Comparison of Proteins Induced in Cells Infected with Rinderpest and Peste des Petits Ruminants Viruses

Transcription

1 J. gen. Virol. (1987), 68, Printed in Great Britain 2033 Key words: rinderpest/peste des petits ruminants/morbillivirus proteins Comparison of Proteins Induced in Cells Infected with Rinderpest and Peste des Petits Ruminants Viruses By A. DIALLO, 1"2 T. BARRETT, l* P.-C. LEFEVRE 2 AND W. P. TAYLOR 3 1Institute for Animal Disease Research, Pirbright Laboratory, Ash Road, Pirbright, Woking GU24 ONF, U.K., 2Institut d'elevage et de Mbdicine Vbtbrinaire des Pays Tropicaux, 10 rue Pierre-Curie, Maisons-Alfort, Paris, France and 3Food and Agricultural Organization, P.O. Box 30570, Nairobi, Kenya (Accepted 2 April 1987) SUMMARY The two morbilliviruses rinderpest virus (RPV) and peste des petits ruminants virus (PPRV) are closely related and cause severe disease in large and small ruminants, respectively. They show distinct epidemiological patterns and are distinguishable by reciprocal cross-neutralization tests. We have analysed the proteins induced by these viruses in infected cells and have shown that they can be distinguished by a very marked difference in the apparent mol. wt. of the nucleocapsid (N) protein. The N protein of PPRV is almost identical in mobility on polyacrylamide gels to the N proteins of measles virus and canine distemper virus (60K). Several strains of RPV and PPRV from widespread geographical locations were studied and found to show this difference in the N protein. Rinderpest virus (RPV) causes a severe, acute disease of cattle and wild bovids in Africa, Asia and the Middle East. Peste des petits ruminants virus (PPRV) causes a disease of sheep and goats which is clinically similar to RPV. It was first described in West Africa (Gargadennec & Lalanne, 1942) where it is still widespread and considered to be the major constraint to increased small ruminant production. It occurs in the Sudan where its impact on livestock is unknown (El Hag Ali & Taylor, 1984; for review, see Taylor, 1984). Clinically the two diseases are characterized by pyrexia, nasal catarrh and ocular discharges, enteritis and severe oral necrosis. The morbidity in susceptible herds is often greater than 90 ~ and the mortality may be very high (60 to 90~). The causative agents are two distinct, though closely related, viruses in the morbillivirus genus (family Paramyxoviridae) which includes measles virus (MV) and canine distemper virus (CDV). All four viruses are closely related serologically (Imagawa, 1968; 0rvell & Norrby, 1974; Gibbs et al., 1979; Hall et al., 1980). Little is known about the genome or proteins of RPV and PPRV, but recently it has been shown that the mrnas induced in RPV and PPRV infections of African green monkey kidney (Vero) cells are identical to those of MV (Barrett & Underwood, 1985). It is difficult to study the proteins of the morbilliviruses, particularly RPV and PPRV proteins, since it is not possible to prepare purified virus, free of host cell contaminants. In addition, these viruses do not shut off host cell protein synthesis efficiently and the virus-specific proteins induced in infected cells are very susceptible to protease degradation. In this study we have tried to overcome these difficulties by labelling the proteins/n vivo in the presence of high concentrations of actinomycin D (AMD), to inhibit host cell mrna synthesis, and have lysed the labelled cells by adding SDS- PAGE loading buffer directly to the cell monolayer and boiling the samples immediately for 3 min. The following viruses were used in this study. Rinderpest virus, RBOK vaccine strain, KAG vaccine strain and wild field isolates from Tanzania (RBT-1), Kuwait (1982), Nigeria (1984), SGM

2 2034 (a) P/H[-- NE Short communication "5 Im All Ill ~Qm U -69 (b) Actin -- ~ ~ ~ ~ --46 N --30 Actin 7... Fig. 1. Analysis of RPV and PPRV proteins synthesized in vivo. All samples in (a) are RPV strains isolated from different geographical areas. Lane 1, uninfected Vero cells; lane 2, RBOK, vaccine; lane 3, KAG vaccine; lane 4, Tanzania (RBT-1): Jane 5, Nigeria (1984); lane 6, Kuwait (1982); lane 7, Oman (1979); lane 8, Saudi Arabia (1981); lane 9, Iraq (1985); lane 10, Yemen (1981); lane 11, Egypt (1984); lane 12, protein tool. wt. markers (Amersham; 10-3). (b) Lane 1, uninfected Vero cells; lane 2, PPRV (Nigeria 75/1); lane 3, Tanzania (RBT-1); lane 4, RPV (Kuwait). Monolayers of Vero cells in 96- well plates were infected with viruses at an mo.i. of 0.5 to 1.0 and incubated at 37 C. At approximately 50~o c.p.e., they were starved for 3 to 6 h in methionine-free medium containing 10 ~tg/ml AMD, washed and labelled for 3 h in medium containing [ 35 S]methionine (Amersham) at 50 ~tci/ml. At the end of the labelling period the medium was removed, 40 ~tl of gel loading buffer was added to each well and the whole plate was boiled for 3 min. Labelled samples were then stored at - 20 C until used. Proteins were separated on 15~ discontinuous gels (a) or 12.5~ /~ gels (b) (Laemmli, 1970). Oman (1979), Saudi Arabia (1981), Iraq (1985), Yemen (1981) and Egypt (1984); PPRV, Nigeria (75/1, 75/2, 75/3, 76/1 strains), Sudan (Meiliq and Sennar strains) and Ghana (Accra strain); MV, Edmonston strain and CDV, Onderstepoort strain. All the viruses were adapted to grow on Vero cells. Virus stocks were harvested when the cells showed 80~ c.p.e. The cells were frozen and thawed once with the culture medium, cellular debris was removed by centrifugation at 1000 g for 10 to 15 min and the supernatants were harvested and stored at -70 C. The mol. wt. of all the proteins induced in MV- and CDV-infected cells are well established in the literature, although there are slight variations depending on the strain of virus used (Rima, 1983). Fig. 1 shows the proteins induced in Vero cells infected with PPRV and various strains of RPV. As in the cases of MV and CDV, the nucleocapsid (N) protein was the most abundant and easily identifiable protein induced in RPV and PPRV infections. However, the N protein of RPV was significantly larger (66K to 68K) than the corresponding protein in PPRV (60K), MV (59K) or CDV (60K). The mol. wt. of MV and CDV N proteins are those determined by translation in vitro (Fig. 2a). Even though the N protein of RPV varied with each strain of virus, all were larger than the N protein of PPRV. Only in one case was the RPV N protein slightly smaller than the average size of the RPV N proteins (Tanzania RBT-I) but it was still significantly larger than the N protein of PPRV (see Fig. 1 a, lane 4; and compare lanes 2, 3 and 4 in Fig. 1 b). This is an example of a series of mild field strains isolated from game in Tanganyika (now Tanzania) in the late 1950s (Robson et al., 1959). Only one strain of PPRV is

3 Short communication 2035 (a) (b) (c) p..... p ~p... _/H? ~ ~lml)--n ~ - - N --]N * ~* -XF? ~. i'~ ~M Fig. 2. (a) In vitro translation products of purified mrnas from each of the morbilliviruses. Lane 1, Mol. wt. marker proteins; lane 2, RPV (Kuwait) mrna; lane 3, PPRV (Nigeria 1/75) mrna; lane 4, CDV mrna; lane 5, MV mrna. (b) Immunoprecipitation of RPV (RBOK) mrna in vitro translation products by rabbit anti-rpv hyperimmune serum (lane 1), rabbit anti-pprv hyperimmune seurm (lane 2) and convalescent serum from an infected calf (lane 3). (c) Comparison of RPV (RBOK) and RPV (Kuwait) in vitro translation products. Lane 1, Mol. wt. protein markers; lane 2, RPV (RBOK); lane 3, RPV (Kuwait). For extraction of RNA viruses were grown in 20 oz bottles and harvested at 80 to 90~ c.p.e. AMD (10 ~g/ml) was added approx. 12 h prior to harvesting (at 50 to 60% c.p.e.) Poly(A) RNA was purified by oligo(dt)-cellulose chromatography and translated in a rabbit reticulocyte lysate (Amersham). The proteins were synthesized in the presence of [3SS]methionine and separated on 12.5 ~ gels. The positions of the virus-specific proteins are marked on the side of the gel. The expected positions of the unglycosylated H and Fo proteins are also indicated. ~own but the N protein was identical in all the strains listed above. In the case of some RPV rains e.g. Kuwait strain, the N protein appeared as two strong bands migrating close to each her. The slower migrating N band was shown to be a phosphorylated form of the N protein. In te strains of the virus showing only one form of the N protein, e.g. RBOK strain, it was not gnificantly phosphorylated (data not shown). Translations of mrna in vitro from both these ruses showed only one N protein band of identical mol. wt. (see below and Fig. 2c). Since it is difficult to identify the other virus-induced proteins in some infections in vivo due to e small amounts produced and the host cell background proteins, translations were carried out vitro using mrna extracted from cells infected with all four morbilliviruses (Fig. 2a). In each tse the major protein synthesized in vitro was identical in mol. wt. to the N protein synthesized vivo by the corresponding virus, Hyperimmune serum raised in rabbits against RPV and PRV immunoprecipitated the polymerase-associated (P), N and matrix (M) proteins of RPV. n the other hand, convalescent serum immunoprecipitated mainly the P and N proteins (Fig. 2). The translations in vitro confirmed the larger apparent mol. wt. of the N proteins of RPV. In tdition, the doublet form of the N protein was not synthesized in translations in vitro of mrna om the Kuwait strain of RPV, again indicating that the upper band represented a postanslational processing event. As expected, only one protein band corresponding to the N -otein was synthesized with the RBOK strain of RPV. (Fig. 2c). Proteins were synthesized which corresponded to the P proteins of each virus. These were entitled as P proteins since they immunoprecipitated with polyclonal antisera, with a P- )ecific monoclonal antibody (MAb) made against MV (kindly provided by Dr P. Giraudon, yons, France) and with antisera raised against a peptide corresponding to a highly conserved gion (amino acid residues 495 to 504) of the MV and CDV P proteins (Fig. 3). It can be seen om Fig. 3 that the P protein-specific MAb also precipitated significant amounts of the N otein. This may be because the MAb recognizes a form of the P protein complexed with the N

4 2036 Short communication Fig. 3. Immunoprecipitation of the P proteins with P-specific antisera. Lane 1, Mol. wt. markers; lane 2, RPV (Kuwait) proteins with P (MV) monoclonal antiserum; lane 3, RPV (Kuwait) proteins with P peptide antiserum; lane 4, mol. wt. markers, lane 5, MV proteins with P peptide antiserum; lane 6, MV proteins with P (MV) monoclonal antiserum. A peptide corresponding to a highly conserved region of the MV and CDV P protein was synthesized using a Beckman 990 peptide synthesizer. Peptide-specific antiserum was induced in rabbits after conjugation of the peptide with bovine serum albumin. The P- peptide-specific antiserum and a P (MV)-specific MAb (gift of Dr. P. Giraudon) were used to immunoprecipitate the translation products in vitro of RPV (Kuwait) and MV mrnas. The resulting precipitates were separated on 12.5% gels. protein. This type of complex has been described in vesicular stomatitis virus (VSV) where the N and NS (equivalent to P) proteins co-precipitate (Davis et al., 1986). It is probable that a conformational change occurs in the P protein when it is complexed with N. It is difficult to identify unequivocally the P protein in infected cell lysates labelled in vivo because this protein is produced in much lower amounts than the N protein and background host proteins can be confusing. Proteolytic cleavage of virus proteins is also a problem when immune precipitations are carried out on this material no matter how carefully the experiments are performed. During translation in vitro, proteolysis is not a problem and the true size of the proteins can be determined. In addition, post-translational phosphorylation of the P protein in vivo may significantly alter its mobility on polyacrylamide gels. The M protein of RPV also showed strain variation. This was confirmed by translation in vitro. The M _protein of RPV (Kuwait) was larger than the M protein of RPV (RBOK) both in vivo (Fig. 1, lanes 2 and 6), and in vitro (Fig. 2c). Rinderpest and PPR epidemics are common in the same areas of the world. At first it was suggested that they were caused by variants of the same virus infecting large and small ruminants, respectively. Later serological studies showed that there are two distinct viruses which are very closely related. There have been no biochemical studies, however, to support and confirm these results. To date there have been only a few studies carried out on the proteins of RPV (Underwood & Brown, 1974; Prakash et al., 1979; Krishnaswamy & Veerabhadrappa, 1982; Sato et al., 1981) and none on PPRV. All of these studies gave a mol, wt. of about 60K for the N protein of RPV, significantly less than we have observed both in vivo and in vitro. It is probable that in all cases, proteolysis of the protein was responsible for the low mol. wt. estimations. A comparative study of the mrnas induced in the different morbillivirus infections showed that both RPV and PPRV induced identical mrnas to MV in infected Vero cells (Barrett &

5 Short communication 2037 Underwood, 1985). We have analysed many strains of these viruses from widespread geographical locations and consistently found RPV to have a larger N protein, as judged by PAGE. The larger size of the N protein of RPV is not reflected in an appreciably larger mrna for this protein and it is probable that the protein migrates anomalously on PAGE, although this is not the case for MV and CDV where the N proteins migrate at a position expected from their predicted tool. wt. deduced from the gene sequence (Rozenblatt et al., 1985). The P proteins of RPV (84K) and PPRV (86K) also showed a slightly greater mol. wt. than the corresponding proteins in MV (76K) and CDV (80K). However, this protein migrates anomalously on PAGE since the actual mol. wt. of the P gene product deduced from the nucleotide sequence of the gene (MV, 53.9K; CDV, 54.9K) is much less than the apparent mol. wt. on PAGE for both MV and CDV (Bellini et al., 1985; Barrett et al., 1985) and Sendal virus (Giorgi et al., 1983). The mol. wt. of these proteins may vary further when they are phosphorylated in vivo. Since these proteins produced m vitro could be, immunoprecipitated with polyclonal antisera raised against both RPV and PPRV, with a P-specific MAb and with antiserum raised against a P gene peptide, we are confident that they represent the true unmodified form of the P proteins. Since the N protein is the most abundant virus-induced protein species in infected cells and can be readily detected even in those infected at a low level, this mobility difference may be of use in rapid differentiation of RPV and PPRV isolates. Complete sequence data on the N genes of RPV and PPRV is required to determine the reason for the slower migration of RPV N proteins. This work is now in progress. We wish to thank Miss M. Barbron for excellent technical assistance, Dr P. Giraudon for providing the measles anti-p MAb, Dr T. Doel for synthesis of the P-specific peptide, Mrs D. O'Keefe for photographic assistance and Dr B. W. J Mahy for encouragement and critical reading of the manuscript. This work was supported by grants from the Wellcome Trust (No /1.5) and the EEC (No ). REFERENCES BARRETT, T. & UNDERWOOD, B. (1985). Comparison of messenger RNA induced in cells infected with each member of the morbillivirus group. Virology 145, BARRETT, T., SHRIMPTON, S. & RUSSELL, H. (1985). N ucleotide sequence of the entire protein coding region of canine distemper virus polymerase-associated (P) protein mrna. Virus Research 3, BELLINI, W. J., ENGLUND, G., ROZENBLATT, S., ARNHEITER, H. & RICHARDSON, C. D. (1985). Measles virus P gene codes for two proteins. Journal of Virology 53, DAVIS, N. L., ARNHEITER, H. & WERTZ, G. W. (1986). Vesicular stomatitis virus N and NS proteins form multiple complexes. Journal of Virology 59, EL HAG ALI, B. & TAYLOR, W. P. (1984). Isolations of peste des petits ruminants virus from the Sudan. Research in Veterinary Science 36, 1~4. GARGADENNEC, L. & LALANNE, A. (1942). La peste des petits ruminants. Bulletin des Services Zootechniques et des Epizooties de l'ajrique Occidentale Franfaise 5, GIBBS, E. J. P., TAYLOR, W. P., LAWMAN, M. I. P. & BRYANT, I. (1979). Classification of peste des petits ruminants virus as the fourth member of the genus morbillivirus. Intervirology 11, GIORGI, C., BLUMBERG, B. M. & KOLAKOFSKY, D. (1983). Sendal virus contains overlapping genes expressed from a single mrna. Cell 35, HALL, W. W., LAMB, R. A. & CHOPPIN, P. W. (1980). The polypeptides of canine distemper virus: synthesis in infected cells and relatedness to the polypeptides of other morbilliviruses. Virology 100, IMAGAWA, D. T. (1968). Relationships among measles, canine distemper and rinderpest viruses. Progress in Medical Virology 10, KRISHNASWAMY, S. & VEERABHADRAPPA, G. (1982). Intracellular polypeptides in the rinderpest virus infected cells. Indian Journal of Comparative Microbiology, Immunology and Infectious Diseases 3, LAEMMLI, V. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, London 227, RVELL, C. & NORRBY, E. (1974). Further studies on the immunological relationships among measles, canine distemper and rinderpest virus. Journal of Immunology 113, PRAKASH, K., ANTHONY, A. & RAMAKRISHNAN, T. (1979). Polypeptides of rinderpest virus and virus-specific polypeptide synthesis in infected cells. Indian Journal of Experimental Biology 17, RIMA, B. (1983). The proteins of morbilliviruses. Journal of General Virology 64, ROBSON, J., ARNOLD, R. M., PLOWRIGHT, W. & SCOTT, G. R. (1959). The isolation from an eland of a strain of rinderpest virus attenuated for cattle. Bulletin of Epizootic Diseases of Africa 7, ROZENBLA'I~I ', S., EIZENBERG, O., BEN-LEVY, R., LAVIE, V. & BELLINI, W. G. (1985). Sequence homology within morbilliviruses. Journal of Virology 53,

6 2038 Short communication SATO, T. A., HAYAM, M. & YAMANOUCHI, R. (1981). Analysis of structural proteins of measles, canine distemper and rinderpest viruses. Japanese Journal of Medical Sciences and Biology 34, TAYLOR, W. P. (1984). The distribution and epidemiology of peste des petits ruminants. Preventive Veterinary Medicine 2, UNDERWOOD, B. O. & BROWN, F. (1974). Physico-chemical characterisation of rinderpest virus. Medical Microbiology and Immunology 160, (Received 4 February 1987)