Veterinary Research Communications, 28 (2004) 623^639 # 2004 Kluwer Academic Publishers. Printed in the Netherlands
Production and Characterization of Monoclonal Antibodies to Peste des Petits Ruminants (PPR) Virus R.P. Singh1*, S.K. Bandyopadhyay2, B.P. Sreenivasa3 and P. Dhar2 1 Division of Virology, Indian Veterinary Research Institute, Mukteswar, Nainital ^ 263 138 (Uttaranchal ), India; 2Indian Veterinary Research Institute, Izatnagar, Bareilly, India; 3Indian Veterinary Research Institute, Hebbal, Bangalore, India *Correspondence: E-mail:
[email protected] Singh, R.P., Bandyopadhyay, S.K., Sreenivasa, B.P. and Dhar, P., 2004. Production and characterization of monoclonal antibodies to peste des petits ruminants (PPR) virus. Veterinary Research Communications, 28(7), 623^639 ABSTRACT Peste des petits ruminants (PPR) is an acute, febrile viral disease of small ruminants, caused by a virus of the genus Morbillivirus. PPR and rinderpest viruses are antigenically related and need to be di¡erentiated serologically. In the present study, 23 mouse monoclonal antibodies were produced by polyethyleneglycol (PEG)-mediated fusion of sensitized lymphocytes and myeloma cells. Among these, two belong to the IgM class and the remaining 21 to various subclasses of IgG. The MAbs from the IgG class designated 4B6 and 4B11 neutralized PPR virus in vitro. In radioimmunoprecipitation assay, 10 MAbs recognized nucleoprotein, 4 recognized the matrix protein and one each haemagglutinin and phosphoprotein. The remaining 7 MAbs failed to precipitate any de¢ned viral protein. The reactivity pattern of the monoclonal antibodies in indirect ELISA indicated a close antigenic relationship within three Indian PPR (lineage 4) virus isolates and also within two rinderpest vaccine strains. All PPR virus isolates could be distinguished from rinderpest vaccine viruses on the basis of the reactivity pattern of all MAbs and anti-N protein MAbs. A set of six monoclonal antibodies speci¢c to PPR virus could also be identi¢ed from the panel. From the panel of MAbs available, two MAbs were selected for diagnostic applications, one each for the detection of antigens and antibodies to PPR virus. Keywords: antigen, monoclonal antibody characterization, peste des petits ruminants virus, rinderpest virus Abbreviations: CPE, cytopathic e¡ect; ELISA, enzyme-linked immunosorbent assay; HRPO, horseradish peroxidase; kDa, kilodalton; MAb, monoclonal antibody; MEM, minimum essential medium; OPD, ortho-phenylene diamine; PEG, polyethyleneglycol; PMSF, phenylmethylsulphonyl £uoride; PPR, peste des petits ruminants; RIPA, radioimmunoprecipitation assay; SDS-PAGE, sodium dodecyl sulphate^polyacrylamide gel electrophoresis; TCID, tissue culture infective dose
INTRODUCTION Peste des petits ruminants (PPR) is an acute, febrile viral disease of small ruminants, characterized by necrotizing and erosive stomatitis, enteritis and pneumonia (Ismail et al., 1995). The disease is caused by an RNA virus (PPR virus) belonging to the genus Morbillivirus of the family Paramyxoviridae. The virus is antigenically related to other morbilliviruses (Gibbs et al., 1979), e.g. rinderpest virus (RPV), canine distemper virus (CDV) and measles virus (MV). Among the morbillivirus group, PPR and rinderpest 623
624
viruses have been found to cause similar diseases in small ruminants (Diallo et al., 1989). The two viruses show very close antigenic and genetic relationship. The genome of these viruses is divided into six transcriptional units encoding two non-structural proteins (V, C) and six structural proteins. The surface glycoproteins include fusion (F) and haemagglutinin (H) proteins. Internal virion proteins include the matrix (M) protein, the nucleoprotein (N), and the phosphoprotein (P), which forms the polymerase complex in association with the large (L) protein (Crowley et al., 1988; Sidhu et al., 1993; Diallo et al., 1994). Morbilliviruses di¡er considerably in epitope homology of H proteins, while epitopes in F protein are relatively conserved. The internal virion proteins such as N and M have been shown to have both speci¢c and cross-reacting epitopes (McCullough et al., 1986; Sheshberadaran et al., 1986). Therefore, anti-N protein monoclonal antibodies to rinderpest virus alone were used to establish relationships between di¡erent morbilliviruses. The nucleocapsid gene and the protein have been the focus of attention for di¡erentiation of PPR and rinderpest viruses by various workers (McCullough et al., 1986; Diallo et al., 1987, 1989). The main objective of the investigation of the properties of monoclonal antibodies (MAbs) to PPR virus was to select and identify these for antigenic characterization, and detection of antigen and antibodies to PPR virus using suitable ELISA systems. Recent epidemiology of PPR viruses in India and more broadly in Asia, suggests that all the viruses belong to lineage 4 (Dhar et al., 2002) with the exception of `India/TN/ 92', a lineage 3 virus (Shaila et al., 1996). In a changing epidemiological scenario, a well-characterized panel of monoclonal antibodies could be an important tool for morbillivirus research. Further, the use of monoclonal antibodies to PPR viruses, though reported scantily, will lead to better understanding of antigenic relatedness between various members of the morbillivirus genus. In the present study, we report (i) the production of 23 MAbs from a single fusion using a crude antigen derived from an Indian isolate of PPR virus; (ii) the characterization of MAbs for class and subclass, virus neutralization ability, antigenic speci¢city using radioimmunoprecipitation assay and reactivity with various PPR and rinderpest virus isolates in ELISA; and (iii) the subsequent use of MAbs to determine antigenic relationships between established rinderpest viruses and PPR virus isolates. MATERIALS AND METHODS Cell lines Vero cells, B95a cells and myeloma cells used in the present study were available in the authors' laboratory. Vero cells between passages 130^160, propagated in Earle's MEM with 10% fetal calf sera, were used to prepare the antigen for immunization of mice. B95a cells are lymphoblastoid cells from the marmoset monkey (Kobune et al., 1991). These cells were propagated in RPMI medium supplemented with 10% fetal calf serum and maintained in the same medium with 2% fetal calf serum. B95a and Vero cells were used for production of PPR virus antigen for screening of primary hybridoma, characterization of monoclonal antibodies and virus neutralization test. Myeloma cells
625
(sp2/0) were propagated in Iscove's modi¢ed Dulbecco's medium (IMDM) having 20% hybridoma-tested fetal calf serum (Sigma-Aldrich, St Louis, MO, USA) under 5% carbon dioxide. These cells were used as fusion partners with sensitized splenocytes. Reference viruses PPR virus isolates `PPRV Sungri-96', Vero/57-62 (Sreenivasa et al., 2000), `PPRV Chirgaon-99' Vero/32 (unpublished information) and `PPRV Izatnagar-94' B95a/5 and Vero/9 (Sreenivasa et al., 2002) were used in the study. Rinderpest virus isolates included attenuated `RBOK strain' (Plowright and Ferris, 1962), `Edwards Caprinized' rinderpest virus adapted to grow in Vero cells (Parida and Bandyopadhyay, 1996) and `Lapinized' rinderpest virus (Nakamura et al., 1938). Experimental animals Balb/c mice aged 4^6 weeks (weighing 16^20 g) were used for immunization with PPR virus antigen for the production of hybridomas. Production of monoclonal antibodies Monoclonal antibodies were produced according to standard procedures (Harlow and Lane, 1988; Peters and Baumgarten, 1992) with some modi¢cations. The modi¢cations included use of crude antigen for immunization of mice and feeder cells (peritoneal macrophages) for hybridoma production. PPR antigen used for immunization was prepared as described by Singh and colleagues (2000) with certain modi¢cations. Vero cell-derived, cell-free virus was prepared using a low multiplicity (0.01 TCID50/cell) of PPR virus. It was subjected to polyethyleneglycol (PEG) precipitation, followed by ultracentrifugation at 100 000g for 2 h. Two Balb/c mice were immunized using crude PPR virus antigen. The immunizing dose was adjusted at a virus concentration equivalent to 26105 infected Vero cells (Sugiyama et al., 1989). Three intraperitoneal inoculations were made: the ¢rst (day 0) with Freund's complete adjuvant, the second (day 35) and third (day 42) with Freund's incomplete adjuvant. Subsequently, three immunizations were carried out intravenously without adjuvant on three consecutive days (days 49, 50 and 51) prior to fusion (day 52). However, before fusion, the mice were test bled on the 45th day of the immunization schedule to detect immune response against the inoculated antigen. Myeloma cells and spleen lymphocytes were prepared for fusion as described by Singh (2002). Polyethyleneglycol (50% PEG, MW 1300^1600)-mediated fusion of myeloma and lymphocytes (cell ratio 1:10) was performed. Cells were seeded in 96well microtitre plates. Growth of hybridoma was observed regularly under a selection medium containing aminopterin. Well-to-well screening of hybridoma clones, employing an indirect ELISA was performed at the appropriate stage of growth (25% of the well surface covered by cells). Positive clones were ampli¢ed, followed by single-cell
626
cloning and subcloning of PPR-positive hybridoma using the limiting-dilution method (Harlow and Lane, 1988; Peters and Baumgarten, 1992). Single-cell cloned hybrids were screened at the appropriate stage of the growth. Single-cell cloning was repeated at least twice so as to prove the monoclonality of the hybridoma culture supernatants. Characterization of monoclonal antibodies The isotypes of the MAbs were determined by testing the hybridoma cell culture £uid using a commercial ELISA kit (Sigma-Aldrich). Capture ELISA was performed as recommended by the manufacturers of the kit. Although most of the culture supernatants were used undiluted, monoclonal antibodies such as 4A3, 4D11, 4G2, 4H9, 4H10, 4I3 and 4F13 were diluted 1:100 for determination of their isotypes as these MAbs reacted with all the isotypes in undiluted form. The virus neutralization ability of the speci¢c MAbs was determined according to the procedure described earlier (Singh et al., 2000; Bandyopadhyay et al., 1999) with certain modi¢cations. For this purpose, 2-fold dilutions of the monoclonal antibodies were incubated with 100 TCID50/well of PPR virus (PPRV-Sungri 96) overnight at 48C. Infectivity of virus was tested using a 48-h-old monolayer of B95a cells. The cells were observed daily for virus-speci¢c cytopathic e¡ect (CPE) and ¢nal reading were made 5 days post infection. The protein speci¢city of MAbs was studied by radioimmunoprecipitation assay (RIPA). Labelling of the viral antigen was done according to the procedure described by Sugiyama and colleagues (1989) with certain modi¢cations (Singh, 2002). Modi¢cations included the use of Vero cells in place of NA cells, the labelling of viral proteins for 6 h instead of 2 h and the use of a modi¢ed RIPA bu¡er (0.15 mol/L NaCl, 5 mmol/L EDTA, 0.6 mol/L KCl, 0.01 mol/L Tris, 2% Triton X-100, 3 mmol/L PMSF, 2.5 mmol/L iodoacetamide, and protease inhibitor cocktail from Sigma). Immuunoprecipitation of viral antigens was carried out according to the procedure described by Libeau and Lefevre (1990) using protein A sepharose CL-4B fast £ow (100 ml of 50% beads) with 500 ml of hybridoma culture £uid and 100 ml of infected Vero cell lysate. PPR convalescent serum was also used to precipitate radiolabelled viral proteins in a similar fashion. The monoclonal antibodies, which did not precipitate any protein as observed after autoradiography, were incubated with protein A beads to which rabbit anti-mouse immunoglobulins (Dako, Glostrup, Denmark) were bound. For this purpose, protein A beads were ¢rst incubated with 20 ml of rabbit anti-mouse immunoglobulins, followed by incubation with monoclonal antibodies and radiolabelled antigens as described before. Each sample, including prestained protein molecular weight marker (Sigma-Aldrich) was boiled for 3 min in 25 ml of sample loading bu¡er (Laemmli, 1970) before electrophoresis (12.5% polyacrylamide gels under reducing conditions). Intracellular labelled antigens were also used in each gel. Gels were ¢xed (30% methanol, 10% glacial acetic acid) for 30 min and dried in a gel dryer at 808C for 1 h under vacuum. X-ray ¢lms were exposed at ^208C for 3^4 days. Autoradiographs were developed and images were scanned and aligned with the prestained molecular weight markers.
627
Antibody titres of individual MAbs were determined using an indirect ELISA. Brie£y, the ELISA plates (Nunc, Maxisorp, Roskilde, Denmark) were coated with PPR virus antigen. The plates were incubated at 378C for 1 h under constant orbital shaking. Unbound antigen was washed three times using diluted PBS (1:4). Fifty ml of blocking bu¡er (PBS with 0.1% Tween-20) was added to each well. Further, 50 ml of 2-fold-diluted hybridoma culture supernatants were incubated for 1 h at 378C and washed as before. Anti-mouse HRPO conjugate (Dako) diluted 1:1000 in blocking bu¡er was added at the rate of 50 ml to each well. Plates were further incubated for 1 h at 378C under constant shaking. Plates were washed again and OPD containing H2O2 was added to the wells. Colour reaction was observed for 10 min, followed by stopping of the reaction with an equal volume of 1 mol/L H2SO4. The optical densities of the wells were recorded at 492 nm in an ELISA-Reader. The dilution of the monoclonal antibodies that gave 75% absorbance (A492) of the plateau of the titration curve (maximum A492) was considered as the titre of the monoclonal antibodies (Saliki et al., 1993). For determination of antigenic pro¢le of the viruses, the characterized panel of monoclonal antibodies were screened in indirect ELISA with three di¡erent Indian isolates of PPR viruses, vaccine strain of rinderpest virus (RBOK strain) and Edwards `caprinized' rinderpest virus derived from Vero cells. Similarly, di¡erent virus passages of an attenuated PPR virus of Indian origin (PPRV Sungri 96) were also screened for ELISA reactivity using the panel of monoclonal antibodies. For checking MAb reactivity, respective antigens derived from Vero cells were used in 1: 2 dilution. The MAbs were also screened at a dilution of 1:2 with all the antigens according to the procedure described for indirect ELISA. Binding of di¡erent monoclonal antibodies to antigen (% reactivity) was calculated in relation to a cross-reacting MAb (clone no. 4G2). Based on reactivity of the MAbs in indirect ELISA, the correlation coe¤cient (r) between antigenic pro¢les of each isolate and also between various passages of PPRV Sungri 96 was calculated using Microsoft Excel computer software. RESULTS Production of monoclonal antibodies There was a signi¢cant increase in indirect ELISA signal after immunization of Balb/c mice when sera samples were tested on the 45th day of the immunization schedule (data not presented). Hybridoma culture supernatants, at the appropriate stage of screening, showed signi¢cantly high absorbance (A492) values (twice) with PPR virusinfected cell culture virus in comparison to the reactivity with mock-infected cell culture antigen. The reactive clones were selected for further investigation. In all, 23 PPR virus-positive primary hybridoma clones were initially identi¢ed on the basis of the reactivity in indirect ELISA. Primary clones were subjected to single-cell cloning and subcloning. The limiting-dilution method of hybridoma clones was adopted for this purpose. Monoclonality of a clone was accepted only when all the wells of a microtitre plate with growing cells gave a positive reaction in indirect ELISA after repeated subcloning.
628
Characterization of monoclonal antibodies Subsequent to production of su¤cient quantity of monoclonal antibodies, these MAbs were characterized. Important characteristics of all the monoclonal antibodies are summarized in Table I. The monoclonal antibodies were subjected to radioimmunoprecipitation assay in order to determine virion protein speci¢city. Of the 16 MAbs characterized for antigenic speci¢city, 10 apparently belong to nucleoprotein (N), which precipitated a viral protein of around 68 kDa, four against matrix (M) protein (40 kDa), one against neutralizing epitope of haemagglutinin (H) protein (80 kDa), and one against phosphoprotein (90 kDa). Interestingly, the two MAbs that were of IgM class did not bind to protein A directly and therefore were characterized indirectly after binding anti-mouse immunoglobulins to protein A. Accordingly, these two MAbs were found to be directed against N protein. The remaining seven MAbs (4A3, 4B6, 4D15, 4F1, 4F13, 4G16 and 4I10) did not precipitate any viral protein in RIPA. Photographs of SDS-PAGE followed by autoradiography of the proteins precipitated by various monoclonal antibodies are shown (Figure 1). All the monoclonal antibodies were tested for reactivity in indirect ELISA using various PPR and rinderpest virus isolates (Figure 2). A critical look at the reactivity of anti-N protein monoclonal antibodies indicates that the clones 4B8, 4E10 and 4G6 are speci¢c with strong reactivity to di¡erent PPR virus isolates. Among these, 4B8 was an exceptionally high-titred clone (1:1024) belonging to the IgM isotype. The clones such as 4D11, 4G2, 4H11 and 4I3 cross-reacted even with rinderpest virus isolates. Among these, 4G2 had equal a¤nity with all the PPR and rinderpest virus isolates. This monoclonal antibody was therefore used to compare the reactivity of other MAbs in ELISA. Among the four monoclonal antibodies directed against matrix (M) protein, two (4D16 and 4H4) were speci¢c and reacted only with PPR viruses. The other two monoclonal antibodies (4H9 and 4H10) were cross-reactive to rinderpest virus isolates as well. Anti-M protein antibodies, in general, had poor reactivity with PPR virus Izatnagar isolate compared to the other two isolates. The virus-neutralizing anti-H MAb (clone 4B11) was speci¢c to PPR virus in ELISA and, at the same time, it did not neutralize rinderpest vaccine virus (RBOK strain) in vitro. The monoclonal antibodies for which antigenic speci¢city could not be determined were either PPR-speci¢c (4B6 and 4G16) or cross-reactive. On the basis of the results described above, a set of PPRspeci¢c MAbs were identi¢ed from the available panel for di¡erentiation of PPR viruses from rinderpest viruses in indirect-ELISA (Figure 3). Correlation matrix studies on the basis of ELISA reactivity of various monoclonal antibodies showed strong antigenic correlation between all the three Indian isolates of PPR viruses. The correlation matrix could clearly di¡erentiate all the three PPR viruses from two established strains of rinderpest viruses (RBOK strain and Edwards `caprinized' rinderpest virus) based on the reactivity of all 23 MAbs (Figure 4a). Correlation coe¤cient (r values) improved when this study was limited only to anti-N protein MAbs (Figure 4b). Similar studies were also conducted to determine the relationship between various passages of parent virus (PPRV Sungri) in cell culture. The correlation coe¤cient (r) between various Vero cell passaged viruses was always
4B8 4C14 4D11 4D14 4E9 4E10 4G2 4G6 4H11 4I3 4D16 4H4 4H9 4H10 4A18 4B11d 4A3 4B6d 4D15 4F1 4F13 4G16 4I10
N
IgM IgG1 IgG1 IgG1 IgG2a IgG2a IgG1 IgG1 IgG1 IgM IgG1 IgG2a IgG2b IgG2b IgG2b IgG1 IgG1 IgG2b IgG2b IgG1 IgG1 IgG2a IgG1
Isotype 10 4 2 3 52 4 4 5 2 4 4 7 2 2 7 9 8 52 2 6 7 52 52
ELISA titre (log2)b + + ++ ++ ++ ++ ++ ++ ++ + ++ ++ ++ + + ++ ++ ++ ++ ++ ^ + +
++ + ++ ++ ++ ++ ++ ++ ++ + ++ ++ ++ + + ++ NT ++ ++ ++ ^ + ^
+ ^ ++ ++ ++ ++ ++ ++ + + + + + + ^ + NT + + + ^ + ^
^ ^ + + ^ ^ ++ ^ + ++ ^ ^ + + + ^ NT ^ + ++ ++ ^ ^
^ ^ + ^ + ^ ++ ^ + ++ ^ ^ ++ ++ ^ ^ ^ ^ ^ ++ ++ ^ ^
Reactivity with PPR and rinderpest virus strains/isolates c ööööööööööööööööööööööööööööö PPR viruses Rinderpest viruses öööööööööööööööö öööööööööö PPRV PPRV PPRV RPV RPV Sungri Chirgaon Izatnagar RBOK GTV
Antigenic speci¢city as determined in radioimmunoprecipitation assay; bELISA titre was taken as the dilution of a monoclonal antibody, which correspond to 75% of the plateau/saturation absorbance (A492) value in an indirect ELISA; cGrading of reactivity based on indirect ELISA: ++, distinct (475%); +, weak (25^ 75%); ^, none (525%); NT (not tested); dVirus neutralizing clones
a
P H Undetermined
M
Clone no.
Antigenic speci¢citya
TABLE I Characteristics of monoclonal antibodies to PPR virus
629
Figure 1. Autoradiograms showing reactivity of PPRV proteins from PPR virus-infected Vero cell lysate with MAbs in RIPA under reducing conditions. Immunoprecipitates of [35S]methionine-labelled antigens were analysed on a 12.5% gel. (a) Lane 1, pre-stained molecular weight markers; lane 2, intracellular labelled protein; lane 3, PPR immune sera; lane 4, MAb 4H11; lane 5, MAb 4G2; lane 6, MAb 4G6; lane 7, MAb 4B11. (b) Lane 1, pre-stained molecular weight markers; lane 2, MAb 4D11; lane 3, MAb 4D14; lane 4, MAb 4E9; lane 5, MAb 4E10; lane 6, MAb 4H9; lane 7, MAb 4H10; lane 8, MAb 4A18; lane 9, PPR immune sera; lane 10, intracellular labelled antigen. (c) Lane 1, prestained molecular weight markers; lane 5, MAb 4D16; lane 9, PPR immune sera; lane 10, intracellular labelled antigen. (d) Lane 1, intracellular labelled antigen; lane 2, MAb 4C17; lane 3, MAb 4B8; lane 4, MAb 4I3; lane 5, MAb 4A3; lane 6, MAb 4H4; lane 7, intracellular labelled antigen. Molecular weight markers (pre-stained, nonradioactive), from the same gel were photographed along with the autoradiograms
630
631
Figure 2. Reactivity pattern of di¡erent anti-PPR virus monoclonal antibodies to various PPR and rinderpest virus isolates in indirect ELISA. The absorbance values at 492 nm of the di¡erent antibodies binding to each isolate are expressed as the percentage value of a crossreacting monoclonal antibody 4G2
632
Figure 3. Di¡erentiation of PPRV from RPV using anti-PPRV MAbs in indirect ELISA. The reactivities of PPR virus-speci¢c monoclonal antibodies are represented in relation to a crossreacting MAb, 4G2, that does not discriminate PPR virus from rinderpest virus
very high (more than 0.9), which shows no signi¢cant di¡erence in antigenic pro¢le of the virus with the increase in passage number (between passages 14 and 77) (Figure 5). DISCUSSION PPR is one of the economically important diseases of small ruminants prevalent in several parts of the Asia and Africa. The causative agent, PPR virus, is antigenically very similar to other morbilliviruses, the closest being rinderpest virus. Precise
633
a
b Figure 4. Antigenic relationship between di¡erent PPR and rinderpest virus isolates based on reactivity pattern of all 23 MAbs (a) and only anti-N protein MAbs (b). Shaded areas show correlations. Values shown are correlation coe¤cients (r) between absorbance of di¡erent isolates
634
Figure 5. Antigenic correlations (r) between various passages of PPRV Sungri 96 based on reactivity pattern with all the monoclonal antibodies (n = 23) in indirect ELISA
laboratory diagnosis of rinderpest and PPR diseases therefore require speci¢c techniques. Further, PPR and rinderpest viruses have been found to cause clinically similar disease in small ruminants (Diallo et al., 1989). Monoclonal antibodies are important tools for antigenic characterization of various morbilliviruses and for speci¢c diagnosis of these diseases. Use of monoclonal antibodies raised against PPR virus to study morbilliviruses has been reported little, probably because of the late evidence of emergence of PPR infection in 1942 (Gargadennec and Lalanne, 1942). However, monoclonal antibodies to rinderpest virus, measles virus and canine distemper virus have been used extensively to characterize PPR viruses and other morbilliviruses (McCullough et al., 1986; Sheshberadaran et al., 1986; Sugiyama et al., 1989). In the present investigation, 23 MAbs were produced using a semi-puri¢ed preparation of an Indian virus isolate (PPRV Sungri 96) as immunogen. The majority of MAbs (at least 14) precipitated internal virion proteins (nucleoprotein and matrix protein) in RIPA. These proteins are the most abundant virus-induced proteins in infected cells of morbilliviruses (Diallo et al., 1987). It can be seen that, in the present
635
investigation, barring one (clone 4B11), MAbs against envelope proteins could not be produced. There are several reasons for this, the most important of which is the nature of the antigen used both for immunization and for initial screening. The viral envelope, being fragile in nature, is easily disrupted when subjected to concentration, puri¢cation, etc. As a result, both at the time of immunization and subsequently during screening, these proteins (or MAbs against them) are outnumbered by more abundant nucleocapsid or matrix proteins. It is now clear that, in order to generate MAbs against envelope glycoprotein (e.g. H and F proteins), a di¡erent strategy will be required, such as the use of virus-infected cellular envelope for immunization and initial screening against a formalin/acetone-¢xed infected cell monolayer or by conventional virus neutralization. A similar strategy has been described by Sugiyama and colleagues (1991) with respect to rinderpest virus. The usual principle for adopting a hybridoma-screening strategy is to select a protocol that shows similarity with the subsequent and intended use of the monoclonal antibodies. Since the main objective of the investigation was to apply these MAbs for the diagnosis of PPR virus infection using various ELISA protocols, e.g. competitive ELISA for the detection of antibody and sandwich ELISA for the detection of antigen, indirect ELISA was used extensively for quick selection of appropriate MAbs. However, there are indications that the PPR-speci¢city of the MAbs in indirect ELISA tests may not always translate into PPR-speci¢city in the competitive ELISA (Anderson and Mckay, 1994). Failure to precipitate any PPR virus-speci¢c protein by some MAbs may be attributed to low antibody titres (4B6, 4D15, 4G16 and 4I10) and non-recognition of detergent-sensitive epitopes in RIPA (McCullough et al., 1986). In vitro-translated proteins may be a good option for characterization of MAbs to detergent-sensitive epitopes, as these proteins are not a¡ected by detergents (Sugiyama et al., 1989). That not all MAbs may precipitate viral proteins had also been reported by other workers working with rinderpest and PPR viruses (Sugiyama et al., 1989; Libeau and Lefevre, 1990). Libeau and Lefevre (1990) could characterize only 6 MAbs out of 13 in RIPA for antigenic speci¢city when working with rinderpest virus monoclonal antibodies. However, the remaining MAbs in their study had very similar characteristics to anti-N MAbs. The two IgM antibodies (4B8 and 4I3) of the present study precipitated viral protein (nucleoprotein) only when anti-mouse IgG was bound to protein A. This explains the poor binding a¤nity of IgM antibodies with protein A compared to IgG antibodies. Labelling of viral protein with [35S]methionine was carried out after 70^ 80% cytopathic e¡ect (CPE) induction in the cells. Anti-N protein antibodies coprecipitated actin (DNA binding protein) along with viral nucleoprotein; these are the major background host proteins. This suggests that these viruses do not shut o¡ host cell proteins e¤ciently. Libeau and Lefevre (1990) also observed that MAbs to nucleoprotein partially co-precipitated actin (46 kDa protein) during immunoprecipitation using protein A-conjugated Sepharose beads. Some partially translated or immature N proteins may also be precipitated. Based on the reactivity pattern of di¡erent MAbs in ELISA, each virus isolate in morbilliviruses can be given an antigenic ¢ngerprint (McCullough et al., 1986). This enables di¡erentiation of various members of the morbillivirus genus and also within
636
the virus between various isolates (Libeau and Lefevre, 1990). This information may be even more meaningful than the molecular analysis of viral genomes. The degree of relatedness between di¡erent morbilliviruses can be determined by estimating the amount of antibody bound (% reactivity) to the virus, calculated from ELISA results using a cross-reactive antibody or a polyclonal serum for comparison. Based on the reactivity with PPR and rinderpest virus isolates, MAbs could be categorized into PPR-speci¢c clones (which di¡erentiate rinderpest viruses from PPR viruses) and nondiscriminating clones (which reacted equally with both the viruses). However, marginal variation was observed between di¡erent isolates of the same virus. Among the crossreacting ones, 4G2 attracted attention owing to its uniform reactivity with all the isolates of PPR and rinderpest viruses. The monoclonal antibodies directed against internal virion proteins (N and M) recognized both virus-speci¢c and cross-reactive epitopes, as also reported by other workers (Sheshberaderan et al., 1986; McCullough et al., 1986, 1991). Interestingly, the monoclonal antibodies against matrix protein showed di¡erent reactivity within PPR virus isolates. Therefore, these MAbs may be important tools for characterizing di¡erent PPR virus isolates originating from various geographical areas or lineages. Similar observations were made with rinderpest antimatrix protein MAbs (McCullough et al., 1986), with some of the MAbs being extremely restricted in their reactivity, recognizing only homologous rinderpest virus. In fact, most of anti-M protein MAbs in that study recognized only small number of RPV isolates. Further, matrix protein of morbilliviruses shows partial or least epitope homology compared to N, P and F proteins (Sheshberaderan et al., 1986). Another group of two MAbs (clones nos. 4 F13 and 4I3) surprisingly reacted poorly with PPR virus but strongly with both the attenuated strains of rinderpest viruses. This phenomenon could not be explained but was similar to an observation made by McCullough and colleagues (1986). It is possible that the epitopes of the proteins, against which these MAbs are directed, are located in rinderpest virus in a more expressive manner for optimal reactivity in ELISA than they are in PPR viruses. In spite of all these observations, further studies are needed using all lineages of both rinderpest and PPR viruses to con¢rm the overall speci¢city and sensitivity of the panel of monoclonal antibodies generated. Antigenic correlations between the three Indian PPR virus isolates and two rinderpest virus strains could clearly distinguish PPR from rinderpest viruses. Although one of the low-passaged PPR viruses (PPRV Izatnagar, passage 9) showed a slightly di¡erent correlation coe¤cient (r) in the reactivity of monoclonal antibodies compared to other two isolates. The passage level (antigenic expression) of these isolates in Vero cells may have had an e¡ect in this marginal di¡erence. The correlation matrix based on the reactivity of all the monoclonal antibodies with various passages of the same virus (PPRV Sungri-96) derived from Vero cells does not apparently di¡er in antigenic pro¢le with the increase in virus passages (from 14 to 77). This virus has been shown to be completely attenuated at passage 59 (unpublished observation). Working with the N gene of the Nigeria 75/1 PPR virus strain, 8.9% divergence at nucleotide level and 5% divergence at amino acid level has been reported between the original virulent virus and its tissue culture-attenuated strain (Ismail et al., 1995). In contradiction to the changes at nucleotide and amino acid level observed with Nigeria
637
75/1 virus, our ¢ndings indicate that the variations at epitope/antigenic level may not be evident while passaging PPR virus Sungri-96 isolate in cell culture for attenuation. The sole objective of the production of MAbs against PPR virus in the present investigation was to use the selected MAbs for the diagnosis of PPR virus infection. The extensive studies of ELISA reactivity were directed to achieving this objective, so that these ¢ndings may help in selecting appropriate monoclonal antibodies. Competitive ELISA is the most commonly used test for detection of antibodies to PPR and rinderpest viruses (Anderson et al., 1991; Singh et al., 2000). The clone designated as 4B11 in the present study, an anti-H MAb, was found to be most suitable for the stated purpose when employed in testing of more than 3000 serum samples using a competitive ELISA test. This test has a diagnostic sensitivity and speci¢city comparable with the virus neutralization test (Singh et al., 2004). Anti-N monoclonal antibody (clone no. 4G6) was found to be useful for speci¢c diagnosis of PPR virus infection from clinical samples using a sandwich ELISA (Singh, 2002). This test also had sensitivity and speci¢city similar to those of a commercially available immunocapture ELISA test (Libeau et al., 1994). N protein of morbilliviruses has always been the focus of attention for di¡erentiation of closely related viruses. The main advantage of using N protein-based diagnostics is the abundance of this protein in clinical materials (Diallo et al., 1994; Ismail et al., 1995) and the availability of speci¢c as well as crossreacting epitopes (Libeau et al., 1994) in this protein. The cross-reacting epitopes help in capturing the antigen through antibodies directed to common epitopes, while speci¢c epitopes help in detection of viruses using PPR-speci¢c monoclonal antibodies. The MAbs developed during the present investigation may prove to be an important tool for research on various morbilliviruses. Further, these MAbs are likely to be useful for immunohistochemical studies of PPR-infected tissues and for developing suitable pen-side diagnostic tests such as a chromatographic strip test or a latex agglutination test as available for rinderpest (Bruning et al., 1999; Wambura et al., 2000). Such tests are becoming increasingly popular for the diagnosis of infectious diseases. ACKNOWLEDGEMENTS This study was partly funded by National Agricultural Technology Programme (NATP) of the Indian Council of Agricultural Research. The authors are grateful to the Director, Indian Veterinary Research Institute for the facilities and to Dr R.K. Singh for the critical reading of the manuscript and suggestions. REFERENCES Anderson, J. and McKay, J.A. 1994. The identi¢cation of antibodies against peste des petits ruminants virus in cattle, sheep and goats and the possible implications to rinderpest control programs. Epidemiology and Infection, 112, 225^231 Anderson, J., McKay, J.A. and Butcher, R.N., 1991. The use of monoclonal antibodies in competition ELISA for detection of antibodies to rinderpest and peste des petits ruminants viruses. In: The Seromonitoring of Rinderpest throughout Africa: Phase I, (IAEA, Vienna), 43^53
638 Bandyopadhyay, S.K., Singh, R.P. and Chandra, U., 1999. E¤cacy of the micro-method of the assessment of neutralizing antibodies following vaccination/infection with rinderpest virus. Indian Journal of Animal Sciences, 69, 82^84 Bruning, A., Ballamy, K., Talbot, D. and Anderson, J., 1999. A rapid chromatographic strip test for the pen side diagnosis of rinderpest virus. Journal of Virological Methods, 81, 143^154 Crowley, J., Dowling, P., Mennoma, J., Silverman, J.I., Schuback, D., Cook, S.D. and Blumberg, B.M., 1988. Sequence variability and function of measles virus 3', 5' and intercistronic regions. Virology, 164, 498^506 Dhar, P., Sreenivasa, B.P., Barret, T., Corteyn, M., Singh, R.P. and Bandyopadhyay, S.K., 2002. Recent epidemiology of peste des petits ruminants. Veterinary Microbiology, 88, 153^159 Diallo, A., Barrett, T., Lefevre, P.C. and Taylor, W.P., 1987. Comparison of proteins induced in cells infected with rinderpest and peste des petits ruminants virus. Journal of General Virology, 68, 2033^ 2038 Diallo, A., Barrett, T., Barbron, M., Shaila, M., Subbarao, S.M. and Taylor, W.P., 1989. Di¡erentiation of rinderpest and peste des petits ruminants viruses using speci¢c cDNA clones. Journal of Virological Methods, 23, 127^136 Diallo, A., Barrett, T., Barbron, M., Meyer, G. and Lefevere, P.C., 1994. Cloning of the nucleocapsid protein gene of peste des petits ruminants virus: relationship to other morbilliviruses. Journal of General Virology, 7, 233^237 Gargadennec, L. and Lalanne, A., 1942. La peste des petits ruminants. Bulletin des Services Zootechniques et des Epizooties de l'Afrique Occidentale Francaise, 5, 16^21 Gibbs, E.P.J., Taylor, W.P., Lawman, M.J.P. and Bryant, J., 1979. Classi¢cation of peste des petits ruminants virus as fourth member of the genus morbillivirus. Intervirology, 11, 268^274 Harlow, E. and Lane, D., 1988. Antibodies: A Laboratory Manual, (Cold Spring Harbor Laboratory, NY, Cold Spring Harbor Laboratory Press), 148^242 Ismail, T.M., Yamanaka, M.K., Saliki, J.T., EL-Kholy, A. Mebus, C. and Yilma, T., 1995. Cloning and expression of the nucleoprotein of peste des petits ruminants virus in baculovirus for use in serological diagnosis. Virology, 20, 776^778 Kobune, F., Sakata, H., Sugiyama, M. and Sugiura, A., 1991. B95a, a marmoset lymphoblastoid cell line, as a sensitive host for rinderpest virus. Journal of General Virology, 72, 687^692 Laemmli, U.K., 1970. Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature, 277, 680^685 Libeau, G. and Lefevre, P.C., 1990. Comparison of rinderpest and peste des petits ruminants viruses using anti-nucleoprotein monoclonal antibodies. Veterinary Microbiology, 25, 1^15 Libeau, G., Diallo, A., Colas. F. and Guerre, L., 1994. Rapid di¡erential diagnosis of rinderpest and peste des petits ruminants using an immunocapture ELISA. The Veterinary Record, 134, 300^304 McCullough, K.C., Sheshberadaran, H., Norrby, E., Oat, T.U. and Crowther, J.R., 1986. Monoclonal antibodies against Morbilli viruses. Revue Scienti¢que et Technique de l'O¤ce International des Epizooties, 5, 411^427 McCullough, K.C., Obi, T.U. and Sheshberadaran, H., 1991. Identi¢cation of epitope(s) on the internal virion proteins of rinderpest virus which are absent from peste des petits ruminants virus. Veterinary Microbiology, 26, 313^321 Nakamura, J., Wagatuma, S. and Fukusho, K., 1938. On the experimental infection with rinderpest virus in the rabbit. I. Some fundamental experiments. Journal of Japanese Veterinary Science, 17, 185^204 Parida, M.M. and Bandyopadhyay, S.K., 1996. Adaptation of caprinised rinderpest virus to grow in Vero cells in vitro. Acta Virologica, 40, 45^48 Peters, J.H. and Baumgarten, H., 1992. Monoclonal Antibodies, (Berlin: Springer Verlag), 51^390 Plowright, W. and Ferris, R.D., 1962. Studies with rinderpest virus in tissue-culture. The use of attenuated culture virus as a vaccine for cattle. Research in Veterinary Science, 3, 172^182 Saliki, J.T., Libeau, G., House, J.A., Mebus, C.A. and Dubovi, E.J., 1993. A monoclonal antibody based blocking ELISA for speci¢c detection and titration of peste des petits ruminants antibody in caprine and ovine sera. Journal of Clinical Microbiology, 31, 1075^1082 Shaila, M.S., Shamaki, D., Forsyth, M.A., Diallo, A., Goatley, L., Kitching, R.P. and Barrett, T., 1996. Geographic distribution of peste des petits ruminants viruses. Virus Research, 43, 149^153 Sheshberadaran H., Norrby, E., McCullough, K.C., Carpenter, W.C. and Orvell, C., 1986. The antigenic relationship between measles, canine distemper and rinderpest viruses studied with monoclonal antibodies. Journal of General Virology, 67, 1381^1392 Sidhu, M.S., Husar, W., Cook, S.D., Dowling, P.C. and Udem, S.A., 1993. Canine distemper terminal and intergenic non-protein coding neucleotide sequences: completion of the entire CDV genome sequence. Virology, 193, 66^72
639 Singh, R.P., 2002. Production and characterization of monoclonal antibodies to peste des petits ruminants (PPR) virus, (Ph.D thesis, Deemed University, Indian Veterinary Research Institute, Izatnagar) Singh, R.P., Sreenivasa, B.P., Dhar, P., Roy, R.N. and Bandyopadhyay, S.K., 2000. Development and evaluation of a monoclonal antibody based competitive enzyme-linked immunosorbent assay for the detection of the rinderpest virus antibodies. Revue Scienti¢c et Technique de l'O¤ce International des Epizooties, 19, 754^763 Singh, R.P., Sreenivasa, B.P., Dhar, P. Shah, L.C. and Bandyopadhyay, S.K., 2004. Development of monoclonal antibody based competitive-ELISA for detection and titration of antibodies to peste des petits ruminants (PPR) virus. Veterinary Microbiology, 98, 3^15 Sreenivasa, B.P., Dhar, P., Singh, R.P., and Bandyopadhyay, S.K., 2000. Evaluation of an indigenously developed homologous live-attenuated cell culture vaccine against peste des petits ruminants infection of small ruminants. In: Proceedings of the XX Annual Conference of Indian Association of Veterinary Microbiologists, Immunologists and Specialists in Infectious Diseases and National Symposium on trends in Vaccinology for Animal Diseases, (College of Veterinary Science, G.B. Pant University of Agriculture and Technology, Pantnagar, Uttaranchal, India), 84 Sreenivasa, B.P., Dhar, P., Singh, R.P. and Bandyopadhyay, S.K., 2002. Development of peste des petits ruminants (PPR) challenge virus from a ¢eld isolate. Presented at the XIV Annual Conference and National Seminar on Management of Viral Diseases with Emphasis on Global Trade and WTO Regime of the Indian Virological Society, 18^20 January, 2002, IVRI, Hebbal, Bangalore Sugiyama, M., Minamoto, N., Kinjo, T., Hirayama, N., Sasaki, H., Yoshikawa, Y. and Yamanouchi, K., 1989. Characterization of monoclonal antibodies against four structural proteins of rinderpest virus. Journal of General Virology, 70, 2605^2613 Sugiyama, M., Minamoto, N., Kinjo, T., Hirayama, N., Asano, K., Tsokiyama-Kohara, Yoshikwa, Y. and Yamanouchi, K., 1991. Antigenic and functional characterization of rinderpest virus envelope proteins using monoclonal antibodies. Journal of General Virology, 72, 1863^1869 Wambura, P.N., Moshy, D.W., Mbise, A.N., Mollel, G.O., Taylor, W.P., Anderson, J. and Bruning, A., 2000. Diagnosis of rinderpest in Tanzania by a rapid chromatographic strip test. Tropical Animal Health and Production, 32, 141^145 (Accepted: 24 October 2003)
