Journal of General Virology (1995), 76, 3051 3058. Printed in Great Britain 3051

The phosphoprotein gene of a dolphin morbillivirus isolate exhibits genomic variation at the editing site

Gert Bol t ,* Soren A l e x a n d e r s e n t and Mere te B l ixenkrone-Mal l er

Laboratory of Virology and Immunology, Department of Veterinary Microbiology, The Royal Veterinary and Agricultural University, Biilowsvej 13, 1870 Frederiksberg C, Denmark

The nucleotide sequence of the phosphoprotein (P) gene of a dolphin morbillivirus (DMV) isolate was deter- mined. Like those of other morbilliviruses the DMV P gene encoded P and C proteins in overlapping open reading frames and V protein by editing the P gene transcript. Among P mRNA based clones the editing site variants GGGC, G G G G , GAGC and G G G G G G C predicting a P protein, and the variants G G G G C and G G G G G predicting a V protein, were found. Surpris- ingly, the three variants GGGC, G G G G and GAGC were also found among clones generated from genomic RNA of the DMV isolate. Thus, more than one viral genome type appeared to be present in cells infected

with the DMV isolate. By a similar analysis of the virus genomes in the tissue from which the DMV isolate was obtained, only the G G G C type was found, indicating that the G G G G and GAGC types arose during ad- aptation of the virus to growth in cell cultures. No editing site variants likely to have arisen by editing the G A G C type were encountered, and it remains to be determined whether mRNA encoding V protein can be transcribed from genomes with this editing site. Using antisera raised against the common N terminus and unique C termini of the predicted P and V proteins, the in vivo expression of these proteins was demonstrated.

Introduct ion

The phosphoprotein (P) genes in members of the genus Morbillivirus contain two overlapping long open reading frames (ORFs). One encodes the structural P protein, whereas the other encodes the non-structural C protein. During transcription of the P gene non-templated G residues can be incorporated in the mRNA at a specific highly conserved editing site. Insertion of one such G residue causes a frameshift in the P ORF, by which it encodes the non-structural V protein. Thus, the first 231 amino acids of the P and V proteins are identical, whereas the remaining residues are different. Throughout the text the common part of the P and V proteins will be referred to as the common N terminus, while the unique parts will be referred to as the P or V protein C termini.

The situation described above has been predicted by nucleic acid sequence analyses for measles virus (MV) (Bellini et al., 1985; Cattaneo et al., 1989), rinderpest

* Author for correspondence. Fax +45 35 28 27 42.

t Present address : Danish Veterinary Institute of Virus Research, Lindholm, Dk-4771 Kalvehave, Denmark.

The dolphin morbillivirus P gene nucleotide sequence reported in this paper has been submitted to the EMBL/GenBank databases and assigned the accession number Z47758.

virus (RPV) (Baron et al., 1993), canine distemper virus (CDV) (Barrett et al., 1985; Cattaneo et al., 1989) and phocine distemper virus (PDV) (Blixenkrone-Meller et al., 1992; Curran & Rima, 1992). However, in vivo expression of all three proteins has only been proved for MV (Bellini et al., 1985; Wardrop & Briedis, 1991). For RPV and CDV, in vivo expression has only been definitively demonstrated for the P protein (Orvell et al., 1985; Sheshberadaran et al., 1986). For PDV, in vivo expression of both the P and V proteins has been documented (A. Hu & E. Norrby, unpublished results; Rima et al., 1990; Visser et al., 1990).

In 1990-1992 a hitherto unknown morbillivirus caused a serious epidemic among Mediterranean striped dol- phins (Stenella coeruleoalba) (Domingo et al., 1990; Van Bresssem et al., 1993). We reported previously the N, M and F gene nucleotide sequences of an isolate of this dolphin morbillivirus (DMV) (Blixenkrone-Meller et al., 1994; Bolt et al., 1994). In this paper we analyse the DMV P gene and its gene products.

M e t h o d s

Sequencing strategy. Isolation and adaptation of DMV to Vero cells, total RNA extraction from cell cultures infected with the DMV isolate after it had been plaque purified four times in Vero cells, construction of an mRNA based cDNA library and screening for clones with cDNA inserts of P mRNA have been described previously (Blixenkrone-

0001-3414 © 1995 SGM

3052 G. Bol t and others

T a b l e 1. Primers used f o r reverse transcription and /or P C R o f D M V P gene sequences

Primer no. Nucleotide position Sequence (5 ' -Y)* Sense

1 458-474 2 1167-1151 3 60-74 4 82-96 5 71 0-696 6 776-790 7 777-791 8 1602-1585

GGCAGTGATGATGATGC + AGTCTGGACAGGTGGCC TT(CCCGGG) I(GGATCC)ZATGGCAGAGGAGCAG + TT(CCCGGG)a(GGATCC) 2ATGTCAATAAGGGAC + TT(CTGCAG)a(GTCGAC)4(TCA)SGGTTGGAATTGGTGG - TT(CCCGGG) 1 (GAATTC)6TGGGACGGAGACCGA + TT(CCCGGG) I(GAATTC)6GGGACGGAGACCGAG + TT(CTCGAG)V(GTCGAC)4CAATAAGATGTAGGTTGG

* 1, SmaI site; 2, BamHI site; 3, PstI site; 4, Sall site; 5, Stop codon; 6, EcoRI site; 7, XhoI site.

Moiler et al., 1994; Bolt et al., 1994). A total of 18 such clones was selected. The insert of one of these clones was sequenced completely. Of the remaining 17 P clones only the editing site regions were sequenced.

A genomic P gene region containing the editing site was PCR amplified from total RNA extracted from Vero cells infected with the DMV isolate and from total RNA extracted from the dolphin lung tissue sample that was also used for obtaining the DMV isolate. Initially, reverse transcription was carried out with primer 1 (Table 1) alone using the First-Strand cDNA synthesis kit according to the recommendations of the manufacturer (Pharmacia). In order to prevent P mRNA from acting as template in the subsequent steps 40 lag RNase A was added to the eDNA synthesis product, which was incubated at 37 °C for 30 min. The reaction was then heated to 95 °C for 5 min, primer 2 (Table 1) and Taq polymerase (AmpliTaq; Perkin Elmer) were added, and amplification was carried out under standard cycling conditions (Blixenkrone-Moller et al., 1994). The PCR products were extracted from agarose gel (GeneClean II kit; Bit 101) and cloned into pGEM-T vector (Promega). The editing sites of 24 such clones with inserts amplified from DMV infected cell culture and ten clones with inserts amplified from dolphin lung tissue were sequenced. As a control for the effect of RNase addition, the editing sites of 17 clones with inserts amplified from DMV infected cell culture without addition of RNase after eDNA synthesis were also sequenced.

All sequencing was carried out by the dideoxynucleotide chain- termination technique (Sanger et al., 1977) using the Sequenase 2.0 kit (USB).

Computer analysis of the nucleotide and deduced amino acid sequences was done as described previously (Bolt et al., 1994).

In vitro transcription and in vitro translation. Two pSPORT 1 vectors (BRL) with inserts encoding the P/C and V/C proteins, respectively, were linearized with NaeI and transcribed in vitro with T7 RNA polymerase using the Riboprobe Oemini system (Promega). The run- off transcripts were translated in vitro with nuclease treated rabbit reticulocyte lysates (Promega) and [35S]methionine. All reactions were carried out according to the recommendations of the manufacturer.

Production of anlisera against the predicted proteins of the D M V P gene. Fragments of the DMV P gene encoding protein parts were PCR amplified from linearized plasmid DNA using specific primers with restriction sites at their 5' termini. The gene fragments encoding the first 217 amino acid residues of the predicted common DMV P/V protein N terminus, the last 63 amino acids of the V protein C terminus, the last 266 amino acids of the P protein C terminus and the 177 amino acids of the entire C protein were amplified with the primer pairs 3/5, 6/8, 7/8 and 4/5 (Table 1), respectively. The PCR products were cut with the appropriate restriction enzymes and cloned into the pMAL-c2 vector (New England Biolabs) so that the DMV protein parts were encoded in continuation of the ORF encoding the maltose binding protein of Eseherichia coli. The recombinant fusion proteins were extracted from bacterial lysates using amylose resin as recommended by the manufacturer (New England Biolabs).

Rabbits were immunized subcutaneously with 10-200 lag of purified fusion protein at days 0, 14 and 28. A 1 ml volume o fa 1:1 suspension in Freund's complete adjuvant was used on day 0 and a similar suspension in Freund's incomplete adjuvant was used on days 14 and 28. The rabbits were bled on day 42.

Indirect immunofluorescence assay (IFA), Western blotting and radio- immunoprecipitation assay (RIPA). The antisera raised against the fusion proteins were used for IFA, Western blotting and RIPA.

IFA was carried out on acetone fixed cytospins of DMV infected Vero cells essentially as described previously (Blixenkrone-Moller, 1989). Blocking was done with 5 % swine serum. The antisera were used at a 1:15 dilution. Fluorescein conjugated swine anti-rabbit IgG (DAKO) diluted 1 : 25 was used as the secondary antibody.

Western blotting was done as described by Christensen et al. (1993), except that the RIPA lysis buffer described below was used for harvesting the DMV infected cells.

DMV infected Vero cells were radiolabelled with [35S]methionine as described by A. Hu & E. Norrby (unpublished). The labelled cells were harvested in R1PA lysis buffer (50 mM-Tris ItC1 pH 8.0, 150 mM-NaC1, 1 mM-EDTA, 0.1% SDS, 0.5% sodium deoxycholate, 1% Nonidet P- 40, 0.1% gelatin, 1 mM-PMSF and 0-1% aprotinin) and incubated for 1 h on ice before freezing at - 2 0 °C. Before in vivo synthesized proteins were immunoprecipitated, they were denatured by increasing the SDS concentration of 135 gl of cell lysate to 2% and heating the lysate at 70 °C for 30 min. The heated lysate was diluted to 3 ml with lysis buffer and 30 lal of antiserum was added. Immunoprecipitation of in vitro synthesized proteins was performed as described by A. Hn & E. Norrby (unpublished) but with the above lysis buffer without aprotinin. The antigen~ntibody reactions were rotated at 4 °C overnight. Binding of the immune complexes to Protein A-conjugated Sepharose was done as described by A. Hu & E. Norrby (unpublished), except that the above lysis buffer without PMSF and aprotinin was used for all suspension and washing steps. The precipitated proteins were analysed by SDS- PAGE and fluorography.

Resul ts

Nucleot ide sequence analyses

T h e n u c l e o t i d e s e q u e n c e o f t h e P g e n e o f t h e D M V

i s o l a t e is s h o w n in Fig . 1. T h e m R N A b a s e d c l o n e d i d

n o t c o n t a i n t he f i rs t 21 n u c l e o t i d e s o f t h e gene . T h e

s e q u e n c e o f t h i s s e g m e n t a n d t h e n u m b e r o f A r e s i d u e s

e n c o d e d a t t he e n d o f t h e g e n e h a v e b e e n d e t e r m i n e d

p r e v i o u s l y ( B l i x e n k r o n e - M o l l e r et al., 1994; B o l t et aI.,

1994).

The DMV P gene contains two overlapping long ORFs encoding the P/V and C proteins, respectively. As in other morbilliviruses, the start signals of these two

Genomic variation at the editing site of D M V 3053

RFI C RF3 P

RFI C RF3 P

A ~ A C ~ G T C ~ A C ~ C ~ T ~ C A T C ~ C C G A ~ C ~ C C ~ T ~ G A ~ A G ~ C ~ A T ~ T ~ A ~ A G ~ T ~ T ~ G A I 120 M S I R D L S V S N L S E 13

M A E E Q A Y H V N K G L E C L K S L R 20

~ T C ~ C C ~ A ~ T ~ ~ G C C ~ A T C ~ G ~ C ~ C ~ G ~ T ~ G C G A G A G ~ C ~ A ~ C ~ A A / L ~ A G A ~ C ' f T G A ~ 240 K I R P M L S K L R K P K L S E A R P P A K N Q A R V I T R T T P K K T L L I S 53

E N P P D A V E I K E A Q I I R S K A A C E E S S E E H H Q D N S E K D T L D F 60

RF1 C RF3 P

A ~ ~ G ~ A G A C ~ G ~ T A C ~ C A ~ A C T T G G ~ A ~ A T A ~ A ~ A G A G ~ C ~ A ~ C ~ A G A G C C ~ A ~ A G A ~ T ~ 360 T N H A L Q Q L D Q K R T A C Y L V M I Q D L E H Q V T S L M K E S P S Q E T S 93

D E S C S S A I R P E T Y R M L L G D D T G F R A P G Y I P N E G E P E P G D I I00

RF1 C RF3 P

G ~ C ~ G T A C G A ~ A C ~ A ~ A T ~ C ~ T ~ G A ~ A G T ~ A ~ A C ~ ~ C ~ G ~ A ~ A ~ A ~ G ~ T 4 8 0 E R R N L Q Y D V T M F M I T A V K R L K E S R M L T C S W F Q Q A V M M M Q N 133

G K E E P A V R C Y H V Y D H G G Q E V E G V K D A D L L V V P T G S D D D A E 140

RFI C RF3 P

T ~ G A ~ A G A ~ A G A G ~ A G A G ~ A ~ T ~ ~ A T A C ~ G A ~ C C T ~ A ~ C C ~ A ~ T ~ A G A G A T C G T ~ 600 S E T E M R A L S R A M V N L A L L I P E E I L P L T G D L L P G L R S R D R L 173

F R D G D E S S L E S D G E S G T V D T R G N S S S N R G S A P R I K V E R S S 180

RFI C RF3 P

RF2 V RF3 P

RF2 V RF3 P

RF2 V RF3 P

RF3 P

RF3 P

RF3 P

ACGI~DGAGACTATR~%GCAGTGAAGAGCTACAAGGACTGA~VFAGATCTCAGAGT~ CATAATGGATq'rGGAGTAGACAGA~C ~ T C C ~ C ~ C ~ C ~ G ~ C ~ C 720 T L R L 177

D V E T I S S E E L Q G L I R S Q S Q K H N G F G V D R F L K V P P I[ P T S V P 220

~ A C C C ~ C C ~ T C ~ A A A A A ~ A G A G A G A T ~ G C ~ A T C T G G G A ~ A G A C ~ A G ~ A ~ ~ C C ~ A ~ G ~ T ~ G A ~ 840 H R R E I S L I W D G D R V F I D R W ~ N P T ~ S R I K M G 2 6 1

L D P A P K S I K K G T G E R S A L S G T E T E F S L T G G A T R L A Q E S R W 260

~ T ~ G A G T ~ G ~ C ~ A G ~ T C ~ C ~ G T ~ A ~ G A G A ~ A C C ~ G ~ C C C C ~ A T ~ T A C ~ G C ~ C ~ G ~ T C C ~ G ~ 960 I V R V K ~ T ~ G E ~ P P V ~ D E ~ R E D P E T P T R I W Y H S L P E I P E Q W 3 0 1

A S S E S S A P A E N V R Q S V T N A E R T Q K P P Q G S G T T A S Q K S Q N N 300

G C ~ A ~ A G T A ~ G A ~ C F F r L ~ A ~ T A ~ A G A ~ C A G ~ A ~ G A ~ G A T ~ T ~ G ~ G A T ~ T A T C A A A A ~ A ~ A T ~ f PF

G H S D D E Y E D E L F b E V Q E I K T A I T K I N E D N Q Q I I S K L D S I M

1080 303 340

TACTAAAGGGTGAAA~X~AATCTATCAAGAAA CAGATCAATAAGCAAAATATCACTATATCAACTATCGAAGGC CACCTGTC~G~T~ATAG C ~ ~A~C~T~GATC 1200 L L K G E I E S I K K Q I N K Q N I T I S T I E G H L S S I M I A I P G F G K D 380

C CAATGATCCTA CTGCAGACGTAGAACTCAATC C CGATCTGAGACCCATAATTAG C CGTGATGCAGGAAGAGCTCTAGCTGAGGTCCTCAAGAGGCCAGCAGTCGAGAGAAATCCAAA~ 1320 P N D P T A D V E L N P D L R P I I S R D A G R A L A E V L K R P A V E R N P K 420

TCACCCCAAAGGTC CA~2CAGGATCCAAGGGG CAGATTC~AGGGATCTGCAACTTAAGCCGGTAGACAGGAAAA~AG~~A~TC C ~ A ~ A T ~ C ~ T ~ C 1440 V T P K V H P G S K G Q I L R D L Q L K P V D R K M S S A V G F V P T D D L P S 460

GGAGTGTGC~CG CTC CATGATTAAGTC CAG CAATCTrGl%ATCAGAACACAAA CGAAG C A T G A T A G G G C T C ' I ' I ' G A A T G A I " G T ~ G ~ A T ~ ~ A T ~ G A ~ 1560 S V L R S M I K S S N L E $ E H K R S M I G L L N D V K S G K D L G E F Y Q M ~ 500

rGA~-~+cArc~G~c~Acc~c6rAcATcrrA~Gc~rGca~crAGc'~rAG~AGr~rGrTaGTc~Ar~Ar~ 16 s 5 RF3 P V K K I I K 506

Fig. h Nucleotide sequence of the DMV P gene and the deduced amino acid sequences of the P, V and C proteins encoded by their various reading frames (RFs). The start and stop codons of the long P, V and C ORFs are in bold type• The presumed editing site is overlined and the C residues of the V protein C terminus are underlined•

ORFs are in approximately equal contexts for translation initiation (Kozak, 1991).

The sequence of the editing site (nucleotides 749-752) varied between the 18 m R N A based clones examined. Six editing site variants were encountered (Table 2); four predicted a P protein and two caused a frameshift predicting a V protein. In order to determine the genomic sequence of the editing site, 24 clones amplified from genomic R N A of the isolate were analysed, and to our surprise three editing site variants were found among these clones (Table 2). However, none of these variants caused a frameshift. We then examined the editing sites of ten clones amplified from genomic R N A in the dolphin lung tissue sample that was also used for virus isolation. Among these clones only the GGGC editing site, determined for other morbillivirus genomes (Baron et al., 1993; Blixenkrone-Moller et al., 1992; Cattaneo et al., 1989), was encountered.

The amplification of c D N A from genomic R N A was based upon the positive sense of the primer used for reverse transcription and the addition of large amounts of RNase before adding a negative sense primer. To check if the added RNase prevented m R N A from being reverse transcribed and amplified, we also cloned PCR products from an identical amplification reaction to which no RNase had been added• Among 16 such clones, we found the three editing site variants that were also found after RNase had been added plus the G G G G G variant. This indicates that the addition of RNase prior to PCR actually prevented amplification of m R N A sequences. In another control experiment an identical amount of RNase was added prior to c D N A synthesis. No D N A could be amplified or cloned from this reaction, which shows that the RNase dosage used was capable of degrading all potential template RNA. Non-specific priming on m R N A appears very unlikely, since the PCR

3054 G. Bolt and others

Table 2. Editing site variants and their frequencies among the examined clones

Nucleotide and deduced Cell culture mRNA Cell culture genomic Dolphin tissue amino acid sequence Predicted protein clones (18) clones (24) genomic clones (10)

AAAAAGGGCACAGGA 1 (6 %) 4 (17 %) l0 (100 %) K K G T G P variant 1

AAAAAGGGGACAGGA 8 (44%) 17 (71%) 0 K K G T G P variant 1

AAAAAGAGCACAGGA 4 (22 %) 3 (12 %) 0 K K S T G P variant 2

AAAAAGGGGCACAGGA 2 (11%) 0 0 K K G H R V variant 1

AAAAAGGGGGACAGGA 2 (11%) 0 0 K K G D R V variant 2

AAAAAGGGGGGCACAGGA 1 (6 %) 0 0 K K G G T G P variant 3

Table 3. Percentage identity between the nucleotide and deduced amino acid sequences o f the P gene o f D M V and those o f other morbilliviruses*

Sequence RPV 1 MV 2 CDV 3 PDV 4

P gene 66 65 64 64 P protein 49 48 48 47 P protein C terminus 58 60 57 58 V protein 45 41 41 39 V protein C terminus 69 65 54 59 C protein 46 40 40 40

* Sources: 1, Baron et al. (1993); 2, Bellini et al. (1985); Cattaneo et al. (1989); 3, Barrett et al. (1985, 1993); Sidhu et al. (1993); 4, Blixenkrone-Moller et al. (1992).

products were gel purified, and only distinct D N A bands of the correct size were used for cloning. Furthermore, only those clones with inserts of the same correct size were sequenced.

Thus, it seems that the sequence of the P gene editing site varied among the D M V genomes present in a cell culture infected with the D M V isolate. The three editing site variants, G G G C , G G G G and G A G C were found to be encoded by D M V genomes. These variants, which all predict a P protein, represented 72 % of the P m R N A based clones examined (Table 2). The m R N A variants G G G G C and G G G G G G C are probably editing pro- ducts of the G G G C genomic type, whereas the G G G G G m R N A variant appears to have been produced by editing the G G G G genomic type. No variants likely to have arisen f rom editing of the G A G C genomic type were found. Also, no variants with two or five extra G residues, predicting a truncated P protein, were encoun- tered.

The percentage nucleotide identity of the D M V P gene to those of other morbilliviruses is shown in Table 3. The results of phylogenetic analyses of the presently available morbillivirus P gene sequences and their predicted proteins (data not shown) were in agreement with previous phylogenetic analyses of D M V (Barrett et al., 1993; Blixenkrone-Moller et al., 1994; Bolt et al.,

1994), showing that D M V is not derived from any known morbillivirus but represents an independent morbillivirus lineage.

Predicted P, V and C proteins o f the D M V P gene

The amino acid sequences of the deduced proteins of the D M V P gene are shown in Fig. 1. These proteins share the characteristics of the predicted P, V and C proteins of other morbilliviruses.

The molecular mass of the deduced P protein is approximately 55 kDa. Among morbilliviruses, the C termini of the P proteins are more conserved than the N termini (Table 3). In Sendai virus the former region is believed to bind to both the nucleocapsid and the L protein (Ryan et al., 1991; Smallwood et al., 1994)

The molecular mass of the predicted V protein is approximately 35 kDa. The C terminus is well conserved, when compared to other morbilliviruses (Table 3). The seven cysteine residues of this domain are completely conserved among a considerable number of viruses belonging to the Paramyxoviridae family (Baron et al., 1993). The arrangement of these cysteine residues exhibits some resemblance to metal binding protein regions (Thomas et al., 1988), and recently the C terminus of the V protein of MV was reported to bind zinc (Liston & Briedis, 1994).

The predicted C protein has a molecular mass of approximately 20kDa . The region from residues 102-128 is very well conserved with respect to those of other morbilliviruses, but does not comply completely with the consensus sequence stated by Curran & Rima (1992).

Demonstration o f the P and V proteins in vitro and in vivo

Direct S D S - P A G E analysis of in vitro synthesis reactions showed that proteins of approximately 72 kDa and 51 kDa were produced from m R N A based clones

Genomic variation at the editing site o f D M V 3055

(a)

(b)

kDa

9 4 - - 67

43

3 0 - -

2 1 - -

14.4

kDz 94 67

43

30

21

14.4

.~--... p

• iii • ill• i !i! i! :::i/

GAGC G G G G G

1 2 3 1 2 3

• P

• ~ V

(a) DMV Mock

Immune Pre-immune Immune sera sera sera

kDa 1 2 3 1 2 3 1 2 3

21

14.4

(b)

kDa

9 4 - - 6 7 - -

4 3 - -

3 0 - -

2 1 - -

14.4 - -

DMV

Imm Pre

1 2 1 2

P

V

Fig. 2. Analysis of radiolabelled proteins synthesized in vitro from plasmids with DMV P m R N A based inserts. The positions of molecular mass markers (Pharmacia) and putative P and V proteins are indicated. (a) Direct SDS-PAGE analysis of a translation reaction of RNA in vitro transcribed from plasmids with the editing site GAGC or G G G G G or of a control reaction without RNA. (b) Immunoprecipi- tation assay of the above translation products using antisera from rabbits immunized against the common N terminus of the predicted DMV P and V proteins (lanes 1), the C terminus of the V protein (lanes 2) or the C terminus of the P protein (lanes 3).

with editing sites predicting P and V proteins, re- spectively. Proteins the size of the deduced C protein were not detected (Fig. 2a).

RIPA of the in vitro synthesized proteins demonstrated that the 51 kDa protein, as expected for a V protein, was precipitated by the antisera raised against the common N terminus of the predicted DMV P and V proteins and by the antiserum raised against the C terminus of the V protein but not by the antiserum raised against the P protein C terminus (Fig. 2b).

As expected for a P protein, the 72 kDa protein was precipitated by the antiserum raised against the common P/V N terminus and by the antiserum raised against the P protein C terminus. Surprisingly however, the 72 kDa protein was also precipitated by the antiserum raised against the V protein C terminus (Fig. 2b).

Fig. 3. Analysis of in vivo synthesized proteins. Lanes 1 3 specify the immune or pre-immune sera as described in the legend to Fig. 2 (b). The positions of molecular mass markers and of putative P and V proteins are indicated. (a) Immunoprecipitation assay of lysates of radiolabelled DMV or mock infected Vero cells. (b) Western blots of non-labelled lysates of DMV infected Vero cells.

Similar results were obtained by RIPA of in vivo synthesized proteins (Fig. 3a). However, in Western blotting the antiserum against the V protein C terminus reacted with a protein the size of the V protein, whereas no reactivity with the P protein could be detected (Fig. 3b).

Besides the putative P protein, the antiserum against the P protein C terminus also precipitated proteins giving bands of various intensities in the range 10-40 kDa from the in vitro translation reaction of the clone predicting a P protein (Fig. 2 b) and in vivo synthesized proteins in the range 30~0 kDa (Fig. 3 a). In analyses ofparamyxovirus P genes, the in vitro translation system used is known to give rise to products of internal initiation (Thomas et al., 1988). However, the antiserum against the V protein C terminus did not precipitate any comparable proteins from the in vitro translation reaction of the clone predicting a V protein (Fig. 2b). Thus, besides the P protein, the proteins precipitated by the antiserum

3056 G. Bolt and others

Fig. 4. lmmtmofluorescence assay of cytospins of DMV infected Vero cells tested 120-144 h post-infection with antiserum raised against the common N terminus of the predicted DMV P and V proteins (a), the C terminus of the V protein (b), the C terminus of the P protein (c) and pre-immune serum fi-om the rabbit immunized against the common N terminus of the P and V proteins (d). Results similar to (d) were obtained when DMV infected cells were tested with pre-immune sera of the other two rabbits and when mock infected cells were tested with the above three antisera. The bars represent 17.4 ~.tm (a), 38-1 lam (b)~ 29-1 j.tm (c) and 47.1 lam (d).

against the P protein C terminus are probably protcolysis products of the P protein. The identity of the bands at approximately 21 kDa in Western blotting (Fig. 3b) is unknown.

The molecular masses of the in vitro and in vivo synthesized DMV P and V proteins assessed by mobility in SDS PAGE are within thc ranges dctcrmined for the same proteins of other morbilliviruses. As observed for other morbilliviruses the estimates of molecular mass

based on SDS PAGE varied considerably from the molecular masses of the proteins predicted from the nucleotide sequence data. This is generally attributed to phosphorylation of the proteins, which inhibits binding of SDS, causing them to migrate more slowly in SDS PAGE than non-phosphorylated proteins of the same size (Bellini et al., 1985; Cattaneo et al., 1989; A. Hu & E. Norrby, unpublished results; Lamb & Paterson, 199 l ; Orvell et al., 1985; Wardrop & Briedis, 1991).

Genomic variation at the editing site o f D M V 3057

By IFA all three antisera raised against DMV P and V protein regions gave rise to a specific diffuse cytoplasmic fluorescence in DMV infected cells (Fig. 4).

The antiserum raised against the C protein of DMV failed to react in either Western blotting, RIPA or IFA. Some degradation of the fusion protein used for immunization against the C protein was observed upon SDS-PAGE analysis and Coomassie blue staining of the purified fusion proteins (data not shown). The degra- dation may be an explanation for the negative results obtained with this antiserum.

Discussion

We sequenced the P gene of a DMV isolate, examined P gene editing site variants among mRNA and genomic RNA, and analysed the expression of the genes encoded by the P gene.

A fragment of the P gene of another DMV isolate, corresponding to nucleotides 411-799 of the present P gene, has been sequenced previously (Barrett et al., 1993). Nucleotide differences between this and the present sequence are found at positions 412, 596, 597 and 780. All four nucleotide alterations predict amino acid differences.

The reason for the apparent cross-reactivity between the intended V protein specific antiserum and the P protein in immunoprecipitation assay but not in Western blotting is unknown. Stretches of amino acid homology between the C terminus of the predicted V protein and the entire deduced P protein capable of causing cross- reactions do not seem to exist. Other conceivable explanations for the reactivity of antiserum against the V protein C terminus with a protein the size of the P protein are ribosomal frameshifting or the existence of an alternative editing site. However, ribosomal frame- shifting is believed to require formation of an RNA pseudoknot (Brierley et al., 1989; Craigen & Caskey, 1987) and the DMV P gene appears to contain neither sequences capable of forming this structure nor an alternative editing site.

Despite the inexplicable reactivity of the intended V protein specific antiserum our results show that the P gene of DMV, like those of MV and PDV, directs the synthesis of the two proteins P and V, with common N termini and unique C termini.

In MV and PDV the P mRNA editing site variant GGGGGC with two non-templated G residues has been found. This variant predicts a truncated P/V protein called the I protein (Blixenkrone-Moller et at., 1992; Gombart et al., 1992; A. Hu & E. Norrby, unpublished results). In the present study, no mRNA based clones predicting a I protein were encountered, and no putative I proteins were detected by immunoprecipitation. Thus it

seems that the editing process of DMV is strictly controlled. A similar conclusion can be drawn for RPV (Baron et al., 1993).

In immunofluorescence assays of MV the P protein was attached to distinct cytoplasmic granules that were known to contain virus nucleocapsids, whereas the V protein seemed to be diffusely dispersed throughout the entire cytoplasm (Bellini et al., 1985; Wardrop & Briedis, 1991). In the present study all three antisera towards the P and/or V proteins of DMV gave rise to a diffuse fluorescence over the entire cytoplasm. Whether this represents a difference between DMV and MV or is simply due to the fact that we were unable to detect cells at an early stage of infection remains to be determined.

It is surprising that in the present study C protein was not detected after in vitro translation, since C protein was demonstrated using similar translation systems with in vitro or in vivo produced P mRNA of other members of the Paramyxoviridae family (Bellini et al., 1985; Cattaneo et al., 1989; Curran & Kolakofsky, 1988; Galinski et al., 1992). Furthermore, we were unable to demonstrate in vivo expression of the DMV C protein. However, this does not allow us to conclude that the DMV C protein is not expressed in vivo, especially since the antigen used for obtaining the antiserum against the C protein was not of optimal quality.

Nucleotide sequence analyses demonstrated that the three genomic editing site variants, GGGC, GGGG and GAGC were present in Vero cells infected with the DMV isolate. To our knowledge this is the first report of a morbillivirus isolate with more than one viral genome type. In all previously examined morbillivirus genomes only the GGGC type was found, and among the P mRNAs of these viruses only editing site variants likely to have arisen from insertion of extra G residues to the GGGC sequence were encountered (Baron et al., 1993; Blixenkrone-Moller et al., 1992; Cattaneo et al., 1989).

The apparent absence of the GGGG and GAGC editing site types among clones generated from DMV genomes in dolphin lung tissue indicates that these types arose during adaptation of the virus to Vero cells. The GGGG type was also amplified from a P gene segment in mRNA sense of another isolate of DMV (Barrett et al., 1993; T. Barrett, personal communication). Thus, DMV may be more likely to mutate in the editing site during adaptation than other morbilliviruses.

No editing site variants likely to be editing products of the GAGC genomic type were encountered among the mRNA based clones. Whether the absence of such variants is due to examination of an insufficient number of clones, or whether the GAGC type is unable to be edited remains to be determined. If the GAGC type does not induce editing, no V protein can be produced from genomes with this P gene editing site. If so, it would be

3058 G. Bolt and others

interesting to examine whether viruses with GAGC type genomes are able to multiply without co-infection with viruses with genomic editing sites that induce editing. Such experiments could be relevant to the study of V protein function.

A. Friis, B. Nissen, T. Storgaard, J. Christensen, U. Toftegaard, T. Dannemann, B. Viuff and B. Aasted are thanked for indispensable advice and help. Tissue samples from infected dolphins were kindly provided by M. Domingo. The comments of T. Barrett, M. D. Baron and I.R. Pedersen were very much appreciated. The study was financially supported by The Danish Agricultural and Veterinary Research Council and by VetFond, Logumkloster, Denmark.

References

BARON, M. D., SHAILA, M.S. & BARRETT, T. (1993). Cloning and sequence analysis of the phosphoprotein gene of rinderpest virus. Journal of General Virology 74, 299-304.

BARRETT, T., SHRIMPTON, S. B. & RUSSELL, S. E. H. (1985). Nucleotide sequence of the entire protein coding region of canine distemper virus polymerase-associated (P) protein mRNA. Virus Research 3, 367 372.

BARRETT, T., VISSER, I. K. G., MAMAEV, L., GOATLEY, L., VAN BRESSEM, M.-F. & OSTERHAUS, A. D. M. E. (1993). Dolphin and porpoise morbilliviruses are genetically distinct from phocine distemper virus. Virology 193, 101(~1012.

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, 908-919.

BEIXENKRONE-MOLLER, M. (1989). Detection of intracellular canine distemper virus antigen in mink inoculated with an attenuated or a virulent strain of canine distemper virus. American Journal of Veterinary Research 50, 1616-1620.

BLIXENKRONE-MOLLER, M., SHARMA, B., VARSANYI, T. M., HE, m., NORRBY, E. & KtVAMEES, J. (1992). Sequence analysis of the genes encoding the nucleocapsid protein and phosphoprotein (P) ofphocid distemper virus, and editing of the P gene transcript. Journal of General Virology 73, 885-893.

BLIXENKRONE-MOLLER, M., BOLT, G., GOTTSCHALCK, E. & KENTER, M. (1994). Comparative analysis of the gene encoding the nucleocapsid protein of dolphin morbillivirus reveals its distant evolutionary relationship to measles virus and ruminant morbilliviruses. Journal of General Virology 75, 2829-2834.

BOLT, G., BLIXENKRONE-MOLLER, M., GOTTSCHALCK, E., WISHAUPT, R. G.A., WELSH, M.J., EARLE, J. A. P. & RIMA, B.K. (1994). Nucleotide and deduced amino acid sequences of the matrix (M) and fusion (F) protein genes of cetacean morbilliviruses isolated from a porpoise and a dolphin. Virus Research 34, 291-304.

BRIERLEY, I., DIGaRD, P. & INGLIS, S. C. (1989). Characterization of an efficient coronavirus ribosomal frameshifting signal : requirement for an RNA pseudoknot. Cell 57, 537-547.

CATTANEO, R., KAELIN, K., BACZKO, K. & BILLETER, M. (1989). Measles virus editing provides an additional cysteine-rich protein. Cell 56, 759 764.

CHRISTENSEN, J., STORGAARD, T., BLOCH, B., ALEXANDERSEN, S. & AaSTED, B. (1993). Expression of Aleutian mink disease parvovirus proteins in a baculovirus vector system. Journal of Virology 67, 229 238.

CRAIGEN, W. J. & CaSKEY, C. T. (1987). Translational frameshifting: where will it stop? Cell 50, 1 2.

CURRAN, J. & KOLaKOESKY, D. (1988). Ribosomal initiation from an

ACG codon in the Sendal virus P/C mRNA. EMBO Journal 7, 245-251.

CURRAN, M. D. & RIia, B. K. (1992). The genes encoding the phospo- and matrix proteins of phocine distemper virus. Journal of General Virology 73, 1587-1591.

DOMINGO, M., FERRER, L., PUMAROLA, M., MARCO, A. J., PLANA, J., KENNEDY, S., McALISKEY, M. & RIMA, B. K. (1990). Morbillivirus in dolphins. Nature 348, 21.

GALINSKI, M. S., TROY, R. M. & BANERJEE, A. K. (1992). RNA editing in the phosphoprotein gene of the human parainfluenza virus type 3. Virology 186, 543-550.

GOMBART, A. F., HIRANO, A. & WONG, T. (1992). Expression and properties of the V protein in acute measles virus and subacute sclerosing panencephalitis virus strains. Virus Research 25, 63 78.

KOZAK, M. (1991). Structural features in eukaryotic mRNAs that modulate the initiation of translation. Journal of Biological Chemistry 266, 19867-19870.

LAMB, R. A. & PATERSON, R. G. (1991). The nonstructural proteins of paramyxoviruses. In The Paramyxoviruses, pp. 181 2•4. Edited by D. W. Kingsbury. New York: Plenum Press.

LISTON, P. & BRmDIS, D. J. (1994). Measles virus V protein bind zinc. Virology 198, 399-404.

()RVELL, C., SHESFIBERADARAN, H. & NORRBY, E. (1985). Preparation and characterization of monoclonal antibodies directed against four structural components of canine distemper virus. Journal of General Virology 66, 443-456.

RIMA, B.K., COSBY, S.L., DLrFFY, N., LYONS, C., O'LOAN, D., KENNEDY, S., McCULLOUGH, S. J., SMYTH, J. A. & McNEILLY, F. (1990). Humoral immune responses in seals infected by phocine distemper virus. Research in Veterinary Science 49, 114-116.

RYAN, K.W., MORGAN, E.M. & POR~ER, A. (1991). Two non- contiguous regions of Sendal virus P protein combine to form a single nucleocapsid binding domain. Virology 180, 126-134.

SANGER, F., NICKELEN, S. & COULSON, m. R. (1977). DNA sequencing with chain terminating inhibitors. Proceedings of the National Academy of Sciences, USA 74, 5463 5467.

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, 1381-1392.

SIDHU, M. S., HUSAR, W., COOK, S. D., DOWLING, P. C. & UDEM, S. A. (1993). Canine distemper terminal and intergenic non-protein coding nucleotide sequences: completion of the entire CDV genome sequence. Virology 193, 66-72.

SMALLWOOD, S., RYAN, K.W. & MOYER, S.A. (1994). Deletion analysis defines a carboxyl-proximal region of Sendal virus P protein that binds to the polymerase L protein. Virology 202, t 54-163.

THOMAS, S. M., LAMB, R. A. & PATERSON, R. G. (1988). Two mRNAs that differ by two nontemplated nucleotides encode the amino coterminal proteins P and V of the paramyxovirus SV5. Cell 54, 891-902.

VAN BRESSEM, M.-F., VtSSER, I. K. G., DE SWARX, R. L., ORVELL, C., STANZANI, L., ANDROUKAKI, E., SIAKAVARA, K. t~ OSTERHAUS, A. D. M.E. (1993). Dolphin morbillivirus in different parts of the Mediterranean Sea. Archives of Virology 129, 235-242.

VISSER, I. K. G., GRACHEV, M. A., ORVELL, 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. n. M. E. (1990). Comparison of two morbilliviruses from seals during outbreaks of distemper in Northwest Europe and Siberia. Archives of Virology 111, 149-164.

WARDROP, E. A. & BRIEDIS, D. J. (1991). Characterization of V protein in measles virus-infected cells. Journal of Virology 65, 3421-3428.

(Received 31 May 1995; Accepted 28 July 1995)