Immunological relationships between phocid and canine distemper virus studied with monoclonal antibodies

Transcription

1 Journal of General Virology (1990), 71, Printed in Great Britain 2085 Immunological relationships between phocid and canine distemper virus studied with monoclonal antibodies Claes Orveli, ~* Merete Bfixenkronc-MOHer, 2 Vilhjalmur Svansson z and Per Have 3 1 Department of Virology, National Bacteriological Laboratory and Karolinska Institute, S Stockholm, Sweden, 2 Department of Veterinary Virology, Royal Veterinary and Agricultural University, D-1870 Frederiksberg, Denmark and 3State Veterinary Institute of Virus Research, Lindholm, D-4771 Kalvehaves, Denmark The immunological relationships between distemper viruses, isolated from a seal and mink in Denmark and from a dog in Greenland, were investigated with 39 previously developed monoclonal antibodies (MAbs) directed against four major structural proteins of canine distemper virus (CDV). They were also investigated with 16 newly developed MAbs directed against the fusion (F) and large glycoprotein (named H in analogy with measles virus) of phocid distemper virus (PDV) isolated from a harbour seal (Phoca vitulina). These MAbs were reacted with the three different isolated viruses and with the LEC strain of measles virus, in ELISA and immunofluorescence tests. In addition, immunoprecipitation tests were carried out with some of the cross-reacting antibodies. All 55 MAbs reacted identically with distemper virus isolated from seals or mink. When the MAbs produced against CDV were tested, 37 of 39 antibodies reacted with a virus isolated from a sled dog diseased in an outbreak of distemper in Greenland prior to the epizootic among seals in the North Sea. Of the 39 antibodies, 25 reacted with PDV and distemper virus isolated from mink. Of these antibodies, only three of the nine antibodies directed against the H protein of CDV cross-reacted with PDV and distemper virus from mink. Eleven MAbs, reacting with six epitopes of the H protein of PDV, were produced. All 11 antibodies reacted with distemper virus from mink, two of the antibodies reacted with CDV and none reacted with measles virus. All five antibodies reacting with three different epitopes of the F protein of PDV reacted with distemper virus from mink and CDV. Of these five antibodies three, directed against two epitopes, reacted with measles virus. Of the two envelope proteins, the H protein shows pronounced immunological differences between PDV and CDV. In contrast, immunologically the F protein appears to be well conserved among morbilliviruses. It is concluded that the virus causing the epizootic in seals in the North Sea in 1988 may have infected mink on land, or, alternatively, the virus in the sea may have originated from virus-infected mink. Introduction In April 1988 an outbreak of an infectious disease started among harbour seals (Phoca vitulina) in the Kattegat area. The epizootic rapidly spread south to Danish waters and the Baltic Sea and west to the North Sea and the Wadden Sea. The clinical and pathological signs in the seals, including respiratory and gastrointestinal symptoms, neurological signs, cutaneous lesions and post-mortem findings resembled those found in canine distemper (Osterhaus & Vedder, 1988). The infection in seals coincided in time with the appearance of serum neutralizing antibodies against canine distemper virus (CDV). Further proof that the newly discovered virus was closely related to CDV came from experiments in which specific pathogen-flee dogs and mink were infected with materials from diseased seals. Clinical signs, the appearance of serum neutralizing antibodies against CDV and morphological identification of herring bone-like nucleocapsid structures in macrophages, all indicated the existence of a virus closely related to CDV (Osterhaus et al., 1988; Blixenkrone-M011er et al., 1989). Vaccination of seals with two inactivated CDV vaccines, a whole inactivated virus and a subunit immune-stimulating complex preparation protected the animals against the disease caused by the virus (Visser et al., 1989). The newly discovered virus was named phocid or phocine distemper virus (PDV). Immunological characterization of PDV has been carried out by immunofluorescence (IF) tests with monoclonal antibodies (MAbs) directed against CDV (Cosby et al., 1988 ; Osterhaus et al. I989b) SGM

2 2086 C. Orvell and others Twelve out of 17, and 17 out of 32 MAbs directed against CDV reacted with the two different isolates of PDV, respectively. The origin of the virus is unknown. An outbreak of an acute disease in Baikal seals (Phoca siberica) prior to the epizootic in the North Sea was found to have been caused by a virus more closely related to CDV than to PDV isolated in the North Sea (Grachev et al., 1989; Osterhaus et al., 1989a,b). There is another hypothesis for the origin of PDV that has not been proved (Dickson, 1988). An outbreak of distemper virus in Arctic foxes and huskies occurred in Greenland in the winter of 1987 to Some of the carcasses of the dead huskies were reportedly thrown into the sea and may have infected the seal population. The aim of the present study was to prepare and characterize serologicauy a number of MAbs against the large glycoprotein (H) and fusion (F) proteins of PDV. With the aid of these antibodies and previously developed antibodies against CDV (Orvell et al., 1985), an attempt was made to compare PDV to CDV and measles virus. Attempts were also made to try to find the origin of PDV by comparing it with distemper virus from mink and a virus isolate from a distemper virus-diseased sled dog in Greenland. Methods Virus strains and preparation of virus materials. Five morbillivirus strains were used in the study. PDV was isolated from a diseased seal from Gedser on the south coast of Denmark (Blixenkrone-M611er et al., 1989). The virus was first passaged in mink. Virus isolates la, 13a and 4a, from mink, were then adapted to Vero cells and used in the study. Virus isolates collected in 1989 from two diseased mink at an outbreak of distemper in a Danish mink farm located close to the sea, were also adapted to Vero cells. An isolate of distemper virus was obtained from a sled dog in Northern Greenland in The Convac strain of CDV and the LEC strain of measles virus were also used (Orvell, 1980; Sheshberadaran et al., 1986). All virus strains were propagated in Vero cells maintained in Eagle's MEM containing 2~ foetal calf serum. When the cell cultures showed pronounced cytopathic effects the medium was harvested and extracellular virions were purified according to the method described previously (Orvell, 1978). Production of hybridoma cell lines. Vero cell-grown purified virions, from mink passage 4a (Blixenkrone-M611er et al., 1989) of PDV, at a protein concentration of 0.5 mg to 1.0 mg per ml, were treated with 1 of the non-ionic detergent octyl-fl-d-glucopyranoside (octyl glycoside, Calbiochem-Behring) for 15 min at room temperature. After the treatment, two consecutive ultracentrifugations were performed, and the superuatant, which contained purified H and F proteins, was used for immunization (Orvell & Norrby, 1980). The proceciure for immunization of BALB/c mice and for establishment of mouse hybridoma cell lines have been previously described (0rvell & Grandien, 1982), with the exception that the fusions were performed with the myeloma cell line SP2/0. ELISA. This technique has been described previously (0rvell & Grandien, 1982). Purified virions of the five strains, diluted in phosphate-buffered saline to a final protein concentration of 2 gg in 100 ~tl, were used to coat wells of plastic plates. Ascites material of individual hybridoma cell lines was diluted in 10-fold dilution steps, and tested against the homologous and heterologous virus strains at the same time. All 100-fold or higher differences in titres between the homologous and heterologous virus were considered as a significant difference. A 10-fold difference in titres was considered as a positive but weak reaction with the heterologous virus strain. Plastic plates coated with parainfluenza type 3 virus (protein concentration 2 ~g in 100 ~tl per well) served as a negative control in tests with all ascites materials. ELISA was also used to determine the Ig class and subclass of individual clones, as previously described (()rvell, 1984; ()rvell et al., 1985). An ascites sample of each of the newly developed clones was added at a 10-fold dilution to five different wells of plastic plates coated with PDV. After incubation at 37 C for 1 h, the plates were washed and five different peroxidase-labelled rabbit anti-mouse immunoglobulin classes and subclasses, at a carefully titrated dilution, were added to each of the five wells. After one more hour of incubation, the plates were washed and the substrate 5-aminosalicylate was added. Radioimmune precipitation assay (RIPA). [35S]Methionine-labelled purified virions, prepared as previously described, were used as the antigen in RIPA (Orvell & Norrby, 1980). The extracellular [35S]meth. ionine-labelled virions were adjusted to RIPA buffer composition (0-1% SDS, 1% sodium deoxycholate, 1% Triton X- 100 in 0.15 M-NaC1, 0.01 M-Tris-HCI ph 7.4) in an ice bath. A 2 p.1 sample of each ascites material was added to 1 x 105 c.p.m. [35S]methionine-labelled purified virions at 0 C. The samples were processed as described previously (Orvell & Norrby, 1980). The details of the procedures for polyacrylamide gel electrophoresis and scintillation autofluorography have been described previously (Orvell, 1978). IF analysis. The technique for IF has been described previously (Norrby et al., 1982). Vero cell cultures on coverslips were infected with the five different morbillivirus strains. When cytopathic effects started to appear, the cells were fixed in cold (-20 C) acetone for 5 min and then air-dried. The different ascites materials from each of the 55 clones included in the study were used at a dilution of 1 : 10, 1 : 100 and 1 : 1000 to stain the coverslip cell cultures infected with the five virus strains. Before examination in a u.v. microscope, the ceils were counter-stained with Evans' blue. Competitition experiments using peroxidase-conjugated MAbs in ELISA. The periodate method was used to conjugate horseradish peroxidase to ammonium sulphate-precipitated amtibodies (Wilson & Nakane, 1978). Unlabelled ascites material was allowed to compete with peroxidase-labelled antibodies on virus-coated plastic plates (Convac strain or passage 4a of PDV, protein concentration 20 ~tg/ml) as has been described in detail previously (Orvell & Grandien, 1982; ()rvell, 1984). The highest 10-fold dilution of unlabelled clonal Ig that could reduce by 50% or more the A45 o caused by the labelled clonal Ig was considered to be the end titre. Results The reaction of 22 MAbs directed against the nucleocapsid protein (NP) and phospho (P) protein of CD V in serological testing with different morbillivirus strains In a previous study, 149 antibody-producing hybridoma cell lines against the Convac strain of CDV were produced and characterized by different serological tests

3 i m m m Immunological relationship of PD V and CD V 2087 Table 1. Reaction of ascites material from 22 clones reacting with the NP and P protein of CDV in serological tests with different morbillivirus strains Virus CDV DV Measles Convac Mink iso- CDV virus LEC strain PDV-D late Greenland strain Designation ELISA~ ELISA PDV-I* ELISA ELISA ELISA of clone Specificity IF IF IF IF IF IF Rinderpest virus RBOK straint IF NP - NT NPlll - rcr NP NP NT NP NP NT NP3 NT NP NP5 ~rr NP rcr total 10/10 6/10 2/3 6/10 10/ P P P3 - NT P P P P NT P5 -- NT P P1 N'r P1 NT P1 NT total 12/12 9/12 5/6 9/12 11/12 NT NT NT 0/10 1/7 - (weak) -- NT o/12 1/11 * Results obtained in a previous study by Cosby et al. (1988). t Results obtained in the study by Sheshberadaran et al (1986). J~ Test by which reactivity of antibody was assayed. NT, Not tested. II The number denotes a particular epitope of the protein. (Orvell et al., 1985). By RIPA tests with [35S]methioninelabelled purified virions, 57 clones were found to produce antibodies against NP, 22 against the P protein, 10 against the F protein and nine against the H protein. These antibodies were tested in ELISA, IF and RIPA tests with different strains of CDV, and also with measles and rinderpest viruses (t3rvell et al., 1985; Sheshberadaran et al., 1986). In the present study, MAbs that cross-reacted with measles virus were not used for the NP, P and H protein whereas, in the case of those against the F protein, antibodies that cross-reacted with measles virus were included in the study. From all the MAbs described in the previous studies, a panel of 10 antibodies against NP, 12 against the P protein, eight against the F protein and nine against the H protein, was selected for study in the present investigation. The results obtained with MAbs directed against NP and the P protein can be seen in Table 1. From competition experiments performed previously by ELISA, the antibodies against NP and P protein represented at least five and six epitopes, respectively (Orvell et al., 1985). The epitope specificity of four antibodies against NP could not be determined. None of the antibodies directed against the NP and P protein reacted with the LEC strain of measles virus in ELISA and IF tests. Of the seven antibodies directed against NP and 11 antibodies directed against protein P, one from each group reacted by IF with the RBOK strain of rinderpest virus in the study by Sheshberadaran et al. (1986). The virus isolate from a sled dog in Greenland (CDV Greenland) did not react with antibody (epitope P3), but reacted with the other 21 antibodies. Six of 10 antibodies against NP, representing a minimum of three epitopes, and nine of 12 antibodies against the P protein, representing at least four epitopes, reacted identically with mink passage PDV la, 4a and 13a, and

4 m 2088 C. Orvell and others Table 2. The reaction of ascites material from 17 clones reacting with the H and F proteins of CD V in serological tests with different morbillivirus strains CDV DV Measles Convac Mink iso- CDV virus LEC Rinderpest strain PDV-D late Greenland strain virus RBOK Designation ELISA:~ ELISA PDV-I* ELISA ELISA ELISA strain t of clone Specificity IF IF IF IF IF IF IF Virus HI H2 _ NTII w H H3 - NT H H H5 yr H6 - rcr H7 - N'r -- total 9/9 3/9 1/4 3/9 8/ F F F1 w F2 F2 w N'r NT w F1 yr w F F3 - ~rr - total 8/8 7/8 4/4 7/8 8/8 0/9 0/9 w 5/8 5/8 * Results obtained in the study by Cosby et al. (1988). t Results obtained in the study by Sheshberadaran et al. (1986). Test by which reactivity of antibody was assayed. The number denotes a particular protein epitope. ]l ~rr, Not tested. w, Weak reactivity. with the two different isolates of distemper virus from mink. Of the 22 antibodies, nine were used in the study by Cosby et al. (1988) in IF tests with an isolate of PDV from the Irish coast. Of these nine, eight antibodies reacted similarly in the two studies; only antibody reacted differently. The reaction of 17 MAbs directed against the H and F proteins of CD V in serological testing with different morbillivirus strains The antibodies against the H and F proteins were demonstrated to react with seven and three epitopes of these proteins, respectively (Orvell et al., 1985; Table 2). None of the nine antibodies directed against H reacted with measles virus or rinderpest virus in IF and ELISA tests. In contrast, antibodies against two of the three epitopes of the F protein reacted with measles and rinderpest virus. Sixteen of 17 antibodies reacted with CDV Greenland, but only a weak reaction was observed with antibody clones (H2), (F1) and (F1), The reactions of the MAbs with the three PDV mink passage isolates and the virus isolates from two distemper-diseased mink from the mink farm were identical. Only three of the nine antibodies against the H protein, representing two epitopes, reacted with these viruses in ELISA and IF tests. In contrast, seven of the eight antibodies against the F protein, representing all three epitopes of the protein, reacted with these viruses. The presence and absence of reaction of clones (epitope F3) and (epitope F3), respectively, with PDV by ELISA and IF show that these two antibodies must react with different epitopes of the protein, a result that could not be demonstrated by competitive ELISA ((grvell et al., 1985). Of the 17 MAbs, eight were used in the study by Cosby et al. (1988), and these antibodies reacted similarly with the Irish isolate of PDV. RIPA with MAbs against CDV in tests with CDV and PDV PDV-infected Vero cell bottles were labelled with [35S]methionine and extracellular virions were purified from the tissue culture fluid. In most experiments, the H

5 Immunological relationship of PD V and CD V produce antibodies directed against the H protein, five against the F protein and three against the NP protein (Fig. 1). Competition experiments using peroxidase-conjugated MAbs against the H and F protein in ELISA Fig. 1. Immune precipitation of [35S]methionine-labelled purified virions of the Danish strain of PDV with MAbs against major virus structural proteins. Lanes 4 and 11, purified PDV virions; lanes 1, 2, 3 and 5, purified PDV virions precipitated with MAbs prepared against CDV; clone directed against NP (lane 1); clones and directed against H (2 and 3); clone directed against F (5). Lanes 6 to 10 and 12 to 15 contain precipitates of purified PDV virions with MAbs prepared against that virus. Lanes 6, 8, 10 and 12 to 15 show precipitation by MAbs 1.067E5, 1.069D9, 1.070B5, 1.071E5, 1.072C4, 1.085C4 and 1.063E9, respectively. Lanes 7 and 9 show the precipitation obtained with MAbs 1.068B2 and 1.069G2, respectively. protein of PDV in extracellular virions could not be clearly identified (Fig. 1). The reason for this is not known. A selected number of MAbs against the NP, H and F proteins was used in RIPA tests with CDV and PDV. MAb clones (NP2), (H1), (H5), (H5), (F1), (F2) and (F3) were shown to precipitate three different protein bands of PDV (Fig. 1). Preparation and characterization of MAbs against PD V Study of the immunological relationships between these viruses with MAbs against PDV was also considered to be of interest. Envelope preparations were produced by treatment of purified preparations of PDV with octyl glycoside, followed by ultracentrifugations (see Methods). These preparations contained predominantly the H and F proteins of the virus, but also contained a certain amount of NP (data not shown). BALB/c mice were immunized with such antigenic preparations and their spleens were used for preparation of MAbs. From these experiments, 19 MAbs directed against PDV were obtained. By RIPA tests, 11 of these MAbs were found to Competition experiments with antibodies against the H protein are shown in Table 3. In these experiments, five antibodies were considered to react with one epitope, represented by antibodies 1.062G5 and 1.063C3 in Table 3. These antibodies were also blocked by antibody clones 1.067E5, 1.068F2, 1.070B5 and 1.072C4, which were considered to react with epitope 2 (1.067E5), epitope 3 (1.068F2) and epitope 4 (1.070B5 and 1.072C4), respectively. Antibodies to epitope 4 (1.070B5 and C4) could also block the binding of the labelled clones 1.067E2 (epitope 2) and 1.085C4 (epitope 5). The labelled clones 1.068F2 (epitope 3) and D11 (epitope 6) were not blocked by any antibody reacting with the other five epitopes. Competition experiments with the five antibodies directed against the F protein are shown in Table 4. Antibody clones 1.062E2 and 1.068B2 were considered to react with one epitope. These antibodies were blocked by each other, and also by antibody clone 1.067D2. The labelled antibody clone 1.067D2 was not blocked by any other clone, but it could itself block the binding to the ELISA plate of all other labelled clones. Finally, antibody clones 1.073G10 and 1.092C3 represented antibodies reacting with the third epitope. They were blocked by each other and by clone 1.067D2, but they were unable to block the binding to the plate of the labelled clones reacting with the other two epitopes. The reaction of 16 MAbs directed against the H and F proteins of PDV in serological testing with different morbillivirus strains The reactions of the 16 MAbs directed against the H and F proteins of PDV in tests with the morbillivirus strains are shown in Table 5. None of the 16 reacted with parainfluenza type 3 virus in ELISA. All 11 of the antibodies directed against the H protein reacted with mink distemper virus. Clones 1.069D9 (epitope H1) and Dll (epitope H6) reacted with the virus isolate from Greenland and the Convac strain of CDV. Four other antibodies, reacting with epitope H1 by competitive ELISA, did not react with CDV. This result shows that the group of antibodies reacting with epitope H1 by competitive ELISA must in fact comprise antibodies against at least two epitopes. Of the 11 antibodies directed against the H protein, none reacted with measles virus. Antibodies directed against the F protein

6 2090 C. Orvell and others Table 3. Competition experiments, using ELISA with mouse ascites material from eight clones producing antibodies against the H protein of PDV, in competition with eight peroxidase-conjugated clones of similar specificity Designation of peroxidase-conjugated clone Designation Antibody Epitope of clone subclass no G C E F B C C Dll 1.062G5 IgG1 1 *(103) (103)1" - -~ C3 IgG1 1 (101) (101) E5 IgG1 2 (103 ) (103 ) (103 ) - (102 ) (103 ) 1.068F2 IgG1 3 (103 ) (103 ) - (103 ) B5 IgG1 4 (103 ) (103 ) (103) - (103 ) (103 ) (102 ) C4 lgg3 4 (103 ) (103 ) (103 ) - (102 ) (103 ) (102 ) C4 lgm (102) - (102) D11 lgg (103) *, Inhibition of binding of peroxidase-conjugated clone. i" Titre of inhibition with inhibiting clone. -, No inhibition of binding of peroxidase-conjugated clone. Table 4. Competition experiments, using ELISA with mouse ascites material from five clones producing antibodies against the F protein of PD V, in competition with five peroxidase-conjugated clones of similar specificity Designation of peroxidase-conjugated clone Designation Antibody Epitope of clone subclass no E D B G C E2 IgG 1 1 *(102)~ -~ (10 ~) D2 IgG2b 2 (102 ) (102) (102) (102 ) (102) 1.068B2 IgG 1 1 ( 103 ) - (103) 1.073G10 IgG (104) (104) 1.092C3 IgG (104) (104) *, Inhibition of binding of peroxidase-conjugated clone. t Titre of inhibition with inhibiting clone. :~ -, No inhibiting of binding of peroxidase-conjugated clone. of PDV reacted more broadly with heterologous viruses than antibodies against the H protein. The five antibodies reacted with distemper virus from mink and the Convac strain of CDV by ELISA and IF tests. Of the five antibodies, three clones 1.062E2 (epitope F1), 1.068B2 (epitope F1) and 1.067D2 (epitope F2) reacted with the virus isolate from the sled dog in Greenland by ELISA and IF tests and three clones 1.067D2 (epitope F2), 1.073G10 (epitope F3) and 1.092C3 (epitope F3), reacted with measles virus. Discussion In the present study the immunological relationship between CDV and PDV was studied with a total of 55 MAbs against the two viruses; 39 MAbs were directed against four proteins of CDV and 16 against two proteins of PDV. With the exception of a few MAbs directed against the F protein of CDV, MAbs directed against CDV, and which also reacted with measles virus, were not used in the study. In this way, a stricter immunological comparison between CDV and PDV could be performed with antibodies directed against CDV unique epitopes. Of the thirty-nine MAbs directed against CDV, 25 also reacted with PDV. The majority of antibodies directed against the NP, P and F proteins cross-reacted with PDV. The antibodies directed against the H protein were the exception as only three of the nine MAbs directed against this protein reacted with PDV. When the 16 newly developed antibodies against the H and F protein of PDV were investigated, similar results were obtained. Only two of the 11 MAbs directed against the H protein of PDV reacted with the Convac strain of CDV. When antibodies against the F protein of PDV were tried, all five reacted with the Convac strain of CDV, and three of the five antibodies reacted with the virus isolate from the sled dog. The immunological relationship between the H protein of CDV and PDV therefore resembles the situation with other morbilli-

7 m B B m Immunological relationship of PDV and CDV 2091 Table 5. Reaction of ascites material from 16 clones reacting with the H or F protein of PD V in serological testing with different morbillivirus strains and parainfluenza virus type 3 Virus CDV Measles DV mink Convac CDV virus PDV isolate strain Greenland LEC strain Designation ELISA* ELISA ELISA ELISA ELISA of clone Specificity IF IF IF IF IF 1.062G5 H1 t C3 H E9 H D9 H E5 H E5 H F2 H B5 H C4 H C4 H D 11 H6 - total 11/11 11/11 2/11 2/I 1 0/ E2 FI B2 FI D2 F G10 F C3 F3 - total 5/5 5/5 5/5 3/5 3/5 Parainfluenza type 3 virus ELISA 0/11 0/5 * Test by which reactivity of antibody was assayed. t Number denotes a particular protein epitope. viruses, in that the H proteins are relatively unrelated immunologically (Orvell & Norrby, 1974, 1980; Norrby et al., 1985; Sheshberadaran et al., 1986). However, the immunological relationship between the H proteins of CDV and PDV seems to be closer than that found between CDV and measles virus (Table 2 and Table 5), as antibodies against PDV have been shown to neutralize the infectivity of CDV (Osterhaus & Vedder, 1988; Osterhaus et al., 1988; Blixenkrone-M6Uer et al., 1989). The other envelope protein of paramyxoviruses, the F protein, appears to be immunologically conserved between CDV and PDV. Immunological conservation of the F protein of CDV and PDV appears to extend to other members of morbilliviruses, as has been previously reported ((}rvell & Norrby, 1980; Hall et al., 1980; Norrby et al., 1985; Sheshberadaran et al., 1986). A parallel situation to PDV and CDV appears to exist with human respiratory syncytial (RS) virus, of the pneumovirus genus of paramyxoviruses, which has been divided into subgroups A and B. The large glycoprotein (named G in RS virus) differs immunologically between the two subgroups, whereas the other proteins of the virus show a close immunological relationship between subgroups (Mufson et al., 1985; Anderson et al, 1985; Orvell et al., 1987). The virus isolate from the seal was passaged in healthy mink (Blixenkrone-M611er et al., 1989). Different numbers of passages of the isolate in mink did not change the antigenic composition of the virus in tests with the MAbs. Two other isolates of PDV from the North Sea have been analysed with the MAbs against CDV (Cosby et al., 1988; Osterhaus et al., 1989b; Visser et al., 1990). With the exception of one MAb (clone 3.991, NP4), the 17 MAbs used in the study by Cosby et al. (1988) reacted identically with the Irish and Danish isolates of PDV. The PDV isolate from Holland (Osterhaus et al., 1989b; Visser et al., 1990) and the Danish PDV isolate reacted identically with the 39 MAbs against CDV and the 16 MAbs against PDV, in simultaneous ELISA testing (C. Orvell et al., unpublished data). On the basis of these results it appears likely that the three PDV isolates from the North Sea are very closely related or identical immunologically. In contrast to these virus isolates, the virus isolate from diseased seals in the Baikal sea appears to be more related to CDV than to PDV (Osterhaus et al., 1989b; Visser et al., 1990). In the present investigation, it has been demonstrated that the virus isolated from a sled dog in Greenland is closer immunologicauy to CDV than to PDV. It is therefore unlikely that the epizootic of PDV in the North and

8 2092 C. Orvell and others Baltic Seas started either in Baikal Lake or Greenland. Isolates from two distemper-diseased mink, from a mink farm in Denmark, reacted identically to the PDV isolate in tests with all 55 MAbs directed against CDV and PDV. The identical reaction of the nine MAbs against the H protein of the Convac strain of CDV and the 11 MAbs against the H protein of PDV especially, appears to be strong proof that the distemper virus from the two mink and PDV are in fact identical viruses, since only five of a total of 20 antibodies reacted with both CDV and PDV. As the mink farm was located in close proximity to the sea, and the isolates from the mink were collected after the disease became evident in the sea, the virus from seals may have infected the mink. However, at present the opposite possibility cannot be completely ruled out. It would therefore be of interest to study more virus isolates, collected before 1988, from distemperdiseased mink and other terrestrial carnivores from different geographical locations, with a larger panel of MAbs against PDV and CDV. The sequencing of the nucleotide order of the H gene of PDV, and of different isolates of distemper virus from different animal species in Europe and in different parts of the world, will also give more accurate information about where and when the virus was brought into the sea. The authors are grateful to Ms Vanoohi Shahnazarian for excellent technical assistance. References ANDERSON, L. J., HIERHOLZER, J. C., Tsou, C., HENDRY, R. M., FERNIE, B. F., STONE, Y. & MCINTOSH, K. (1985). Antigenic characterization of respiratory syncytial virus strains with monoclonal antibodies. Journal of Infectious Diseases 151, BLIXENKRONE-M6LLER, M., SVANSSON, V., HAVE, P., B6TNER, A. & NIELSEN, J. (1989). Infection studies in mink with seal-derived morbiuivirus. Archives of Virology 106, COSBY, S. L., MCQUAID, S., DUFFY, N., LYONS, C., RIMA, B. K., ALLAN, G. M., MCCULLOUGH, S. J., KENNEDY, S., SMYTH, J. A., MCNEILLY, F., CRAIG, C. & (~RVELL, C. (1988). Characterization of a seal morbillivirus. Nature, London 336, DICKSON, D. (1988). Canine distemper virus may be killing North Sea seals. Science 241, GRACHEV, M. A., KUMAREV, V. P, MAMAEV, L. V., ZORIN, V. L., BARANOVA, L. V., DENIKINA, N. N., BELIKOV, S. J., PETROV, E. A., KOLESNIK, V. S., KOLESNIK, R. S., DOROFEEV, V. M., BEIM, A. M., KUDELiN, V. N., NAGIEVA, F. G. & SiDOROV, V. N. (1989). Distemper virus in Baikal seals. Nature, London 338, 209. HALL, W. W., LAMB, R. A. & CHOI'VIN, P. W. (1980). The polypeptides of canine distemper virus: synthesis in infected cells and relatedness to the other pol..ypeptides of morbilliviruses. Virology 100, MUFSON, M. A., ORVELL, C., RAFNAR, B. & NORRBY, E. (1985). Two distinct subtypes of human respiratory syncytial virus. Journal of General Virology 66, NORRBY, E., CHEN, S.-N., TOGASHi, T., SHESBERADARAN, H. & JOHNSON, K. P. (1982). Five measles virus antigens demonstrated by use of mouse hybridoma antibodies in productively infected tissue culture cells. Archives of Virology 71, NORRBY, E., SHESHBERADARAN, H., MCCULLOUGH, K. C., CARPENTER, W, C. & ORVELL, C. (1985). Is rinderpest virus the archevirus of the morbillivirus genus? lntervirology 23, RVELL, C. (1978). Structural polypeptides of mumps virus. Journal of General Virology 41, ORVELL, C. (1980). Structural polypeptides of canine distemper virus. Archives of Virology 66, ORVELL, C. (1984). The reactions of monoclonal antibodies with structural proteins of mumps virus. Journaloflrnmunology 132, ORVELL, C. & GRANDIEN, M. (1982). The effects of monoclonal antibodies on biologic activities of structural proteins of Sendai virus. Journal of Immunology 129, ORVELL, C. & NORRBV, E. (1974). Further studies on the immunological relationships between measles, distemper and rinderpest viruses. Journal of lmmunology 113, ORVELL, C. & NORRBY, E. (1980). Immunological relationships between homologous structural polypeptides of measles and canine distemper virus. Journal of General Virology.50, ORVELL, C., SHESHBERADARAN, H. & NORRBY, E. (1985). Preparation and characterization of monoclonal antibodies directed against four structural components of canine distemper virus. Journal of General Virology 66, ORVELL, C., NORRBY, E. & MUFSON, M. A. (1987). Preparation and characterization of monoclonal antibodies directed against five structural components of human respiratory syncytial virus subgroup B. Journal of General Virology 68, OSTERHAUS, A. D. M. E. & VEDDER, E. J. (1988). Identification of a virus causing recent seal death. Nature, London 335, 20. OSTERHAUS, A. D. M. E., GROEN, J., DE VRIES, P., UYTDEHAAG, F. G. M. C., KLINGEBORN, B. & ZARNKE, R. (1988). Canine distemper virus in seals. Nature, London 335, OSTERHAUS, A. D. M. E., GROEN, J., UYTDEHAAG, F. G. M. C., VISSER, I. K. G., VAN DE BILDT, M. W., BERGMAN, A. & KLINGEBORN, B. (1989a). Distemper virus in Baikal seals. Nature, London 338, OSTERHAUS, A. D. M. E., BROEDERS, H. W. J., GROEN, J., UVTDEHAAG, F. G. M. C., VISSER, I. K. G., VAN DE BtLDX, M. W. G., ORVELL, C., KUMAREV, V. P. & ZORIN, V. L. (1989b). Different morbilliviruses in European and Siberian seals. Veterinary Record 125, SHESHBERADARAN, H., NORRBY, E., McCULLOUGH, K. C., CARPENTER, W. C. & ORVELL, C. (1986). The antigenic relationship between measles, canine distemper and rinderpest viruses studied with monoclonal antibodies. Journal of General Virology 67, VISSER, I. K. G., VAN DE BILDT, M. W. G., BRUGGE, H. N., REIJNDERS, P. J. H., VEDDER, E. J., KUtPER, J., DE VRIES, P., GROEN, J., WALVOORT, H. C., UYTDEHAAG, F. G. C. M. & OSTERHAUS, A. D. M. E. (1989). Vaccination of harbour seals (Phoca vitulina) against phocid distemper with two different inactivated distemper virus (CDV) vaccines. Vaccine 7, VISSER, I. K. G., KUMAREV, V. P., (~RVELL, C., DE VRIES, P., BROEDERS, H. W. J., VAN DE BILDT, M. W. G., GROEN, J., TEPPEMA, J. S., BURGER, M. C., UYTDEHAAG, F. G. C. M. &. OSTERHAUS, A. D. M. E. (1990). Comparison of two morbilliviruses isolated from seals during outbreaks of distemper in North-western Europe and Siberia. Archives of Virology 111, WILSON, B. M. & NAKANE, P. K. (1978). Recent developments in the periodate method of conjugating horseradish peroxidase (HRPO) to antibodies. In Immunofluorescence and Related Staining Techniques, p Edited by W. Knapp, K. Holubat & G. Wick. Amsterdam: Elsevier/North-Holland. (Received 15 February 1990; Accepted 4 May 1990)