key: cord-0004905-wyzb8v4a authors: Forsyth, M.; Al-Nakib, W.; Chadwick, P.; Stanway, G.; Hughes, P. J.; Leckie, G.; Almond, J. W.; Tyrrell, D. A. J. title: Rhinovirus detection using probes from the 5′ and 3′ end of the genome date: 1989 journal: Arch Virol DOI: 10.1007/bf01313878 sha: 505b69e3a6ab2be5e26ef5a86febcd4d5d4abb14 doc_id: 4905 cord_uid: wyzb8v4a This study investigated the abilities of cDNA probes from the 5′ and 3′ ends of the genome of human rhinoviruses (HRV-) 14, 9, and 1B to detect RNA from 59 rhinovirus serotypes. The results show that probes from the 5′ end of the genomes of HRV-14, 9, and 1B detected a large number of serotypes but the detection rate was variable and depended on the degree of homology with the particular probe. In contrast, all the 3′ end probes were specific for the homologous virus. However, along HRV-9 probe detected a large number of serotypes. It was concluded that such cDNA probes would not detect all serotypes with equal efficiency. Synthetic oligonucleotides corresponding to short but highly conserved regions in the 5′ non coding region may overcome this problem. Rhinoviruses are the major causative agents of the common cold [9] . In the majority of healthy individuals, the infection results in a short illness of some 3-5 days duration characterized by rhinorrhoea, nasal obstruction, sore throat and pharyngitis [4] . However, in immunocompromised individuals particularly children and in patients with obstructive airways disease, rhinovirus infection may result in more serious lower respiratory tract involvement [11, 14] . Furthermore, recent community studies in Michigan, U.S.A., suggested that rhinoviruses can be isolated from up to 70 percent of adults (over 40 years of age) with lower respiratory tract involvement [15] . In these individuals the median duration of illness was as long as 3 weeks [15] . We have recently shown that a new synthetic anti-rhinovirus agent, R61837, can sucessfully suppress illness in volunteers challenged with a rhinovirus [5] . It is anticipated that with further progress in the field it may be possible to treat these infections. However, rapid virus identification prior to treatment would be essential since these antivirals are specific for rhinoviruses. Until recently, rhinoviruses could only be identified by growth in a sensitive cell culture. Such procedures are time consuming, labour intensive and require considerable expertise. Although, it is now possible to detect rhinovirus antigens directly in nasal washings using immunologically based methods such as ELISA [10] , the diversity of serotypes, recently estimated to be around 100 [12] makes efficient detection of all serotypes difficult. We have therefore, attempted to overcome this problem by developing procedures based on R N A detection. A previous study has shown that a c D N A probe from the 5' non-coding region of HRV-14 detects 96.4% of the 54 rhinoviruses investigated, but the sensitivity of detection was variable and presumably depended on the degree of genomic homology of particular serotypes with HRV-14 [2] . In this study we therefore increased the number of probes used to include the 3' ends of HRV-14, 1B and 9 and the 5' ends of H R V -1 B [13] and 9 [Leckie et al., in prep.] . The aim was to find a probe that would hybridise with R N A from various serotypes with equal efficiency and we therefore studied the reactions of the new probes with R N A from 59 rhinovirus serotypes. Stocks of rhinoviruses including animal rhinoviruses such as the calf rhinovirus SDI and bovine rhinovirus EC 11 and other control viruses, namely influenza A and B (FLU A and B), coronavirus 229E and coxsackie A21 (COXA21) were prepared as previously described [2] . Each stock was titrated in microtitre plates and titres expressed as TCIDs0/ml. Table 1 shows the final titres of the virus stocks as used in this study. Viral RNA was extracted using the method of Rotbart et al. [16, 17] . Briefly, 0.2ml of each virus preparation was mixed with an equal volume of a 3 : 2 mixture of 20 x SSC--37% formaldehyde. The mixture was then spotted onto nitrocellulose filters that had been pre-soaked in 20 x SSC as described earlier [2] . Details of the methods used to prepare and label probes used in this study have been described previously [2, 3] . Briefly, probes were produced from M 13 templates containing rhinovirus cDNA cloned in the appropriate orientation. The HRV-14 5' construct contained nucleotides 1-802 and was produced by cutting a recombinant plasmid with PstI and BglII and ligating into the M13 mpl9 PstI and BamHI sites. The corresponding 3' probe comprised positions 6336-7167 contained within an HpaI fragment ligated into M 13 cut with Sinai. The HRV-1B probes represented positions 1-846 (5') located within a PstI HindIII fragment which was ligated into these sites of M 13 and 6338-7133 (3') located in a PstI fragment. The HRV-9 5' end probe was prepared from a 3.3 kb cDNA clone of HRV-9 designated pRg112. A 472 base pair fragment representing nucleotides 1-472 of the HRV-9 sequence (unpublished) was subcloned into M 13 mp 18 in the positive sense orientation. Both of the 3' end probes used in this study were prepared from a 1.1 kb HRV-9 cDNA clone pR9193 covering the 3' region of the genome. The short 3' end probe was prepared from a 331 base pair fragment representing nucleotides 6768-7098. The total length of the HRV-9 genome is 7128 excluding poly A tail. The long 3' end probe was prepared from a 769 base pair fragment from pR9193 covering nucleotides 6005-6773. In both cases the fragments were subcloned into the phage vector M 13 mp 18 in the positive sense orientation. The templates were used to produce radioactive cDNA probes complementary to viral sense RNA by extension of an M 13 universal primer in a reaction performed by the Klenow fragment of DNA polymerase 1. Annealing of the primer/template was achieved by boiling together (3 min) in a mixture (20 lal) comprising primer (5 ng), template (1 lag), 15 mM Tris-HC1, pH 8.0, and 7.5 mM MgCI2. The solution was allowed to cool to room temperature and to it was added dGTP, dCTP dTTP (to a final concentration of 0.5 raM), 32p_ dATP (20 laCi) Klenow fragment (5 units) and water to give a final volume of 50 ~tl. The reaction took place at room temperature for 30 minutes after which the radioactive DNA was separated from unincorporated nucleotides by passage through a Sephadex G 100 column. The probes were hybridized with viral RNA as described previously [2, 3] . The strength of the hybridization signals was assessed visually by two independent observes and the signal was classified as very strong (+ + + +), strong (+ + + ), good (+ +), positive (+), weak (4-), or no signal (0). Comparison of the hybridization results with the titres of the virus stocks (Table 1) shows that generally there is no direct relationship between the titres and the strength of the hybridization reaction. For example, the HRV-14 probe hybridised very strongly (+ + + +) with HRV-49 even though the titre was low (< 104 TCIDs0/ml) but only weakly with HRV-1A ( + ) which had one of the highest titres of the viruses tested (> 108 TCIDs0/ml) ( Table 2) . It therefore appears that the efficiency of detection is more directly related to other factors, the most important of which is probably the degree of RNA homology between the RNA of the different rhinoviruses. Table 2 shows the strength of the signal observed using the various viruses and seven probes. It can be seen that the reactions varied greatly in intensity. Thus, the HRV-14, 5' end probe gave a very strong signal (+ + + +) when hybridized with RNA from HRV-3, 4, and 49 and a strong signal (+ + +) when reacted with RNA from HRV-2, 41, 47, 56, 62, 72, and 85. Similarly, a 5' end probe from HRV-9 gave a very strong signal (+ + + +) with RNA from HRV-15, 31, and 32. The 5' end probe from HRV-1 B hybridized very strongly (signal + + + + ) with RNA from HRV-49 and 85 and strongly (+ + + ) with RNA from HRV-1A, 15, and 19. In contrast to the 5' end probes, those from the 3' end (831,795, and 331 nucleofides in length for HRV-14, 1 B, and 9, respectively) detected only the homologous virus in the conditions of the assay. However, the longer probe (768 nucleofides in length) from HRV-9, detected many more viruses (Table 2) . Indeed, this probe hybridized very strongly (+ + + + ) with RNA from HRV-13, 24, 27, 32, 64, 73, and 75 and reacted strongly (signal + + +) with RNA from HRV-11, 15, 18, 41, and 65 suggesting that these viruses are closely related to HRV-9 in the 3' end region of the genome and perhaps reflects the conservation of the polymerase sequences among these viruses. Both 5' and 3' end probes from HRV-14, -9, and -1B reacted with their respective viruses very strongly (+ + + + ). Table 3 shows the percentage of rhinovirus serotypes detected by each probe according to the strength o f the hybridization signal. Thus 93.2, 66, and 74.5% of viruses investigated were detected (signal > + ) by HRV-14, 9, and 1B, 5' end probe, respectively, while 71% were detected by the long 3' end HRV-9 probe. Similarly, 45.8, 20.3, and 37.3% of rhinoviruses gave a good hybridization signal ( > + + ) with 5' end probes from HRV-14, 9, and 1 B, respectively, while 35.6% gave a similar signal with the long HRV-9, 3' end probe. As can be seen from Table 2 , none of the control respiratory viruses such as influenza A, B, and coronavirus 229E, gave positive hybridization signals in any experiments during this study. Coxsackie A21, gave a positive signal with HRV-14 5' end probe suggesting some genomic homology with HRV-14. These hybridization tests were repeated three or more times and the results were shown to be reproducible. The data suggest that c D N A hybridization with different probes show a different relationship between rhinovirus serotypes from that based an other properties. For example, HRV-15 which shares the same cellular receptor as HRV-14 (both are included in the major receptor group) [1, 7] reacted more strongly (signal + + + ) with the 5' end probe from HRV-1 B, a serotype in the minor receptor group than with HRV-14 (signal +). Similarly, HRV-2 which shares the same cellular receptor as HRV-1 B (both are included in the minor receptor group) [1, 7] reacted more strongly (signal + + + ) with the 5' end probe from HRV-14, a serotype from the major receptor group, than with HRV-1B (signal +). Moreover, this relationship is also different from that based antigenic crossreactivity [8] . It is interesting to note that R N A from HRV-49 hybridized extremely well (signal + + + +) with both HRV-14 and 1B 5' end probes suggesting a strong genomic homology between HRV-49 and these two viruses. Similarly, R N A from HRV-15 reacted very well with the 5' end probe from both HRV-9 and 1 B indicating a close genomic relationship between HRV-15 and these two serotypes in the 5' end. Furthermore, RNA from HRV-15 and 32 hybridized extremely well with both 5' and 3' end probes from HRV-9 implying that these viruses have strong genomic homology with HRV-9 in both ends of the genome. In contrast, RNA from HRV-45, 5t, 70, and 82 (signal < +) and 8 and 81 (signal + to 4-) did not hybridize well with any of the probes investigated probably indicating that these viruses are more divergent from HRV-14, 9, and lB. The findings of this study are that probes from the 5' end of the genome of rhinoviruses detect a large number of rhinoviruses, although the detection rate is variable and apparently depends on the strength of genornic homology among the different serotypes. In contrast, probes from the 3' end of the genome (of some 800 nucleotides) of HRV-14 and 1 B detected only the homologous virus under the hybridization conditions of the assay. However, a similar size probe from the 3' end of HRV-9 detected many more serotypes. In contrast, a shorter probe (331 nucleotides in length) also from the 3' end of HRV-9, detected only the homologous virus, thus indicating that the detection rate is highly influenced by probe length. Both 5' and 3' end probes detected the homologous viruses with equal efficiency. It was interesting to note that the 5' end HRV-14 probe was more efficient than the other probes in detecting a larger number of rhinoviruses. This is somewhat surprising since comparative sequence analysis indicates that HRV-14 is relatively diverse from the majority of rhinoviruses studied. It might therefore be thought that probes from HRV-1 B and HRV-9, which are more typical rhinoviruses, would prove to give a higher detection rate. The data presented in this paper are interesting in that they show that cDNA probes are unlikely to be useful in detecting all rhinoviruses with equal efficiency despite an earlier prediction that the 5' end non-coding region was likely to be relatively highly conserved throughout the rhinovirus genus [ 18] . These results are therefore in agreement with our earlier findings with the HRV-14 probe [2] and show that although there is considerable homology in the 5' end noncoding region of many rhinoviruses it is still not sufficient for a probe prepared from this region to detect all the different rhinovirus serotypes with equal efficiency. Furthermore, in tests on clinical material, both the identity of the infecting serotype and its concentration in nasal secretion would vary widely. In this study there were great variations in the signal given by the different serotypes even though the titres of virus used were usually greater than 10 4 TCIDs0/ml which is much higher than that normally found in nasal washing (often < l 0 2 TCIDs0/ml). Recent work with synthetic oligonucleotides corresponding to short but highly conserved regions in the 5' end non-coding region of the rhinovirus genome [18] suggests that such probes will detect all rhinovirus serotypes with equal efficiency [6] . Further studies are in progress. Many rhinovirus serotypes share the same cellular receptor Detection of human rhinoviruses and their molecular relationship using cDNA probes Rhinovirus detection by cDNA: RNA hybridization Common cold viruses--rhinoviruses Suppression of colds in human volunteers challenged with rhinoviruses by a new synthetic drug (R61837) Synthetic oligonucleotides as diagnostic probes for rhinoviruses Isolation of a monoclonal antibody that blocks attachment of the major group of human rhinoviruses Antigenic grouping ofg0 rhinovirus serotypes The common cold: control? Direct detection of rhinoviruses by an enzyme-linked immunosorbent assay Provocation of airflow limitation by viral infection: implication for treatment A collaborative report: rhinovirus---extension of the numbering system from 89 to 100 The nucleotide sequence of human rhinovirus 1 B: molecular relationships within the rhinovirus genus The association of rhinoviruses with lower respiratory tract disease in hospitalized patients Rhinovirus infections in Tecumseh, Michigan: frequency of illness and number of serotypes Use of subgenomic poliovirus DNA hybridization probes to detect the major subgroups of enteroviruses Factors affecting the detection of enteroviruses in cerebrospinal fluid with coxsackievirus B3 and poliovirus I cDNA probes Rhinovirus detection using cDNA probes 63 The complete nucleotide sequence of common cold virus: human rhinovirus 14 Dr. Glyn Stanway is supported by a Grant from the Medical Research Council number G860897CA.