-
+ + -
+
+ +
-
+
+ + _ +
-
+
+ + -
7
54
3643
PAM2
No. of reactive isolates
899 1
+
2 3 4 5 6 7
+ + + + + _
8 9 10 11 12 13 14 15 16 17 18 19 20
+ + + + + + + + + + + + _
Number 69 of reactive isolates
+ + + + + + + + + -
21
4
-
+ _
-
+ + + + + + + + + -
9 3 2 2 2 2 1 1 1 2 11 6 1 1 1 9 8 6 1 1
20
40
71
-
-
18
Neutralization of selected strains
Twenty-seven isolates, selected from different reactivity groups, were subjected to neutralization test against one or more HRV sera to evaluate the correlation between hybridization and serotyping. Antisera against serotypes 13 (anti-HRV 13) and 36 were used to neutralize 5 isolates of pattern 1, anti-HRV 12 for 2 isolates of pattern 6, anti-HRV 2 and 38 for the isolate of pattern 8, anti-HRV 3, 14, 48 and 89 for the 18 isolates of patterns 11,12,13, 14 and 15 and anti-HRV 11 for the isolate of pattern 17. One isolate of reaction pattern 12 was neutralized by antiHRV 14. No neutralization was observed with the other combinations.
Discussion
We used nucleic acid hybridization for grouping rhinoviruses on the basis of their genetic similarity. HRV-2 and HRV-14 5’ probes showed high conservation of this
part of the genome among rhinoviruses. Other parts were more specific in their reactivity forming clusters of closely and more distantly related viruses. Based on the reaction patterns obtained with probes covering the capsid region of HRV-2 and HRV-14, the reference viruses included in this study could roughly be classified into three groups (Table 1). The Iirst group consisted of eight serotypes closely related to HRV-2, four rhinoviruses were detected by HRV-14 capsid fragment, and rhinoviruses lB, 11 and 29 were not reactive. In general, the HRV-14 probes were more likely to hybridize with the enteroviruses than were the HRV-2 clones. This finding is in agreement with data obtained from nucleotide sequencing, which indicate that HRV-14 is more closely related to polioviruses than any of the other rhinoviruses sequenced so far (Stanway et al., 1984). However, the 3’ end of HRV14 was reactive with many rhinoviruses. Our findings agree well with the results of Al-Nakib et al. (1986), with the exception that we observed a closer relationship between HRV-14 and HRV-48. Twenty different reaction patterns were found with the clinical isolates (Table 2). They could be divided into ‘HRV-2 like’ and ‘HRV-14 like’ patterns mainly on the basis of their reactivity with probes covering the Pl and P2 regions. Reaction patterns l-8 belong to the former group, and 11-15 to the latter. Reaction patterns 9 and 10 form intermediates between these two groups while 16-20 do not fall into any of these categories and they apparently represent genetically more distinct viruses. No clustering of isolates of either group was observed during the time the specimens were collected, indicating that viruses of all groups can circulate simultaneously in the population. Fourteen of the clinical isolates were also tested after a polymerase chain reaction (Hyypl et al., in press) using synthetic oligonucleotides as hybridization probes. All the four isolates representing pattern 11 and the one in pattern 13 reacted with an enterovirus probe in addition to the HRV specific probe suggesting a close relation of at least some of the ‘HRV 14’ like isolates to the enteroviruses (data not shown). One isolate of pattern 18 hybridized strongly also with a clone containing the first 784 nucleotides of poliovirus type 3 Sabin (Cann et al., 1983) which reacts extensively with the enterovirus reference strains but less with rhinoviruses (Fig. 2~). A Coxsackievirus B3 probe covering 4.3 kb from the 3’ end (Sttilhandske et al., 1984) detected this isolate as efficiently as the enterovirus reference strains (data not shown). Electron microscopic examination of the virus revealed a small round particle with a diameter of approximately 30 rmr. Neutralization test was done with the WHO enterovirus serum pools A to H using standard procedures but no growth inhibition was observed. The acid lability test was repeated, and the infectivity of the isolate was reduced by only a ten-fold dilution-step after acid treatment. Our results favour the possibility that this isolate was an enterovirus which is not neutralized by the serum pools. However, it also shows some properties of human rhinoviruses. Another clinical isolate did not react with any of the probes, which may be due to a low copy number of the viral genome in the infected cells. The nucleic acid spot hybridization method described here is useful for the identification of rhinoviruses after cell culture isolation and by a combination of probes
68
virtually all clinical isolates can be detected. Provided that the genetical reactivity patterns could be correlated to clinical and epidemiological properties, the hybridization assay is a more practical laboratory method for grouping of HRV isolates than the extremely laborious neutralization test with over one hundred antisera. The optimal assay system for a routine diagnosis of rhinovirus infections would be detection of antigens or nucleic acids directly in clinical specimens without cell culture. However, to our experience with twenty of the nasopharyngeal samples, direct detection by HRV-2 and HRV-14 5’ probes is not sensitive enough without prior amplification of the target in cell cultures or in vitro. By optimization of specimen handling, including transport, storage and manipulation, the sensitivity of the hybridization tests might be elevated to a diagnostically relevant level. Acknowledgements
We thank Mrs Marita Maaronen nical assistance.
and Mrs Tuula Lindholm for excellent tech-
References Al-Nakib, W. et al. (1986) Detection of human rhinoviruses and their molecular relationship using cDNA probes. J. Med. Virol. 20, 28%296. Arola, M., Ziegler, T., Ruuskanen, O., Mertsola, J., Nanto-Salonen, K. and Halonen, P. (1988) Rhinoviruses in acute otitis media. J. Pediat. 693-695. Auvinen, P. et al. (1989) Genetic diversity of enterovirus subgroups. Arch. Virol. 104, 175-186. Bruce, C.B., Al-Nakib, W., Almond, J.W. and Tyrrell, D.A.J. (1989) Use of synthetic oligonucleotide to detect rhinovirus RNA. Arch. Virol. 105, 179-187. Callahan, P.L., Mizutami, S. and Colonno, R.J. (1985) Molecular cloning and complete sequence determination of the RNA genome of human rhinovirus 14. Proc. Natl. Acad. Sci. USA 82,732-736. Cann, A.J., Stanway, G., Hauptmann, R., Minor, P.D., Schild, G.C., Clarke, L.D., Mountford, R.C. and Almond, J.W. (1983) Poliovirus type 3: molecular cloning of the genome and nucleotide sequence of the region encoding the protease and polymerase proteins. Nucl. Acids Res. 11,1267-1281. Duechler, M., Skern, T., Sommergruber, W., Neubauer, C., Gruendler, P., Fogy, I., Blaas, D. and Kuechler, E. (1987) Evolutionary relationships within the human rhinovirus genus: comparison of serotypes 89, 2, and 14. Proc. Natl. Acad. Sci. USA 84, 2605-2609. Gama, R.E., Hughes, P.J., Bruce, C.B. and Stanway, G. (1988) Polymerase chain reaction amplification of rhinovirus nucleic acids from clinical material. Nucl. Acids Res. 16, 9346. Gama, R.E., Horsnell, P.R., Hughes, P.J., North, C., Bruce, C.B., Al-Nakib, W. and Stanway, G. (1988) Amplification of rhinovirus specific nucleic acids from clinical samples using the polymerase chain reaction. J. Med. Virol. 28, 73-77. Hughes, P.J., North, C., Jellis, C.H., Minor, P.D. and Stanway, G. (1988) The nucleotide sequence of human rhinovirus 1B: molecular relationships within the rhinovirus genus. J. Gen. Virol. 69,49-58. Hyypil, T., Auvinen, P. and Maaronen, M. (1989) Polymerase chain reaction for human picornaviruses. J. Gen. Virol., in press. Skern, T., Sommergruber, W., Blaas, D., Gruendler, P., Fraundorfer, F., Pieler, C., Fogy, I. and Kuechler, E. (1985) Human rhinovirus 2: Complete nucleotide sequence and proteolytic processing signals in the capsid protein region. Nucl. Acids Res. 13, 2111-2126. Stanway, G., Hughes, P.J., Mountford, R.C., Minor, P.D. and Almond, J.W. (1984) The complete nucleotide sequence of a common cold virus: human rhinovirus 14. Nucl. Acids Res. 12,7859-7875. Stalhandske, P.O.K. et al. (1984) Replicase gene of coxsackievirus B3. J. Virol. 51, 742-746. Gfverstedt, L.G., Hammarstrom, K., Balgobin, N., Hjerten, S., Pettersson, U. and Chattopadhyaya, J. (1984) Rapid and quantitative recovery of DNA fragments from gels by displacement electrophoresis (isotachophoresis). Biochim. Biophys. Acta 782, 120-126.
Elsevier VIRMET 00966
Identification of rhinoviruses by cDNA probes Petri Auvinenl,
Thedi Ziegler’, Tim Skern2, Ernst Kuechleti, Glyn Stanway and Timo Hyypiti’
‘Department of Virology, University of Turku, Turku, Finland, 21nstitutefor Biochemistry, University of Vienna, Vienna, Austria and ‘Department of Biology, University of Essex, Colchester, U.K.
(Accepted 12 September 1989)
Summary We have used nucleic acid hybridization for the detection and grouping of human rhinoviruses (HRV) according to their genetic relationships. Fifteen rhinovirus reference strains, seventy-one clinical isolates and four enteroviruses were propagated in cell cultures, spotted onto membrane filters and hybridized with radioactively labelled cDNA probes covering different parts of the genomes of HRVlB, HRV-2, HRV-14, HRV-85 and HRV-89. When the rhinovirus and enterovirus reference strains were tested, the 5’ probe of HRV-2 hybridized with thirteen of the fifteen HRV reference strains, with poliovirus type 3 and with ECHO virus 11. The HRV-14 5’ probe reacted with eleven HRV reference strains and with all the enteroviruses studied. Sixty-nine of the 71 clinical isolates were recognised by the HRV-2 5’ probe, whereas the HRV-14 probe from the same part of the genome hybridized with 54 field isolates. One of the two isolates that remained negative with the HRV-2 5’ probe was detected with the HRV-2 probe that derived from the P2 region of the genome, and the other isolate was not detected by any of the probes. Probes from other parts than the 5’ end of the genome were generally more specific, and clusters could be formed based on the reactivity of the HRV strains with these probes. Rhinovirus; Nucleic acid hybridization
Correspondence to: P. Auvinen, University of Turku, Department SF-20520 Turku, Finland.
of Virology, Kiinamyllynkatu 13,
0166~0934/90/$03.50@ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
62
Introduction More than one hundred serologically distinct viruses have been classified as human rhinoviruses (HRV), the most frequent agent causing common cold. The viruses have an icosahedral capsid consisting of 60 copies of each of the four structural proteins which enclose a monocistronic, positive strand RNA genome with an approximate length of 7200 nucleotides. With accumulating knowledge about the genomic composition, it has become evident that the different serotypes share much homology including some common sequences, although HRV-14 is less closely related to the others studied (Stanway et al., 1984; Skern et al., 1985; Callahan et al., 1985; Duechler et al., 1987; Hughes et al., 1988). The laboratory diagnosis of HRV infections is currently based on virus isolation; HRVs are identified by their typical cytopathogenic effect and by the lability to acid treatment. Further typing can be done by neutralization tests. These tests, however, are laborious and time-consuming and this has prompted efforts to develop more rapid diagnostic techniques based on nucleic acid hybridization (AlNakib et al., 1986; Bruce et al., 1989). Detection of HRV using enzymatic amplification by the polymerase chain reaction has been reported recently (Gama et al., 1988, 1989). We describe studies on the applicability of cDNA probes representing HRV-lB, HRV-2, HRV-14, HRV-85 and HRV-89 genomes for the detection and identification of HRV reference strains and clinical isolates in cell culture.
Materials and Methods Cells, reference
viruses, clinical samples and antisera
HRV serotypes lB, 2, 3, 9, 11, 12, 13, 14, 22, 29, 36, 38, 39, 48 and 89 were grown in HeLa-Ohio cells (a kind gift of Dr. D.A.J. Tyrrell, Salisbury, U.K.). Coxsackieviruses A9 and B3, ECHO virus 11 and poliovirus type 3 were propagated in LLC-MK2 cells. All reference strains and specific antisera were obtained from the American Type Culture Collection (Rockville, MD). Antisera against HRV 2, 3, 11, 12, 13, 14, 36, 38, 48 and 89 were used for typing of selected isolates by standard neutralization assay. Specimens were collected from adult and paediatric patients with upper respiratory tract infection during a 15 month period between September 1986 and November 1987. HRVs were isolated in HeLa-Ohio cells as described elsewhere (Arola et al., 1988). Preparation of specimens
Cells were inoculated for nucleic acid hybridization with reference viruses and with first passage material of the clinical isolates at a high multiplicity of infection. The cells were scraped into the supernatant when the cytopathic changes were advanced, pelleted by low speed centrifugation and stored at -70°C until used. Un-
63
-
PlYlI
Fig. 1. Localization and the size of the HRV-2 and HRV-14 probes. The map of rhinovirus genome is shown in the upper part of the figure. The codes of the probes are presented on the right.
infected HeLa and LLC cells were used as controls. Prior to hybridization, cells and clinical samples were treated with proteinase K (100 lqjrnl) and 0.5% SDS at 37°C for 1 h. Nucleic acids were extracted with phenol and chloroform, and material corresponding to 2 X 105 cells was spotted, after denaturation by formamide-formaldehyde (WC for 1 h), to a filter membrane (GeneScreen PlusTM, New England Nuclear, Boston, MA) in a manifold apparatus (Schleicher and Schuell, Dassel, F.R.G.). Hybridization procedure
cDNA probes covering different regions of the genomes of HRV-1B (Hughes et al., 1988), HRV-2 (Skem et al., 1985; Fig. l), HRV-14 (Stanway et al., 1984; Fig. l), HRV-85 (Stanway et al., unpublished data), HRV-89 (Duechler et al., 1987) and poliovirus 3 (Cann et al., 1983) were used for the hybridization analysis. The probes contained both the insert and the vector with the exception that the HRV-14 probes 752, 2817 and 3643 were restriction enzyme fragments separated by agarose gel electrophoresis. DNA was eluted from gel slices by isotachophoresis (Gfverstedt et al., 1984). Probes were labelled by nick translation with CX-~~P dCTP (Amersham International plc, Buckinghamshire, U.K.). Hybridization was carried out at 42°C overnight in a solution containing 50% formamide (Auvinen et al., 1989). After hybridization, filters were washed and exposed to X-ray films (Trimax XD, 3M, Milan, Italy).
64
Results Reactivity of the reference
viruses with the cDNA probes
The reactivity of the HRV-2 and HRV-14 probes with the reference viruses is summarized in Table 1. Clone 109 from the 5’ end of HRV-2 gave a positive signal with 13 of the 15 HRV reference strains and also with poliovirus type 3 and ECHO virus 11 (Fig. 2b). After an extended exposure, a weak positive signal was observed also with HRV-29 and HRV-48, and with all the enteroviruses. HRV-11 hybridized only with the probe localized at the 5’ end of HRV-2. Probe 752 from the 5’ non-coding region of HRV-14 reacted with eleven HRV reference strains and with all the enteroviruses. The probes covering the 5’ end of HRV-lB, HRV85 and HRV-89 genomes also gave a positive signal with a majority of the reference strains. The HRV-2 and HRV-14 probes covering the capsid regions distinguished between two groups of reference viruses. The first group contained eight HRV serotypes (2, 9, 12, 13, 22, 36, 38 and 39) while four HRV strains (3, 14, 48 and 89) were in the other group. HRV-2 probe 96, representing the middle part of the TABLE 1 Reactivity of the HRV-2 and HRV-14 probes with the reference virus strains in the spot hybridization test virus strain virus titre
1B 2 3 9 11 12 13 14 22 29 36 38 39 48 89 CA9” CB3b Echo11 Polio3
10-7 lo-’
HRV-14 probes
HRV-2 probes 109
107 132 899
96
24 44
752
2817
3643
PAM2
+ +
+
+
+ +
+ +
_ -
_
+ +
10-7
+
-
-
-
+
+
+
+
lo-’ lo-’ lo-* lo-’ 10-7 lo-’ 10-s 10-s lo-’ lo-’ 10-6 10-7
+ + + + + + + + + +
+ _
-
+ _ _
-
+ + + -
+ + -
+ _ _ + +
+ -
+ -
+
+ + + + + + + +
_ + +
-
_
+ + + +
+ + +
+ + + _
“Coxsackievirus A9; %Zoxsackievirus p3.
+ + -
-
f +
+ + + + + + + +
+
+ + + +
a 18 2 3 9 11 12 13 14
b
c
22 CA9 29cB3 36 El1 38 P3 ,39 LLC 48 89 HeLa
Fig. 2. Examples of hybridization patterns with reference strains using HRV-2 and poliovirus probes. (a) Picornavirus serotypes used in the study. (b) HRV-2 probe covering nucleotides l-685. (c) Poliovirus 3 probe covering nucleotides l-784. lB-89 are rhinovirus serotypes, CA9 = Coxsackievims A9, CB3 = Coxsackievims B3, El1 = ECHO virus 11, P3 = poliovirus 3, LLC and HeLa are control cells.
nonstructural region, reacted more specifically with a restricted number of strains than did the probe deriving from the part coding for the capsid proteins. The HRV14 3’ probe was broadly reactive and recognized 80% of the rhinovirus reference strains and all the enteroviruses studied (Table 1). A probe mapping to position 2.G3.8 kb of the HRV-89 genome reacted with HRV-36, and a probe covering region 6.0-7.2 kb reacted also with HRV-14 (data not shown). Reactivity of the clinical isolates with the probes
Every probe reacted with some of the clinical isolates (Table 2). Probes deriving from the 5’ end showed the widest reactivity. The HRV-14 5’ probe 752 detected 54 of the 71 clinical isolates (76%), and the HRV-2 probe 109 hybridized with all except two (97%). One of these two negative specimens, however, reacted with the HRV-2 probe 96 deriving from the P2 region. The other one remained negative with all the probes. The probes originating from the capsid and the P2 region reacted more selectively with the clinical isolates. The group of clones from the capsid region of HRV2 gave a positive signal with 21 specimens, and the HRV-14 fragment detected 18 isolates. The highest specificity was observed with the clone 96 from the P2 region which detected only four of the clinical isolates. The broadly reacting HRV-14 3’ probe PAM2 recognized 40 isolates (56%). Twenty of the clinical samples were also tested by direct hybridization. Only one of them was positive when the combination of both HRV-2 and HRV-14 5’ probes were used.
66 TABLE 2 Reactivity patterns of the HRV-2 and HRV-14 probes with clinical isolates Reactivity pattern
HRV-14 probes
HRV-2 probes 109
107 132
96
24 44
752
2817
+
+ +
+ + + -
-
+ + -
+
+ +
-
+
+ + _ +
-
+
+ + -
7
54
3643
PAM2
No. of reactive isolates
899 1
+
2 3 4 5 6 7
+ + + + + _
8 9 10 11 12 13 14 15 16 17 18 19 20
+ + + + + + + + + + + + _
Number 69 of reactive isolates
+ + + + + + + + + -
21
4
-
+ _
-
+ + + + + + + + + -
9 3 2 2 2 2 1 1 1 2 11 6 1 1 1 9 8 6 1 1
20
40
71
-
-
18
Neutralization of selected strains
Twenty-seven isolates, selected from different reactivity groups, were subjected to neutralization test against one or more HRV sera to evaluate the correlation between hybridization and serotyping. Antisera against serotypes 13 (anti-HRV 13) and 36 were used to neutralize 5 isolates of pattern 1, anti-HRV 12 for 2 isolates of pattern 6, anti-HRV 2 and 38 for the isolate of pattern 8, anti-HRV 3, 14, 48 and 89 for the 18 isolates of patterns 11,12,13, 14 and 15 and anti-HRV 11 for the isolate of pattern 17. One isolate of reaction pattern 12 was neutralized by antiHRV 14. No neutralization was observed with the other combinations.
Discussion
We used nucleic acid hybridization for grouping rhinoviruses on the basis of their genetic similarity. HRV-2 and HRV-14 5’ probes showed high conservation of this
part of the genome among rhinoviruses. Other parts were more specific in their reactivity forming clusters of closely and more distantly related viruses. Based on the reaction patterns obtained with probes covering the capsid region of HRV-2 and HRV-14, the reference viruses included in this study could roughly be classified into three groups (Table 1). The Iirst group consisted of eight serotypes closely related to HRV-2, four rhinoviruses were detected by HRV-14 capsid fragment, and rhinoviruses lB, 11 and 29 were not reactive. In general, the HRV-14 probes were more likely to hybridize with the enteroviruses than were the HRV-2 clones. This finding is in agreement with data obtained from nucleotide sequencing, which indicate that HRV-14 is more closely related to polioviruses than any of the other rhinoviruses sequenced so far (Stanway et al., 1984). However, the 3’ end of HRV14 was reactive with many rhinoviruses. Our findings agree well with the results of Al-Nakib et al. (1986), with the exception that we observed a closer relationship between HRV-14 and HRV-48. Twenty different reaction patterns were found with the clinical isolates (Table 2). They could be divided into ‘HRV-2 like’ and ‘HRV-14 like’ patterns mainly on the basis of their reactivity with probes covering the Pl and P2 regions. Reaction patterns l-8 belong to the former group, and 11-15 to the latter. Reaction patterns 9 and 10 form intermediates between these two groups while 16-20 do not fall into any of these categories and they apparently represent genetically more distinct viruses. No clustering of isolates of either group was observed during the time the specimens were collected, indicating that viruses of all groups can circulate simultaneously in the population. Fourteen of the clinical isolates were also tested after a polymerase chain reaction (Hyypl et al., in press) using synthetic oligonucleotides as hybridization probes. All the four isolates representing pattern 11 and the one in pattern 13 reacted with an enterovirus probe in addition to the HRV specific probe suggesting a close relation of at least some of the ‘HRV 14’ like isolates to the enteroviruses (data not shown). One isolate of pattern 18 hybridized strongly also with a clone containing the first 784 nucleotides of poliovirus type 3 Sabin (Cann et al., 1983) which reacts extensively with the enterovirus reference strains but less with rhinoviruses (Fig. 2~). A Coxsackievirus B3 probe covering 4.3 kb from the 3’ end (Sttilhandske et al., 1984) detected this isolate as efficiently as the enterovirus reference strains (data not shown). Electron microscopic examination of the virus revealed a small round particle with a diameter of approximately 30 rmr. Neutralization test was done with the WHO enterovirus serum pools A to H using standard procedures but no growth inhibition was observed. The acid lability test was repeated, and the infectivity of the isolate was reduced by only a ten-fold dilution-step after acid treatment. Our results favour the possibility that this isolate was an enterovirus which is not neutralized by the serum pools. However, it also shows some properties of human rhinoviruses. Another clinical isolate did not react with any of the probes, which may be due to a low copy number of the viral genome in the infected cells. The nucleic acid spot hybridization method described here is useful for the identification of rhinoviruses after cell culture isolation and by a combination of probes
68
virtually all clinical isolates can be detected. Provided that the genetical reactivity patterns could be correlated to clinical and epidemiological properties, the hybridization assay is a more practical laboratory method for grouping of HRV isolates than the extremely laborious neutralization test with over one hundred antisera. The optimal assay system for a routine diagnosis of rhinovirus infections would be detection of antigens or nucleic acids directly in clinical specimens without cell culture. However, to our experience with twenty of the nasopharyngeal samples, direct detection by HRV-2 and HRV-14 5’ probes is not sensitive enough without prior amplification of the target in cell cultures or in vitro. By optimization of specimen handling, including transport, storage and manipulation, the sensitivity of the hybridization tests might be elevated to a diagnostically relevant level. Acknowledgements
We thank Mrs Marita Maaronen nical assistance.
and Mrs Tuula Lindholm for excellent tech-
References Al-Nakib, W. et al. (1986) Detection of human rhinoviruses and their molecular relationship using cDNA probes. J. Med. Virol. 20, 28%296. Arola, M., Ziegler, T., Ruuskanen, O., Mertsola, J., Nanto-Salonen, K. and Halonen, P. (1988) Rhinoviruses in acute otitis media. J. Pediat. 693-695. Auvinen, P. et al. (1989) Genetic diversity of enterovirus subgroups. Arch. Virol. 104, 175-186. Bruce, C.B., Al-Nakib, W., Almond, J.W. and Tyrrell, D.A.J. (1989) Use of synthetic oligonucleotide to detect rhinovirus RNA. Arch. Virol. 105, 179-187. Callahan, P.L., Mizutami, S. and Colonno, R.J. (1985) Molecular cloning and complete sequence determination of the RNA genome of human rhinovirus 14. Proc. Natl. Acad. Sci. USA 82,732-736. Cann, A.J., Stanway, G., Hauptmann, R., Minor, P.D., Schild, G.C., Clarke, L.D., Mountford, R.C. and Almond, J.W. (1983) Poliovirus type 3: molecular cloning of the genome and nucleotide sequence of the region encoding the protease and polymerase proteins. Nucl. Acids Res. 11,1267-1281. Duechler, M., Skern, T., Sommergruber, W., Neubauer, C., Gruendler, P., Fogy, I., Blaas, D. and Kuechler, E. (1987) Evolutionary relationships within the human rhinovirus genus: comparison of serotypes 89, 2, and 14. Proc. Natl. Acad. Sci. USA 84, 2605-2609. Gama, R.E., Hughes, P.J., Bruce, C.B. and Stanway, G. (1988) Polymerase chain reaction amplification of rhinovirus nucleic acids from clinical material. Nucl. Acids Res. 16, 9346. Gama, R.E., Horsnell, P.R., Hughes, P.J., North, C., Bruce, C.B., Al-Nakib, W. and Stanway, G. (1988) Amplification of rhinovirus specific nucleic acids from clinical samples using the polymerase chain reaction. J. Med. Virol. 28, 73-77. Hughes, P.J., North, C., Jellis, C.H., Minor, P.D. and Stanway, G. (1988) The nucleotide sequence of human rhinovirus 1B: molecular relationships within the rhinovirus genus. J. Gen. Virol. 69,49-58. Hyypil, T., Auvinen, P. and Maaronen, M. (1989) Polymerase chain reaction for human picornaviruses. J. Gen. Virol., in press. Skern, T., Sommergruber, W., Blaas, D., Gruendler, P., Fraundorfer, F., Pieler, C., Fogy, I. and Kuechler, E. (1985) Human rhinovirus 2: Complete nucleotide sequence and proteolytic processing signals in the capsid protein region. Nucl. Acids Res. 13, 2111-2126. Stanway, G., Hughes, P.J., Mountford, R.C., Minor, P.D. and Almond, J.W. (1984) The complete nucleotide sequence of a common cold virus: human rhinovirus 14. Nucl. Acids Res. 12,7859-7875. Stalhandske, P.O.K. et al. (1984) Replicase gene of coxsackievirus B3. J. Virol. 51, 742-746. Gfverstedt, L.G., Hammarstrom, K., Balgobin, N., Hjerten, S., Pettersson, U. and Chattopadhyaya, J. (1984) Rapid and quantitative recovery of DNA fragments from gels by displacement electrophoresis (isotachophoresis). Biochim. Biophys. Acta 782, 120-126.