J. gen. Virol. (1986), 67, 1381-1392. Printed in Great Britain
1381
Key words: morbilliviruses/antigenic relationships/rnonoclonal antibodies
The Antigenic Relationship between Measles, Canine Distemper and
Rinderpest Viruses Studied with Monoclonal Antibodies
By H O O S H M A N D S H E S H B E R A D A R A N , I* E R L I N G N O R R B Y , 1
K E N N E T H C. M ¢ C U L L O U G H , 2 W I L L I A M C. C A R P E N T E R 2 AND
CLAES ORVELL 1
1 Department o f Virology, Karolinska Institute, School of Medicine, SBL, S-105 21 Stockholm,
Sweden and 2 Animal Virus Research Institute, Pirbright, Woking, Surrey GU24 ONF, U.K.
(Accepted 10 March 1986)
SUMMARY
Monoclonal antibodies (MAbs) were used to delineate the antigenic relationship
between the three morbillivirus types: measles virus (MV), canine distemper virus
(CDV) and rinderpest virus (RPV). Panels of six to 31 MAbs against the
haemagglutinin (H), fusion (F), nucleocapsid protein (NP), phosphoprotein (P) and
matrix (M) proteins of MV and the H, F, NP and P proteins of CDV were employed.
Nine strains of MV, three strains of CDV and four strains of RPV were examined by
radioimmunoprecipitation assay and immune fluorescence for reactivity with the
heterologous MAbs. Overall, the NP and in particular the F proteins of the
morbilliviruses showed a high degree of epitopic homology; the P and M proteins
showed a partial epitopic homology, with the greatest variation between the M proteins
of CDV and MV; the H proteins showed a low degree of epitopic homology and then
only between MV and RPV. These data indicate that the major cross-protecting
antigen in heterotypic vaccination amongst morbilliviruses is the F antigen. The
epitopic relationships found between morbilliviruses as identified by the MAbs were
classified as follows. (i) Group-specific epitopes were present on all strains of the three
morbillivirus types. (ii) Group-cross-reactive epitopes were present on only some of the
strains from each morbillivirus type (these epitopes identified the presence of
intratypic strain variation in all proteins of all three virus types). (iii) Type-specific
epitopes, i.e. MV unique or CDV unique, were found only on the homologous
morbillivirus type. (iv) CDV-RPV intertypic and MV-RPV intertypic epitopes were,
respectively, epitopes shared by CDV and RPV but not with any MV strain, and
epitopes shared by MV and RPV but not with any CDV strain. These cross-reactivities
and type-specific reactions were obtained with the internal viral proteins (M, P and
NP). The epitopes of the F proteins were mainly group-specific and no CDV-RPV or
MV-RPV intertypic epitopes were found. The epitopes of the H protein were either
type-specific or MV-RPV intertypic. These data support the proposed evolutionary
relationship between the morbilliviruses.
INTRODUCTION
Measles virus (MV), canine distemper virus (CDV) and rinderpest virus (RPV) are members
of the morbillivirus genus of the Paramyxoviridae (Kingsbury et al., 1978, Matthews, 1982). The
immunological relationship of these viruses has been studied both by cross-neutralization in vitro
and cross-protection in vivo, but no coherent picture has emerged from these studies (see
Imagawa, 1968; Appel et al., 1981). Attempts have been made to define the immunological
relationship between individual homologous viral proteins by cross-immunoprecipitation using
polyvalent antisera. Studies on the antigenic relationship of MV to CDV detected pronounced
antigenic differences only between the haemagglutinin (H) protein of measles virus and the
equivalent protein in CDV (Stephenson & ter Meulen, 1979; Hall et al., 1980; Orvell & Norrby,
0000-6993 © 1986 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
1382
H. SHESHBERADARAN AND OTHERS
1980; Wild & Huppert, 1980; Shapshak et al., 1982; Russell, 1983). This is in contrast to peptide
m a p p i n g techniques which showed marked differences between all of the polypeptides of C D V
and MV studied (Hall et al., 1980; Rima et al., 1983). M a n y of these differences probably
represent the non-antigenic framework regions. Thus, accurate identification of homology
between different antigenic regions on morbillivirus proteins requires appropriate i m m u n o logical tests.
Parallel data on the antigenic relationship of RPV to MV and CDV are more limited. Reports
differ on which measles virus and CDV proteins are cross-immunoprecipitated by anti-RPV
sera (Stephenson & ter Meulen, 1979; Hall et al., 1980; Sato et al., 1981 ; Russell, 1983), and to
date RPV antigens have been examined in only one study by this method (Sato et al., 1981).
It is clear from the above that a detailed delineation of the antigenic relationship between the
three morbilliviruses was needed. Epitopic analysis using monoclonal antibodies (MAbs)
enables a refined determination of the degree of relatedness between homologous polypeptides
from closely related viruses or isolates. Such studies have been reported for morbilliviruses, but
they have been limited to one-way analyses using usually a small n u m b e r of anti-MV MAbs and
a single CDV or RPV strain (Birrer et al., 1981 ; G i r a u d o n & Wild, 1981 ; Trudgett et al., 1981 ;
Sato et al., 1985). In order to determine the epitopic relatedness of morbilliviruses and isolates
more confidently we have examined nine strains of MV, three strains of C D V and four strains of
RPV using panels of six to 31 MAbs against the H, fusion (F), nucleocapsid protein (NP),
phosphoprotein (P) and matrix (M) proteins of MV and CDV.
METHODS
Virus and cells. The three strains of CDV studied were Convac (CON), Rockborn (ROC) and Onderstepoort
(OND). They have been described previously (Orvell et aL, 1985). Seven of the nine strains of MV studied, i.e.
Edmonston diluted passage (EDM/DP), Hu2, Woodfolk (WDF), MVO, Mantooth (MAN), Moraten (MOR) and
LEC-KI, have been described previously (Sheshberadaran et al., 1983; LEC-KI was referred to as LEC). The
other two isolates were the subacute sclerosing panencephalitis derived strains Zistev (ZIS) (from Dr J. Sever,
NIH, Bethesda, Md., U.S.A.) and LEC-WI (eight passages in Veto cells; from Dr H. Koprowski, Wistar Institute,
Philadelphia, Pa., U.S.A.). All CDV and MV strains were propagated in Vero cells maintained in Eagle's MEM
containing 2 ~ (v/v) foetal calf serum.
The four strains of RPV studied were rinderpest bovine Old Kabete [RBOK, 107 passages in bovine kidney
(BK) cells and two passages in Veto cells], rinderpest bovine Tanzania-1 (RBT/1, eight passages in BK cells),
rinderpest Oman OX 3 (OMAN, three passages in BK cells and two passages in Vero cells) and rinderpest Yemen
C272/81 (YEMEN, three passages in BK cells, one passage through cattle, two passages in BK cells). RPV stocks
were prepared at the passages above by infecting 75 cm2 confluent BK cell monolayers at 10-2 or 10-3 dilution
depending on the initial infectivity titre of the RPV strain. The infected cells were propagated at 37 °C with
rocking using 5 ml Eagle's medium (Glasgow modification) containing 1~ (v/v) foetal calf serum. The monolayers
were re-fed every second day until viral c.p.e, was observed. Thereafter, the medium was harvested daily for
infectious virus until the monolayer was destroyed. Each of these harvestings was assayed for the presence of both
infectious and non-infectious particles. Harvests containing the lowest particle/infectivity ratio were used for
preparation of radioactive antigens.
Monoclonal antibodies. All the MAbs used in this study (Table 1) with the exception of five anti-measles virus
NP MAbs have been described in detail previously (Norrby et al., 1982; Sheshberadaran et al., 1983, 1985; Orvell
et al., 1985). Briefly, all anti-measles virus MAbs were raised against the LEC-KI strain except for three anti-H
protein MAbs (80 III B2-D, 79 XI C2-D, 79 X V 17-D)and four anti-P protein MAbs (81-53-D,81-153-D,81-36-D,
81-382-D) which were raised against the Edmonston strain; five anti-NP MAbs (AI-B2, R1-B5, R2-B4, R2-C2,
R2-C3) were raised against the Hu2 strain. All the non-LEC-KI MAbs reacted with the LEC-KI strain. The antiCDV MAbs were all raised against the CON strain of CDV.
The minimum number of epitopes (Table 1) detected by each panel of MAbs is operationally defined. Epitopic
grouping of anti-MV H MAbs is based on competitive binding assays and antigenic variants selected in vitro.
Binding of MAb members of a group are either unaffected by binding of the other MAbs or by mutations in the H
protein which abolish reactivity with the other MAbs (Sheshberadaran et al., 1983; Sheshberadaran & Norrby,
1985). Epitopic grouping of the anti-MV F, M, P and NP MAbs is similarly based on competitive binding assays
or differential reactivity with various strains of MV (see Sheshberadaran et al., 1983, 1985; Sheshberadaran &
Norrby, 1984). Epitopic grouping of all the anti-CDV MAb panels is based on competitive binding assays (~}rvell
et al., 1985). Epitopic groupings of the MAb panels are referred to as groups in Tables 2 to 8.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
Epitopes shared by morbilliviruses
1383
Table 1. MAbs to structural proteins of M V and C D V employed in the present study
Specificity of MAb
f
Virus
MV
CDV
Protein
H
P
NP
F
M
H
P
NP
F
No. of MAbs
l0
6
9
II
9
9
17
31
10
Epitopic grouping
of MAbs
8
3
5
6
6
7
6
18
3
Radioimmunoprecipitation assay (RIPA). Intracellular virus antigens radiolabelled with either [3sS]methionine
or a mixture of D-[2-aH(N)]mannoseand D-[l,6-3H(N)]glucosaminehydrochloride were prepared and RIPA
performed basically as described previously (Sheshberadaran et al., 1983). The modifications involved were (i) a
modified RIPA buffer [2% (v/v) Triton X-100, 0.15 M-NaCI, 0.6 M-KCI, 5 raM-sodium EDTA, 3 mMphenylmethylsulphonylfluoride, 1% (w/v) trasylol aprotinin, 2-5 mM-iodoacetamide,0.01 M-Tris-HC1pH 7.8], (ii)
solubilization of radiolabelled virus-infected cells from 175 cm2 tissue culture flasks in 8 ml RIPA buffer, and (iii)
substitution of 0.01 M-Tris-HCl, 0.15 M-NaC1pH 8.0 for 0.01 M-phosphate buffer pH 7.2. Radiolabelled antigens
of all viruses were prepared in Vero cells and the lysates stored at - 20 °C until used. The RPV lysates were
inactivated with acetylethyleneimine(finalconcentration 0.05%, v/v) for 24 h at room temperature in the dark and
their pH adjusted to 7.8 prior to use.
SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Conditions of RIPA sample preparation and
discontinuous slab gel SDS PAGE were as described previously (Sheshberadaran et al., 1983). The spacer gel
consisted of 4.5% (w/v) acrylamide and 0.12% (w/v) N,N'-methylenebisacrylamide (bis) and the separation gel
consisted of either l0 or 12.5% (w/v) acrylamide and 0-27 or 0.34% (w/v) his respectively. Gels were fixed in a
solution of 10% (v/v) glacial acetic acid and 30% (v/v) methanol, processed for fluorography (Bonnet & Laskey,
1974) using 'Enlightening', dried and exposed to Kodak X-Omat AR-5 films at - 7 0 °C.
Immunofluorescence (IF). Infected cells on glass slides were fixed in acetone at - 20 °C for 10 min and air-dried.
IF analysis was performed as described previously (Norrby et al., 1982).
Isotopes and chemicals. [ass]Methionine (sp. act. 1000 to 1200 Ci/mmol), D-[2-3H(N)]mannose(sp. act. 15.8
Ci/mmol), o-[l,6-3H(N)]glucosaminehydrochloride (sp. act. 31 to 33 Ci/mmol), 14C-methylatedprotein molecular
weight markers and 'Enlightening' were purchased from New England Nuclear.
RESULTS
The surface glycoproteins
Of the ten anti-MV H protein MAbs, none reacted with any of the C D V strains in R I P A (Fig.
1) or in IF. Similarly none of these MAbs reacted with any of the RPV strains in RIPA.
However, by I F some of these MAbs did react with the RPV strains and epitopic differences in
the H protein could be detected between some RPV strains (Table 2). Of the nine anti-CDV H
protein MAbs none reacted with any of the MV or RPV strains in either R I P A or IF.
I n contrast to the H protein, a high degree of cross-reactivity was found by R I P A and IF
between the F proteins of CDV, MV and RPV strains (Table 3; R I P A exemplified for CDV,
Fig. 1). Using the anti-MV F protein MAbs, CDV and RPV F proteins exhibited some epitopic
differences from MV F and also showed small differences from each other. A limited difference
between strains of each virus was also detected. Similarly, the anti-CDV F protein MAbs
showed that MV and RPV F proteins were slightly different from CDV F, and strains within a
virus type could also show minor differences.
The matrix protein
As no anti-CDV M protein MAbs were available, only a one-way analysis could be carried out
using the nine anti-MV M protein MAbs. Reactivity in I F was not examined. By R I P A , the M
proteins of the CDV strains had different degrees of homology to the M protein of MV (Table 4,
Fig. 2). The M protein of the C O N strain reacted with five, the ROC strain with one and the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
1384
H. S H E S H B E R A D A R A N AND O T H E R S
(a)
I
(b)
(a)
I I
1
I
I
2
3
4
5
6
7
8
9
10
11
12
I
13
14
15
H
P
NP
F1
M
F2
Fig:l. R e a c t i v i t y o f C D V s u r f a c e c o m p o n e n t s w i t h a n t i - M V F ( l a n e s 3 t o l l ) a n d a n t i - M V H ( l a n e s 13
to 15) MAbs. The C O N strain of CDV was radiolabelled with [3sS]methionine (a) or 3H-sugars (b) and
immunoprecipitated with the following anti-MV MAbs: lane 3, 9-DB10; lane 4, 19-FF10; lane 5, 19GD6; lane 6, 19-HC4; lane 7 19-HB4; lane 8, 19-FF4; lane 9, 16-AG5; lane 10, 16-DC9; lane 11, 16EE8; lane 13,1-12; lane 14, 1-41 ; lane 15, 16-DE6. Lanes 1 and 2 show purified extracellular C O N virus
polypeptide profiles. Lane 12 shows t*C-methylated, mol. wt. marker proteins ( x t0-3): myosin, 200;
phosphorylase b, 97.5; bovine serum albumin, 69; ovalbumin, 46; carbonic anhydrase, 30; tysozyme,
14.3. Samples were run on a 12.5~ gel.
T a b l e 2.
Reactivity of anti-MV H protein MAbs with different R P V strains in
immunofluorescence tests*
MAb
A
t"
Group
no.
la
lb
2
3
4
5
6a
6b
7
8
,
Clone
no.
r
RBOK
16-DE6
80 III B2-D
1-41
1-44
16-CDI 1
7-AG 11
1-12
1-29
79 V V17-D
79 XI C2-D
+
RBT/I
+
+w
-
+~
.
+~
+~
Virus strain
A
OMAN
.
.
+w
_
-
-
-
+~
_
+~
+
+w
+w
.
+~
_
+"
* Grading of reactivity: + , distinct; + " , weak;
YEMEN
+
+~
+,,
-
, none.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
Epitopes shared by morbilliviruses
1385
Table 3. Reactivity of anti-MV F and anti-CD V F protein MAbs with different heterologous
morbillivirus strains in RIPA and IF*
Anti-MV F MAb
A.
(
Group
no.
l
2
3
4
5
6
Anti-CDV F
MAb
CDV strain
•
Clone
no.
9-DB10
16-AE7
16-AG5
16-DC9
16-EE8
19-BG4
19-FF4
19-FFI0
19-GD6
19-HB4
19-HC4
¢
CON
+
-(++)t
+
+
+
+
+
+
(+w)
+~
+,,
RPV strain
"h
ROC, OND
+
+
+
+
+
+
+
(+,,)
¢
RBOK
+
(+w)
+
+
+
+
+~'
+
+~
"~
RBT/1, O M A N
+
+
+
+
+
+~
+
+w
YEMEN
+
_(+w)
+
+
+
+
+
+
+~
_
+
+
+
_
+
+
+
MV strain
h
-~
Group Clone
MAN,
LEC-WI,
•
no.
no. LEC-KI Hu2 EDM/DP:~ MOR:~ MVO, WDF, ZIS§ RBOK
+
+
+
NTII
+
l
3.551
+
+
+
+
+
+
3.584 +
+
+
+
+
+
3.697 +
+~
.
4.068 -(++)
5.027 +
q"4+
NT
+
2
3.608 +
+
+
+
NT
+
3.633 +
+
+
+
NT
"4+
+
+
+
+
5.086 +
3
4.985
5.148 -
RPV strain
~"
RBT/1 OMAN YEMEN
+ w
_(+)
_(+)
.
.
+
_(+)
+w(+)
.
+
+w(_)
+()
(+)
+
+
+
+
+
+
-4-
-f-
+
+
+
+
* Grading of reactivity: +, distinct; +w, weak; -, none.
t IF reactivity is indicated in parentheses only where a difference from RIPA reactivity was noted. Also see
footnotes :~ and §.
:~Not tested in RIPA, IF data only.
§ Not tested in IF, RIPA data only.
1[NT, Not tested.
O N D strain with none of the nine anti-MV M protein MAbs. A partial epitopic homology also
was found by R I P A between the M protein of the four RPV strains and MV (Table 4), however,
no epitopic difference could be detected between different RPV strains. The epitopes shared by
the RPV and MV strains were not always the same as those shared by MV and CDV.
The phosphoprotein
In RIPA, none of the six anti-MV P protein MAbs reacted with the CDV strains and only one
M A b reacted with the RPV strains, although in IF two MAbs reacted weakly with the CDV
strains and strongly with the RPV strains (Table 5).
A similar situation was found using the 17 anti-CDV P protein MAbs. In R I P A only one
reacted weakly with the MV strains and strongly with the R P V strains, whereas in IF additional
reactivity occurred with some MAbs (Table 6). The anti-CDV P protein MAbs also detected
some strain-specific differences among the MV strains as well as the RPV strains in I F (Table 6).
The nucleocapsid protein
The cross-reactivity of the 31 anti-CDV NP MAbs with the RPV and MV strains is shown in
Table 7. Taking the R I P A and IF patterns of cross-reactivity together, four major classes of N P
epitopes emerged. Of the 31 anti-CDV N P MAbs, seven cross-reacted with all RPV and measles
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
1386
H. S H E S H B E R A D A R A N
AND OTHERS
(a)
(b)
J"
I
l
2
3
4
5
6
7
I
8
9
I
l
2
3
5
4
6
MD-
Fig. 2. Reactivity of the M protein of the C O N (a) and R O C (b) strains of C D V with anti-MV M M A b s
in RIPA. Irnmunoprecipitates of [35S]methionine-labelled antigens were analysed on a 1 0 ~ gel. M A b s
used were: lane 1, 10-EFI0; lane 2, 19-AGI0; lane 3, 19-CG6; lane 4, 19-DC5 ; lane 5, 16-BB2; lane 6,
19-DF10; lane 7, 19-HC5; lane 8, 19-HF6. Lane 9 shows mol. wt. marker proteins (see legend to Fig. 1).
The low mol. wt. bands (also seen with MV and RPV preparations) varied between experiments and are
interpreted to be breakdown products of the M protein.
T a b l e 4.
Reactivity of anti-MV M protein MAbs with different strains of CDV and R P V in
RIPA*
MAb
x
Group
no.
Clone
no.
1
10-EFI0
2
3
4
16-BB2
19-AGI0
19-CG6
19-GF6
19-DC5
19-DF10
19-HC5
19-HF6
5
6
CON
+w
+
+.
+w
+
.
.
.
C D V strain
~"
ROC
ONI~
RPV
all
strains t
-
-
+
_
-
_
-
+
+
_
-
_
-
.
.
.
_
+
.
.
.
.
.
.
.
.
.
* Grading of reactivity: + , distinct; + ~ , weak; -, none.
t All four strains (RBOK, RBT/1, O M A N and Y E M E N ) gave the same pattern of reactivity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
Epitopes shared by morbilliviruses
1387
Table 5. Reactivity of anti-MV P protein MAbs with CDV and R P V strains in RIPA and IF*
MAb
Group
Clone
no.
no.
1
C D V all
strainst
16-AFIO
16-BD3
81-53-D
81-153-D
81-36-D
81-382-D
2
3
R P V all
strains~
( )§
-(-)
( + w)
_(+w)
-(-)
-(-)
-(-)
-(-)
+(+)
_(+)
-(-)
-(-)
* G r a d i n g of reactivity; + , distinct; + ~ , weak; -, none.
t N a m e l y C O N , R O C and O N D .
N a m e l y R B O K , RBT/1, O M A N and Y E M E N .
§ I F reactivity is s h o w n in parentheses.
Table 6. Reactivity of anti-CDV P protein MAbs with M V and R P V strains in RIPA and IF*
MAb
MV strain
•
~
Group
•
no.
Clone
no.
LEC-KI,
M A N , Hu2
1
3.576
(-)t
-(-)
_(+w)
+w(+)
-(-)
-(-)
-(-)
-(-)
-(-)
-(-)
.
.
-(-)
.
( )
-(-)
3.698
2
3
4
5
6
3.801
3.803
4.415
4.420
4.174
3.788
3.568
3.802
3.630
4.051
3.768
3.780
4.088
3.695
4.215
RPV strain
¢
WDF,
MOR
LEC-WI, EDM/DP,
M V O , ZIS
_(+)
_(+)
_(+)
+w(+)
.
.
.
-
.
.
.
.
-
RBT/1
YEMEN
_ ( + w)
_ ( + w)
(+)
_
_(_)
_(_)
NT]~
_( )
_(+w)
_(+w)
+w
NT
.
.
NT
NT
+(+)
-
+(+)
-
+(+)
-
-( )
-( )
-(-)
(-)
( )
-(-)
-
-( )
( )
-
-(-)
-(-)
-(-)
G)
-
.
.
"1
RBOK,
OMAN
_
.
.
f
(+)
.
.
.
.
NT
-
.
.
-(-)
* G r a d i n g of reactivity: + , distinct; +w, weak; -, none.
~" I F reactivity was examined only in selected cases and is s h o w n in parentheses.
:~ NT, N o t tested.
virus strains (Class A), 12 cross-reacted with some but not all heterologous virus strains (Class
B), five cross-reacted with all or some RPV strains but not with any measles virus strain (Class
C) and seven did not cross-react with any of the heterologous virus strains (Class D).
A parallel cross-reactivity pattern was found using the anti-measles virus NP MAbs (Table 8).
Of the nine anti-measles virus N P MAbs, two cross-reacted with all RPV and CDV strains
(Class A), one cross-reacted with some but not all heterologous virus strains (Class B), three
cross-reacted with only some RPV strains but not with any CDV strain (Class C) and three did
not cross-react with any heterologous virus strain (Class D).
DISCUSSION
In this study, panels of MAbs against MV and CDV structural polypeptides were used to
investigate the antigenic relationship between the three morbilliviruses MV, CDV and RPV.
MAb cross-reactivities with the heterologous virus strains were assayed by both radioimmunoprecipitation assay and immunofluorescent antibody binding to reduce assay bias in assessing
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
4.966~+/+
5.213 J
4.493
+/+
4.317
+w/+
5.227
+/+
5.045
+w/+
3.570
/+
4.946 7
3.552 ~ - + / +
3.805 3
5.152
-/+
4.359
/+
4.973
+/+
4.304
+/+
4.147
-/+
4.080
/+
3.766 7
3.836 ~- - / +
4,276,)
4.229
+/+
4.289
+/+
3.538"~
3,740 . J - /
3.755
-/+
3.564 7
3.721 |
3.740 |
3.958 ~--/
3,991 l
4.100 /
4.1883
3
4
5
6
u
7
u
8
9
u
10
u
u
u
u
11
11
12
13
u
u
14
u
15
16
17
18
RBOK
1
2
u
u
Clone
no.
A
/
/
/
/
//
-/+
-/+
+/+
+/+
+/+
+/+
+/+
+/+
+w/+
_/+
+/+
+/+
-/+
/+
+w/+
+w/+
+/+
+w/+
+/+
+/+
+/+
_/+
/+
+/+
+w/+
+w/+
+w/+
I+
_/+
/+
+/+
OMAN
+/+
RBT/I
RPV strain
/
-/+
/
+/+
+/+
/+
+/+
+/+
+w/+
+w/+
+~/+
_/+
+/+
-/+
/+
-/+
+w/+
+w/+
+/+
YEMEN
)
-/ .
-/ .
+/+
+w/+
+w/+
+/+
+/+
+ w/_
+w/+
-/+
+/+
+/+
+~/+
+w/+
/+
+w/+
/+
+/+
+w/+
+/+
LEC-KI
.
.
.
.
_
+w
_
_
.
+
+
.
+
+
+~
+
+
+w
_
+
+w
+
EDM/DP
.
+
_
+
+
+w
+
+w
+w
+
+w
+
LEC-W1
.
.
.
+w
+w
_/+
+.
+/+
+,v/+
_/+
+/+
_/_
+
+/+
+w
+/+
Hu2
.
MV
.
.
_
+w
.
-
+
+
+
+
+~,
+
+w
+w
_
+
+w
+
WDF
.
_
+w
+
+
+
+
+,~
+
+w
+w
+
+w
+
MVO
/-
/-
+/+
+/+
+/+
+/+
+/+
+ w/_
+w/+
/+
+/+
+/+
+~/+
+w/+
+w/+
/+
-/+
+/+
+w/+
+/+
MAN
-
+w
-
+
+
+w
+
+
+~
+w
+
+
+w
+
+w
+
ZIS
-
-
_
+~
-
+
+
+
+w
+w
+
_
+w
+~
+
+w
+
MOR
* Grading of reactivity in RIPA: + , distinct; + w, weak; - , none. Grading of reactivity in IF : + , positive; - , negative. Single entry shows R I P A reactivity only
(IF reactivity not determined). Two entries: R I P A / I F reactivities.
l Groups (1-18) show the epitopic grouping of the MAbs as determined by competitive binding studies. The epitopic grouping of Group u MAbs has not been
determined. Classes (A-D) are based on the particular pattern of reactivity of the MAbs with heterologous virus strains. For details see text.
D
C
B
no.
Group
no.
t
Class
MAb t
T a b l e 7. Reactivity o f anti-CDV N P M A b s with R P V and M V strains in R I P A and IF*
t'rl
o,...l
m
Z
>
~Z
t'rJ
Oo
Epitopes shared by morbilliviruses
1389
Table 8. Reactivity of anti-MV NP MAbs with different strains of C D V and R P V in IF and
RIPA*
Class
no.
A
B
C
D
MAb
*
Group
no.
1
u
2
2
3
u
3
4
5
CDV strain
~"
Clone
no.
AI-B2
R2-B4
R2-C3
R1-B5
16-AC5
R2-C2
16-EE9
16-BB8
16-CF7
CON
+
+
+w(_)t
.
ROC
+
+
+w( )
.
.
.
OND
+
+
_
-
+(-)
NT (-)
.
.
.
RPV strain
*
R B T / 1 OMAN
+
+
+
+
+(_)
+( )
+(-)
+( )
NT ( )
NT (-)
~
KBOK
+
+
YEME
+
+
+(_)
+(-)
+( )
+( )
N~
.
.
.
.
.
.
* Grading of reactivity: +, distinct; +w, weak; , none.
~fRIPA reactivity is indicated in parentheses only where a difference with IF reactivity was noted or where IF
was not tested (yz).
epitopic homology. Differential assay reactivity in such comparative studies has been discussed
in detail previously (Yewdell & Gerhard, 1981 ; Sheshberadaran et al., 1983 ; Underwood, 1985).
Of the five structural proteins studied, the H protein was found to be the most distinct
between the morbilliviruses. No epitopic relationship was seen between MV and C D V or C D V
and RPV, but six epitopes related between MV and R P V were found. These six anti-MV H
protein MAbs would not immunoprecipitate the cross-reactive R P V protein, suggesting that
either RPV exhibited a lower density of these epitopes than MV or the affinity of the MAbs was
lower for R P V than for MV. Some of the RPV isolates also showed strain-specific variations in
reactivity with the anti-MV H protein MAbs. Such strain-specific variations in the H proteins of
MV and CDV have been reported previously, in studies on the homologous virus strains using
these MAbs (Sheshberadaran et al., 1983; Orvell et al., 1985).
The above data support previous studies which reported weak or no cross-immunoprecipitation of CDV and MV H protein by the respective antisera (Stephenson & ter Meulen, i979;
Hall et al., 1980; Orvell & Norrby, 1980; Wild & Huppert, 1980; Russell, 1983). Furthermore,
the closer homology between the H proteins of MV and R P V supports previous studies on antiMV haemagglutination inhibiting (HI) activity in anti-RPV and anti-CDV sera (Waterson et
al., 1963; Orvell & Norrby, 1974; Sato et al., 1981). Anti-RPV sera were shown to possess low
levels of anti-MV HI activity but the levels were consistently higher than that observed in antiC D V sera.
In contrast to the H protein, there was considerable epitopic relatedness between the F
proteins of the three morbilliviruses. In fact, the F proteins showed the highest degree of
epitopic conservation of all the structural proteins studied. Only one epitope on the C D V F
protein, and none on MV F protein, was found to be unique to the homologous virus. This
relative epitopic conservation of the F proteins agrees with earlier serological studies (Orvell &
Norrby, 1974; Norrby & Appel, 1980; Sato et al., 1981), but is in contrast to the report of Hall et
al. (1980) who found very little similarity between the F1 components of F of a C D V and a
measles virus strain using tryptic peptide fingerprinting. It is important to note that biochemical
methods such as tryptic peptide or oligonucleotide fingerprinting would show all sequence
changes including those which may be of little or no importance to the antigenic characteristics
of the viral protein under study (framework region changes). Only those alterations in a viral
protein which affect its antigenicity will be relevant to the immunobiology of the disease
associated with that virus, and consequently play important roles in homo- and heterotypic
vaccination. Furthermore, other factors (see below) such as minor strain-specific variations
would further complicate interpretation of the data obtained by the biochemical approaches
mentioned above.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
1390
H. SHESHBERADARAN AND OTHERS
Recently, Appel et al. 0984) proposed that the F protein of MV was primarily responsible for
cross-protection in dogs. The proposal was based on the anamnestic MV haemolysin-inhibiting
antibody response seen in dogs vaccinated with live MV and challenged with virulent CDV. The
data from the present study support this observation, and further indicate that the major crossprotecting antigen between all three morbilliviruses is probably the F protein. It is possible that
mice will recognize a different spectrum of sites on the virion as antigenic compared with
humans (MV), dogs (CDV) or cattle/sheep (RPV). However, post-infection cattle sera have
been shown by competition ELISA to have antibody specificities similar to anti-RPV MAb (T.
Obi & K. C. McCullough, unpublished data).
The epitopic conservation of the F protein compared with the marked variability of the H
protein between morbilliviruses probably reflects functional pressures operative during their
evolution. As the H protein's primary function is cell receptor binding, its characteristics will
therefore be most closeIy related to the effective infectivity and host specificity of the virus. In
contrast, the function of F protein-associated fusion activity in cell-to-cell spread of virus
infection and consequently in pathogenesis in vivo, may represent the common positive selection
pressure on this protein in all morbilliviruses.
As no anti-CDV M protein MAbs were available, only a one-way analysis of CDV and RPV
strains using nine anti-MV M protein MAbs was carried out. The M proteins of CDV and RPV
were found to be partially homologous to that of MV. Although the four RPV strains showed the
same epitopic homologies to MV, the three CDV strains showed a high degree of variation; only
the ROC strain and in particular the CON strain showed epitopic homologies with MV M
protein. Previously it has been shown that a high degree of epitopic variation exists among the
M proteins of different strains of MV (Sheshberadaran et al., 1983). It appears that a similar
situation exists among strains of CDV but not with RPV. Although both MV and CDV can
cause chronic neurological diseases in their natural hosts, which at least in the case of the MVassociated subacute sclerosing panencephalitis is thought to involve an abnormal M protein, no
such disease has been found to date involving RPV (see Appel et al., 1981 ; ter Meulen & Carter,
1982). However, this may be due to the shorter life-span of European cattle because of
agricultural practices. More conclusive results might be obtained from cattle exposed to RPV in
Africa, where they are kept alive much longer. Furthermore, confirmation of a high degree of
epitopic variation in the M protein as a common property of MV and CDV strains not shared by
RPV strains must await generation of anti-CDV and anti-RPV M protein MAbs.
The P proteins of the three morbilliviruses appeared to be epitopically quite distinct. Out of
six anti-MV P and 17 anti-CDV P protein MAbs, only one MAb from each panel cross-reacted
with heterologous viruses in RIPA, although by IF some additional cross-reactivity was
observed. Overall, three P protein epitopes were cross-reactive between the three morbilliviruses, of which one was found on all strains of all viruses. The other two were found on some
but not all strains of each virus, indicating the occurrence of strain-specific variation.
The NP proteins of the three morbilliviruses were more conserved than either the H, M or P
proteins, but the viruses could still be distinguished easily and a number of strain-specific
variations were detected in all three viruses. In a recent study the amino acid sequences (deduced
from the nucleotide sequences) of the NPs of MV and CDV have been compared (Rozenblatt et
al., 1985). It was seen that a stretch of about 100 amino acids at the carboxy end showed only
chance homology while the sequence of the rest of the NP of the two viruses (about 400 amino
acids) showed moderate to high homology. This observation led the authors to propose that
shared as well as unique epitopes must exist on the NPs of the two viruses. The results of the
present study confirm and extend this proposal. Amongst others, virus group-specific, virus
group-cross-reactive and virus type-specific epitopes could be delineated for the NPs of the
morbilliviruses. These epitope groupings are described in detail below.
In previous studies characterizing the MAbs using homologous virus systems it was shown
that in IF some of the anti-CDV NP MAbs stained cytoplasmic inclusions only, whereas others
stained both nuclear and cytoplasmic inclusions (Orvell et al., 1985). The anti-MV NP MAbs
characterized previously by Norrby et al. (1982) stained both nuclear and cytoplasmic
inclusions, but some of the additional anti-measles NP MAbs used in this study stained only
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
Epitopes shared by morbilliviruses
1391
cytoplasmic inclusions in the homologous virus system. In cross-reactivity IF studies with
heterologous morbilliviruses, all the anti-NP MAbs were seen to maintain their respective (i.e.
cytoplasmic only or nuclear plus cytoplasmic staining) specificities. The occurrence of
antigenically differentiable nuclear and cytoplasmic forms of NP is therefore common to CDV,
MV and RPV.
In summary, the F and NP proteins of morbilliviruses show a relatively high degree of
epitopic homology; the P and M proteins show a partial degree of epitopic homology, with the
latter showing greater variation when CDV and MV are compared; the H proteins show a very
low degree of epitopic homology, and then only between MV and RPV. The data indicate that
the epitopes on the structural proteins of the three morbilliviruses can be antigenically grouped
in relation to each other as follows (examples refer to the NP). (i) Virus group-specific epitopes,
i.e. epitopes present on all strains of all three morbillivirus types (Class A, Tables 7 and 8). (ii)
Virus group-cross-reactive epitopes, i.e. epitopes present on some strains of each of the three
morbillivirus types (Class B, Tables 7 and 8). The majority of the strain-specific variation fell
within this class. (iii) Virus type-specific epitopes, i.e. epitopes unique to MV only or CDV only
(Class D, Tables 7 and 8). (iv) CDV-RPV intertypic epitopes, i.e. epitopes recognized by antiCDV MAbs which are shared by CDV and RPV but not with any MV strain (Class C, Table 7).
(v) MV-RPV intertypic epitopes, i.e. epitopes recognized by anti-MV MAbs which are shared
by MV and RPV but not with any CDV strain (Class C, Table 8).
The MAb cross-reactivities obtained for all the viral proteins with one exception fell into the
epitopic classes defined above. The exception was the M protein which was the only protein to
exhibit a small number of MV-CDV intertypic epitopes, i.e. epitopes shared by MV and CDV
and not found on any RPV strain.
Based on the cumulative IF data, details of which are included in this report, we have
previously proposed an evolutionary relationship between MV, CDV and RPV (Norrby et al.,
1985). The present greatly extended study supports this proposal, which is that RPV is the
archevirus from which CDV and MV evolved. That the H proteins of MV and RPV showed
greater epitopic homology than the H protein of MV and CDV or CDV and RPV, was taken to
indicate that CDV branched from RPV earlier than MV and has consequently evolved a more
distinct H protein. It will be of interest to see by similar analyses where the fourth member of the
morbillivirus genus, peste des petits ruminants virus, which is thought to have arisen more
recently fits into this evolutionary scheme.
We thank A n n a Coter, Mariethe Ehnlund and David Parkinson for excellent technical assistance, Dr W. P.
Taylor for providing the seed rinderpest viruses, Drs D. McFarlin and K. W. R a m m o h a n for providing the antiEdmonston strain MAbs, and Mrs Kathrin H~iger for typing the manuscript. This work was supported by a grant
from the Swedish Medical Research Council (Project no. 885-16X-00116-21B).
REFERENCES
APPEL, M. J. G., GIBBS, E. P. J., MARTIN, S. J., TER MEULEN, V., RIMA, B. K., STEPHENSON, J. R, & TAYLOR, W. P. (1981).
Morbillivirus diseases of animals and man. In Comparative Diagnosis of Viral Diseases, vol. 4, pp. 235-297.
Edited by E. Kurstak & C. Kurstak. New York: Academic Press.
APPEL, M. J. G., SHEK, W. R., SHESHBERADARAN,H. & NORRBY, E. (1984). Measles virus and inactivated canine
distemper virus induce incomplete immunity to canine distemper. Archives of Virology 82, 73-82.
BIRRER, M. T., BLOOM,B. R. & UDEM, S. (1981). Characterization of measles polypeptides by monoclonal antibodies.
Virology 108, 381-390.
BONNER, W. M. & LASKEY,R. A. (1974). A film detection method for tritium-labelled proteins and nucleic acids in
polyacrylamide gels. European Journal of Biochemistry 46, 83-88.
GIRAUDON, P. & WILD, T. F. (198 l). Differentiation of measles virus strains and a strain of canine distemper virus by
monoclonal antibodies. Journal of General Virology 57, 179-183.
HALL, W. W., LAMB,R. A. & CHOI'PIN, P. W. (1980). The polypeptides of canine distemper virus: synthesis in infected
cells and relatedness to the polypeptides of other morbilliviruses. Virology 100, 433-449.
IMAGAWA,D. T. (1968). Relationships among measles, canine distemper and rinderpest viruses. Progressin Medical
Virology 10, 160-193.
KINGSBURY, D. W., BRATT, M. A., CHOPPIN, P. W., HANSON, R. P., HOSAKA, Y., TER MEULEN, V., NORRBY, E.,
PLOWRIGHT, W., ROTT, R. & WUNNER, W. 14. (1978). Paramyxoviridae. lntervirology 10, 137-152.
MATTrtEWS, R. E. F. (1982). Paramyxoviridae. Intervirology 17, 104-105.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28
1392
H, S H E S H B E R A D A R A N
AND OTHERS
NORRBY, E. & APPEL, M. J. G. (1980). Humoral immunity to canine distemper after immunization of dogs with
inactivated and live measles virus. Archives oJ Virology 66, 169-177.
NORRBY, E., CHEN, S.-N., TOGASHI, T., SHESHBERADARAN,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, 1 11.
NORRBY, E., SHESHBERADARAN,H., McCULLOUGH,K. C., CARPENTER, W. C. & ORVELL,C. (1985). IS rinderpest virus the
archevirus of the morbillivirus genus? Intervirology 23, 228-232.
6RVELL, C. & NORRBY, E. (1974). Further studies on the immunological relationships between measles, distemper
and rinderpest viruses. Journal of Immunology 113, 1850-1858.
6RVELL, C. & NORRBY, E. (1980). Immunological relationships between homologous polypeptides of measles and
canine distemper virus. Journal of General Virology 50, 231-245.
ORVELL, C., SHESHBERADARAN,H. & NORRBY, E. (1985). Preparation and characterization of monoclonal antibodies
directed against tbur structural components of canine distemper virus. Journal oJGeneral Virology 66, 443
456.
RIMA, B. K. ROBERTS, M. W. & MARTIN, S. J. (1983). Comparison of morbillivirus proteins by limited proteolysis.
Medical Microbiology and Immunology 171,203 213.
ROZENBLATT, S., EIZENBERG, O., BEN-LEVY, R., LAVIE, V. & BELLINI, W. J. (1985). Sequence homology within the
morbilliviruses. Journal ~)[ Virology 53, 684-690.
RUSSELL, W. C. (1983). Paramyxovirus and Morbillivirus infections and their relationship to neurological disease.
Progress in Brain Research 59, 113-132.
SATO,T. A., HAYAMI,M. & YAMANOUCHI,K. (1981). Analysis of structural proteins of measles, canine distemper, and
rinderpest viruses. Japanese Journal of Medieal Science and Biology 34, 355 364.
SATO, T. A., FUKUDA, A & SUGIURA, A. (1985). Characterization of major structural proteins of measles virus with
monoclonal antibodies. Journal of General Virology 66, 1397-1409.
SHAPSHAK, P., GRAVES, M. C. & IMAGAWA,D. T. (1982). Polypeptides of canine distemper virus strains derived from
dogs with chronic neurological diseases. Virology 122, 158-170.
SHESHBERADARAN,H. & NORRBY,E. (1984). Three monoclonal antibodies against measles virus F protein cross-react
with cellular stress proteins. Journal of Virology 52, 995-999.
SHESHBERADARAN, H. & NORRBY, E. (1985). Characterization of epitopes on the measles virus hemagglutinin.
Virology (in press).
SHESHBERADARAN,H., CHEN, S.-N. & NORRBY, E. (1983). Monoclonal antibodies against five structural components
of measles virus. I. Characterization of antigenic determinants on nine strains of measles virus. Virology 128,
341 353.
SHESHBERADARAN, H., NORRBY, E. & RAMMOHAN, K. W. (1985). Monoclonal antibodies against five structural
components of measles virus. I1. Characterization of five cell lines persistently infected with measles virus.
Archives of Virology 83, 251 268.
STEPHENSON, J. R. & TER MEULEN, V. (1979). Antigenic relationship between measles and canine distemper viruses:
comparison of i m m u n e response in animals and h u m a n s to individual virus-specific polypeptides.
Proceedings of the National Academy of Sciences, U.S.A. 76, 6601-6605.
TER MEULEN,V. & CARTER,M. J. (1982). Morbillivirus persistent infections in animals and man. In Virus Persistence,
pp. 97 132. Edited by B. W. J. Mahy, A. C. Minson & G. K. Darby. Cambridge: Cambridge University
Press.
TRUDGETT, A., GOULD, E. A., ARMSTRONG,M., MINGIOLI, E. S. & McFARLIN, D. E. (1981). Antigenic differences in the
hemagglutinin of measles and related viruses. Virology 109, 180-182.
UNDERWOOD, P. A. (1985). Practical considerations of the ability of monoclonal antibodies to detect antigenic
differences between closely related variants. Journal of Immunological Methods 85, 309 323.
WATERSON, A. P., ROTT, R. & ENDERS-RUCKLE,G. (1963). The components of measles virus and their relationship to
rinderpest and distemper. Zeitschrifi fi~r Naturjorschung 18, 377-384.
WILD, T. F. & HUPPERT, J. (1980). Specificity of measles and canine distemper virus antibodies. Annals of Virology
131, 73-84.
YEWDELL, J. W. & GERHARD, W. (1981). Antigenic characterization of viruses by monoclonal antibodies. Annual
Review of Microbiology 35, 185-206.
(Received 9 December 1985)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 05:24:28