Murine Pneumonia Virus: Seroepidemiological Evidence of Widespread Human Infection

Transcription

1 J. gen. Virol. (1986), 67, Printed in Great Britain 975 Key words: P VM/ Paget's bone disease/seroepidemiology Murine Pneumonia Virus: Seroepidemiological Evidence of Widespread Human Infection By C. R. PRINGLE 1. AND R. P. EGLIN 2 l Department of Biological Sciences, University of Warwick, Coventry CV4 7AL and 2Department of Virology, John Radcliffe Hospital, Oxford OX3 9DU, U.K. (Accepted 3 March 1986) SUMMARY More than 75 ~ of a random sample of adult human sera exhibited moderate to high murine pneumonia virus (PVM)-neutralizing activity. There was no correlation between PVM-neutralizing activity and respiratory syncytial virus or parainfluenza type 3 virus-neutralizing activities of the same sera. In children the proportion of sera with moderate to high titres increased with age, indicating early exposure to infection. Seroconversion (i.e. > fourfold increase in titre) was observed in four of 108 paired samples of previously undiagnosed respiratory infections. These observations suggest that the human population is frequently exposed to infection with PVM or an antigenically related virus. The sera of patients suffering from Paget's disease of bone tended to exhibit higher than normal PVM-neutralizing titres in comparison with the sera of patients with other bone diseases. Thus, PVM (or an antigenically related virus) resembles some other parainfluenza viruses in being circumstantially associated with Paget's disease of bone. INTRODUCTION Murine pneumonia virus (PVM) and respiratory syncytial (RS) virus are the sole members of the genus Pneumovirus of the family Paramyxoviridae. Human RS virus is the principal cause of respiratory illness in infancy and is regularly isolated during annual epidemics. Bovine RS virus is considered to be a significant cause of respiratory disease in cattle and is frequently isolated from affected animals. The human and bovine RS viruses are antigenically very similar. Although RS virus antibodies have been detected in several other species of domestic, laboratory and wild animals (Lundgren et al., 1969; Pringle & Cross, 1978; Richardson-Wyatt et al., 1981 ; C. R. Pringle, unpublished data), no RS-related viruses have been isolated from these animals. PVM, which is antigenically distinct from the RS viruses, is associated with latent or inapparent infection of mice (Horsfall & Hahn, 1939, 1940) and is prevalent in most colonies of laboratory mice (Parker et al., 1966; Gannon & Carthew, 1980). Serologically related viruses have been isolated from Syrian hamsters, cotton rats and rabbits (Eaton & van Herick, 1944; Horsfall & Curnen, 1946). PVM-neutralizing activity has been detected in several other species (Horsfall & Hahn, 1940; Horsfall & Curnen, 1946), but no PVM-like viruses have been isolated. In the 1940s, Horsfall and colleagues reported serological evidence of PVM infection in a minority of the human population, and in the search for causative agents responsible for primary atypical pneumonia recovered a number of agents from patients which induced PVMneutralizing antibodies in experimental animals (Horsfall et al., 1943). The human agents were considered to be viruses antigenically related to PVM, but the data are difficult to interpret now since these agents could only be monitored by their ability to induce PVM-neutralizing antibodies in animal hosts known to harbour PVM. Paget's disease of bone is a common chronic disorder of bone metabolism of unknown aetiology affecting the middle aged and elderly. Certain features of the disease suggest that it is the result of slow progressive virus infection (Rebel et al., 1977; Singer, 1980). The lesions are focal with nuclear and cytoplasmic inclusions invariably observed in abnormal osteoclasts SGM

2 976 C. R. PRINGLE AND R. P. EGLIN present in the lesions. Studies with immunofluorescent polyclonal and monoclonal antibodies have demonstrated the presence of antigens of measles virus, simian virus 5 (SV5), parainfluenza type 3 (PI3) virus (Basl~ et al., 1985; N. Watt, unpublished data), and RS virus (Mills et al., 1981, 1982) in the osteoclasts of affected tissue. Basl~ et al. (1986) have reported in addition detection of measles virus RNA sequences in affected tissue by in situ hybridization using a nucleocapsid-specific cdna probe. The significance of these findings remains to be established. In the course of surveying patients with Paget's disease of bone for the presence of anti-paramyxovirus antibodies, it became apparent that the sera of the majority of patients in both the affected and control groups possessed PVM-neutralizing activity. Furthermore, as described below, a more marked association between high virus-neutralizing titres and confirmed Paget's disease of bone was observed with PVM, than was observed previously in screening for PI3 virus and RS virus antibodies (Pringle et al., 1985). Because of the unexpectedly high frequency of PVM-neutralizing activity in the sera of the bone clinic patients, the survey was extended to provide an estimate of the prevalence of PVMneutralizing antibody in the general population. The results suggest that exposure to infection by PVM, or an as yet uncharacterized serologically related virus, is widespread and frequent, and that infection occurs early in life and may be accompanied occasionally by respiratory illness. METHODS Virus. PVM strain 15 was obtained from the American Type Culture Collection and cloned by three sequential isolations from single plaques on BHK-21 cell monolayers (Cash et al., 1979). Cells. The BS-C-I line of African green monkey kidney cells was used for assay of PVM infectivity. Infected monolayers were incubated for 7 days at 37 C under Eagle's MEM (Glasgow modification) with 1 ~ foetal calf serum (Flow Laboratories) and 0.9~ agar. The small deeply stained foci were counted after fixing with 1 glutaraldehyde and staining with crystal violet. Standard neutralization test. PI3 virus- and RS virus-neutralizing activities were assayed essentially as described by Pringle & Cross (1978) and Pringle et al. (1985). In the case of PVM, approximately p.f.u, were mixed with an equal volume of twofold dilutions of the test serum. The dilution fluid was Eagle's MEM with 1 ~ foetal calf serum. The test sera were pretreated for 30 min at 56 C to destroy non-specific inhibitors. The virus-serum mixtures were held at 4 C for 18 h approximately and the residual infectivity was assayed by titration on BS-C-1 monolayers. The dilution of serum producing a 50~ reduction of infectivity was determined graphically and the titre of the serum is expressed as the reciprocal of this value. All determinations were carried out in duplicate. Up to 200 foci could be counted on a 35 mm Petri dish before loss of accuracy occurred due to overlap. Abbreviated neutralization test. For speed and economy an abbreviated neutralization test was employed as described in the text. Two dilutions only were used in place of the twofold dilution series of the standard test. The abbreviated test did not give a precise endpoint but was sufficient to allow reproducible classification of sera into low (< 40), intermediate (> 40 but < 320) and high (> 320) activity classes. Serum samples. The serum samples taken from patients attending the Bone Disease Clinic at the Glasgow Western Infirmary between November 1981 and November 1982 have been described previously (Pringle et al., 1985). Thirty-six samples were obtained from patients with Paget's disease and 137 from patients with other bone conditions. The mean age of the patients in the Paget's group was 67 years for men and 71 years for women, and 59 years for men and 62 years for women in the non-paget's group. The following series of samples were obtained from the Virology Department of the John Radcliffe Hospital: 126 samples from young adult women collected during screening for rubella virus antibody, 198 age-grouped samples from patients attending the John Radcliffe Hospital for various reasons, and 108 paired acute and convalescent serum samples from patients with respiratory illness of undetermined aetiology. RESULTS PVM-neutralizing activity in the sera of patients with Paget's disease of bone A collection of serum samples obtained from 173 patients attending the Bone Disease Clinic of the Western Infirmary, Glasgow from November 1981 to November 1982 was screened for RS virus- and PI3 virus-neutralizing activities to investigate further the possible association of Paget's disease of bone with paramyxovirus infection (Basl6 et al., 1979, 1985; Howatson & Fournier, 1982; Mills et al., 1984; Pringle et al., 1985; Rebel et al., 1980; Singer, 1980; Singer & Mills, 1977, 1983). The same sera, which comprise samples from 36 patients with confirmed

3 PVM-neutralizing activity in human sera ;d-l; L: L; :~ v ~ 10 x x x x ~ x~x~y v v v v ~ o ~)~ ~< /> 1280 Titre Fig. 1. Percentage distribution of anti-pvm neutralization titres in the sera of patients attending a bone disease clinic. [~, Paget's disease of bone patients; II, patients with other conditions. 15.=~ 10 0 < >1280 Titre Fig. 2. Distribution of neutralizing activities against PI3 virus, PVM and RS virus in sera from 23 Paget's patients showing the lack of correlation. II, Anti-PVM neutralization titre; l~, anti-rs virus neutralization titre; [-q, anti-pi3 virus neutralization titre. Paget's disease of bone and 137 sera from patients with other conditions, have now been screened for PVM-neutralizing activity using the standard neutralization test. The results are shown in Fig. 1. In contrast to the results obtained previously for RS virus- and PI3 virusneutralizing activities, there was a marked tendency for Paget's disease patients to have higher PVM-neutralizing titres in their sera than other patients. The data which are plotted as a histogram in Fig. 1 represent the entire collection of sera, but a similar result was obtained using a subset of age- and sex-matched sera (not shown); therefore, the lack of age- and sex-matching did not appear to bias the result. Fig. 2 compares the RS virus-, 'PI3 virus- and PVM-neutralizing activities of the sera of 23 of the patients with Paget's disease of bone. This figure shows that there was no correlation

4 978 C. R. PRINGLE AND R. P. EGLIN Table 1. PVM-neutralizing activity in the sera of randomly selected young adults PVM neutralization titres h f Moderate Sera Total Low (~<40) (>40 <320) High (~>320) n Table 2. PVM-neutralizing titres in children and adults f Distribution of neutralizing activities (~) ~k No. in Moderate Age (years) group Low (~<40) (>40 <320) High (>/320) 0" * Over Total * The peak group in each age group is indicated by bold type. between these three activities, and that the anti-pvm activities appeared to fall into two distinct groups, i.e. those in the range < 40 to 80, and those in the range 320 to > There was also no relationship between the PVM-, PI3 virus- and RS virus-neutralizing activities of the individual sera of normal adults (data not shown). The PVM-neutralizing activity was presumed to be antibody-mediated because in five sera chosen at random it was depleted by adsorption with Staphylococcus aureus, Cowan I strain. These sera were also able to precipitate the virus nucleocapsid protein from PVM-infected cells, and not the RS virus nucleocapsid protein from RS virus-infected cells, which is consistent with the presence of a specific immunoglobulin (data not shown). PVM-neutralizing activity in the sera of young adults Since the greatest proportion of anti-pvm titres were in the high (> 1280) titre range, it was decided to carry out a more extensive survey of the prevalence of anti-pvm neutralizing activity in the general population. An abbreviated neutralization test was devised to carry out this survey quickly and economically, taking account of the precision of the plaque assay for PVM on BS-C- 1 monolayers which permitted accurate counting of up to 100 foci per well of a 12-well Costar plate. This test involved use of two dilutions 0/40 and 1/320) of the test serum and a 1/40 dilution of a negative control. The 50% plaque reduction titres were determined by graphical extrapolation from plaque counts on duplicate plates. The titres have been grouped into three broad categories for ease of presentation: i.e. low titres of 40 or less, moderate titres of greater than 40 and less than 320, and high titres of 320 and above. This abbreviated assay proved to be convenient, economical and accurate in distinguishing sera into low, moderate and high activity categories. An estimate of the neutralization titres of a pair of serum samples could be obtained on a single 12-well Costar plate. The results in Tables 1 to 3 were obtained by this means. Samples from young adult women collected in the course of screening for rubella virus infection in the Oxford area were examined to counterbalance the data in Fig. 1 obtained predominantly from elderly patients residing in the west of Scotland. The data in Table 1 indicate that moderate to high levels of PVM neutralizing activity were frequent in this group of

5 PVM-neutralizing activity in human sera 979 Table 3. Distribution of serum titres in 104 paired acute and convalescent sera from patients with respiratory infections of unknown aetiology and exhibiting no seroconversion PVM neutralization titres A Paired Moderate samples Total Low (~<40) (>40 <320) High (>-320) n Table 4. Ser c nversi ninf ur f 8samp esfr mpatientswithrespirat ryillness funkn wnaeti l gy PVM neutralization titre A Age Acute phase Convalescent Patient (years) Diagnosis serum phase serum Y.D. 56 Pyrexia, URTI A.S. 24 Pneumonia, pleurisy <40 >320 J.S. 50 Pyrexia, URTI < 40 > 320 M.T. 20 Recurrent URTI, 320 > 1280 depression sera also, and that at least 90~ of this population had experienced PVM infection by early adulthood. The prevalence of anti-pvm activity in the sera of this population was similar to that observed in the bone disease clinic patients in Glasgow. Prevalence of PVM-neutralizing activity according to age Table 2 presents an analysis of the PVM-neutralizing activities of 324 individuals grouped by age. Childen of 6 months to 1 year exhibited high and low titres in equal proportion, whereas children 1 to 2 years of age had predominantly low titres. The proportion of individuals with high titres then increased rapidly so that the sera of the majority of children possessed high PVM-neutralizing activity by 3 to 4 years of age. Thereafter, the proportion of high anti-pvm titres fluctuated somewhat, but the proportion of sera with moderate to high activities was close to the level observed in adults. The figures for the over- 15 age group include the data tabulated in Table 1. The data in Tables 1 and 2 are consistent with a hypothesis of waning maternal immunity and exposure to infection early in life. Seroconversion In order to establish whether PVM-neutralizing activity was associated with respiratory infection, a collection of 108 paired serum samples from patients with upper respiratory tract infections (URTI), atypical pneumonia or bronchitis of unknown aetiology was screened for anti-pvm activity. One-hundred and four of the paired samples showed no seroconversion, the majority of the titres being in the high range (Table 3). The four samples listed in Table 4 exhibited significant seroconversions (i.e. more than fourfold difference between acute and convalescent sera). These results suggest that a virus similar to PVM may be responsible for 3 to 4~ of otherwise unaccounted-for respiratory infections. The data in Tables 1 to 4 together suggest frequent exposure of the general population to PVM or a PVM-like agent. DISCUSSION RS virus and PVM, although classified on morphological grounds as the sole members of the genus Pneumovirus, nevertheless exhibit significant differences. In particular, several additional low molecular weight proteins are associated with PVM (Cash et al., 1979). The recent molecular cloning and subsequent determination of the genetic organization of RS virus (Collins et al., 1984; Huang et al., 1985) suggest now that these morphologically similar viruses

6 980 C. R. PRINGLE AND R. P. EGLIN Table 5. Comparison of the properties and functions of the polypeptides of PVM and RS virus RS virus polypeptides* A PVM polypeptidest Designation Mol. wt. Mol. wt. ( X 10-3) Function Designation ( X 10-3) Function L G Polymerase Attachment I II Core protein Glycoprotein (without glycosylation, 36K mol. wt.) F0 68 Fusion (cleaved III 62 Glycoprotein to F1/48K and F2/22K) N 42 Nucleoprotein IV 42 Nucleoprotein P (ex 42) Core protein, V VI 40 36? Core protein phosphorylated F2 22 Cleavage product IX 17? M1 M Matrix (major) Matrix (minor) VII VIII 28 26?? of F0 NS 14 Non-structural X 16 Non-structural NS NS Non-structural Non-structural XI XII 14 12? Non-structural * Data from Collins et al. (1984), Huang et al. (1985) and Pringle et al. (1981). t Data from Cash et al. (1979). (Berthiaume et al., 1974) are more closely related than previously supposed. Table 5 illustrates that the 12 PVM-specified polypeptides can be correlated with the 12 primary or secondary products of the RS virus genome. Furthermore, the results presented here indicate that PVM, or an antigenically related virus, is able to infect man. Exposure to infection appears to be frequent and widespread, since moderate or high levels of anti-pvm activity were present in the majority of serum samples from differently constituted groups in the west of Scotland and in Oxfordshire. A similar incidence of high PVM-neutralizing titres has been found in the sera of blood donors in Nigeria (C. U. Iroegbu & C. R. Pringle, unpublished data) suggesting that PVM like RS virus is ubiquitous. As in the case of RS virus, first exposure to virus appears to occur early in life with the majority of children possessing moderate or high levels of PVM-neutralizing activity by their 3rd birthday. However, unlike RS virus where antibody is universal in adults, a small proportion (8~) of individuals in the 15 and over age group had low or no anti-pvm activity in their sera. These results confirm the original observations of Horsfall & Hahn (1940) and Horsfall & Curnen (1946) who found evidence of antibody in man as well as in mice and a variety of other animals. Thirty-three ~o and 27 ~o of human sera respectively were found to be positive in these two surveys. Due to the sporadic occurrence of PVM in mouse colonies it was suggested that the virus might have originated in mice from human contact. The absence of any record of isolation of PVM from man during the intervening 40 years since its first isolation is surprising in view of the seroepidemiological data presented here which suggest its apparent ubiquity at the present time. However, the slow growth of the virus in cultured cells and its instability may partly explain this. It is possible also that the presence of antibodies in the animal sera used in culture media has prevented isolation of this virus. The evidence for association with human disease is not strong; the four instances of seroconversion among 108 respiratory infections of undetermined aetiology could have been fortuitous. However, it is possible that PVM does contribute to the load of human respiratory disease. Seroconversions in two renal transplant patients have also been observed in addition to the results reported here, perhaps emphasizing the prevalence of the virus rather than its disease-producing potential. Both of these patients were experiencing a respiratory tract infection during the sampling period; however, a simultaneous rise in RS virus-neutralizing activity was also observed in one patient during the same period. A

7 PVM-neutralizing activity in human sera 981 The apparent association of PVM with Paget's disease of bone is difficult to evaluate, part!cularly as the bimodal distribution observed at the start of the investigation (Fig. 2) was not typical of the wider population. The patients with Paget's disease of bone were predominantly elderly and female reflecting the normal sex distribution late in life. Age- and sex-matching within this group, however, did not affect the result. The number of cases studied is not sufficient to examine whether this introduced any bias to the result obtained. The matter is under further investigation and attempts are also being made to detect PVM antigens in Pagetic lesions and normal bone tissue. Evidence associating the other pneumovirus, RS virus, with Paget's disease of bone (Mills et al., 1981, 1982, 1984; Singer & Mills, 1983) has yet to be confirmed independently. Presumptive epidemiological evidence for an association between canine distemper virus and Paget's disease of bone has also been described recently (O'Driscoll & Anderson, 1985). Despite the reports of the sporadic occurrence of other paramyxovirus (measles virus, PI3 virus, SV5) antigens in Pagetic lesions (Rebel et al., 1977, 1980; Basl~ et al., 1979, 1985), the role of paramyxoviruses in Paget's disease of bone remains obscure. REFERENCES BASLE, M. F., REBEL, A., POUPLARD, A., KOUYOUMDJIAN, S., FILMON, R. & LEPATEZOUR, A. (1979). Mise en 6vidence d'antig6nes viraux de rougeole dans les ost6oclastes de la maladie osseuse de Paget. Comptes rendus hebdomadaires des s~ances de l'acad~mie des sciences, s~rie D 289, BASLf], M. F., RUSSELL, w. C., GOSWAMI, K. K. A., REBEL, A., GIRAUDON, P., WILD, F. & FILMON, R. (1985). Paramyxovirus antigens in osteoclasts from Paget's bone tissue detected by monoclonal antibodies. Journal of General Virology 66, BASLIB, M. F., FOURNIER, J. G., ROZENBLATT, S., REBEL, A. & BOUTEILLE, M. (1986). Measles virus RNA detected in Paget's disease bone tissue by in situ hybridization. Journal of General Virology 67, BERTHIAUME, L., JONCAS, J. & PARILANIS, V. (1974). Comparative structure, morphogenesis and biological characteristics of the respiratory syncytial (RS) virus and the pneumonia virus of mice (PVM). ArchivJ~r die gesamte Virusforschung 45, CASH, P., PRESTON, C. M. & PRINGLE, C. R. (1979). Characterisation of murine pneumonia virus proteins. Virology 96, COLLINS, P. L., HUANG, Y. T. & WERTZ, G. W. (1984). Identification of a tenth mrna of respiratory syncytial virus and assignment of polypeptides to the 10 viral genes. Journal of Virology 49, EATON, M. D. & VAN HERICK, W. (1944). Demonstration in cotton rats and rabbits of latent virus related to pneumonia virus of mice. Proceedings of the Society for Experimental Biology and Medicine 57, GANNON, J. & CARTHEW, P. (1980). Prevalence of indigenous viruses in laboratory animal colonies in the UK Laboratory Animals 14, HORSFALL, F. L. & CURNEN, E. C. (1946). Studies on pneumonia virus of mice (PVM). II. Immunological evidence of latent infection with the virus in numerous mammalian species. Journal of Experimental Medicine 83, HORSFALL, F. L. & HAHN, R. G. (1939). A pneumonia virus of Swiss mice. Proceedings of the Society for Experimental Biology and Medicine 40, HORSFALL, F. L. & HAHN, R. G. (1940). A latent virus in normal mice capable of producing pneumonia in its natural host. Journal of Experimental Medicine 71, HORSFALL, F. L., CURNEN, E. C., MIRICK, G. S., THOMAS, L. & ZIEGLER, J. E. (1943). m virus recovered from patients with primary atypical pneumonia. Science 97, HOWATSON, A. F. & FOURNIER, V. L. (1982). Microfilaments associated with Paget's disease of bone. Comparison with nucleocapsids of measles virus and respiratory syncytial virus. Intervirology 18, HUANG, Y. T., COLLINS, P. L. & WERTZ, G. W. (1985). Characterisation of the 10 proteins of human respiratory syncytial virus: identification of a fourth envelope-associated protein. Virus Research 2, LUNDGREN, D. L., MAGNUSON, M. G. & CLAPPER, W. E. (1969). A serological survey in dogs for antibody to human respiratory viruses. Laboratory Animal Care 19, MILLS, B. G., SINGER, F. R., WERNER, L. P. & HOLST, P. A. (1981). Immunohistological demonstration of respiratory syncytial virus antigens in Paget's disease of bone. Proceedings of the National Academy of Sciences, U.S.A. 78, MILLS, B. G., STABILE, E., HOLST, P. A. & GRAHAM, C. (1982). Antigens of two different viruses in Paget's disease of bone. Journal of Dental Research 61, MILLS, B. G., SINGER, F. R., WERNER, L. P., SUFFEN, S. C., STABILE, E & HOLST, P. (1984). Evidence for both respiratory syncytial virus and measles virus antigens in the osteoclasts of patients with Paget's disease of bone. Clinical Orthopaedics and Related Research 183, O'DRISCOLL, J. B. & ANDERSON, D. C. (1985). Past pets and Paget's disease. Lancet i, PARKER, J. C., TENNANT, R. W. & WARD, T. G. (1966). Prevalence of viruses in mouse colonies. National Cancer Institute Monographs 20, PRINGLE, C. R. & CROSS, A. (1978). Neutralisation of respiratory syncytial virus by cat serum. Nature, London 276,

8 982 C. R. PRINGLE AND R. P. EGLIN PRINGLE, C. R., SHIRODARIA, P. V., GIMENEZ, H. B. & LEVINE, S. (1981). Antigen and polypeptide synthesis by temperature-sensitive mutants of respiratory syncytial virus. Journal of General Virology 54, PRINGLE, C. R., WILK1E, M. L. & ELLIOTT, R. M. (1985). A survey of respiratory syncytial virus and parainfluenza virus type 3 neutralising and precipitating antibodies in relation to Paget's disease. Journal of Medical Virology 17, REBEL, A., MALKANI, g., BASLI~, M. & BREGEON, CH. (1977). IS Paget's disease of bone a viral infection? Calcified Tissue Research 22 supplement, REBEL, A., BASLE, M. F., POUPLARD, A., KOUYOUMDJIAN, S., FILMON, R. & LEPATEZOUR, A. (1980). Viral antigens in osteoclasts from Paget's disease of bone. Lancet i, 34~346. RICHARDSON-WYATT, L. S., BELSHE, R. B., LONDON, W. T., SLY, D. L., CAMARGO, E. & CHANOCK, R. M. (1981). Respiratory syncytial virus antibodies in nonhuman primates and domestic animals. Laboratory Animal Science 31, SINGER, F. R. (1980). Paget's disease of bone: a slow virus infection? Calcified Tissue International 31, SINGER, F. R. & MILLS, B. G. (1977). The aetiology of Paget's disease of bone. Clinical Orthopaedics and Related Research 127, SINGER, V. R. & MILLS, B. G. (1983). Evidence for a viral etiology of Paget's disease of bone. Clinical Orthopaedics and Related Research 178, (Received 25 November 1985)