VIROLOGY
187,472-479
(1992)
Mapping
of the Antigenic Determinants Recognized by Monoclonal against the M2 Protein of Rabies Virus
Antibodies
KAZUFUMI HIRAMATSU,* KUMATO MIFUNE,*s’ KAZUAKI MANNEN,* AKIRA NISHIZONO,* HIROSHI KAWANO,* YUJI ITO,” AND AKIHIKO KAWAI *Department
of Microbiology, Medical College of Oita, ‘Hazama-cho, Oita 879-55, Japan; and Department of Molecular Faculty of Pharmaceutical Sciences, Kyoto University, Shimoadachicho, Sakyo-ku, Kyoto 606, /apan Received September
3, 199 1; accepted
December
Microbiology,
9, 199 1
Twenty-one hybridomas producing monoclonal antibodies (moAbs) against the M2 protein of the Nishigahara (RCEH) strain of rabies virus were prepared using the SDS-polyacrylamide gel-purified M2 protein as the immunogen. All moAbs reacted with the protein after Western blotting of rabies virus. By combinations of competitive binding assays, examination of the reactivity of moAbs to the cells infected with parent RCEH and two other strains, CVS and HEP-Flury, and immunoprecipitation with in vitro translation products derived from full-length and truncated cDNAs of the M2 gene, these moAbs could be classified into seven epitope groups. Of these, 20 moAbs belonging to six epitope groups were suggested to recognize an antigenic determinant in the amino-terminal region, from the 1st to the 72nd amino acid of the protein (8 moAbs from two groups directed to amino acids 1 to 72; 2 moAbs from a group directed to amino acids 9 to 72; 5 moAbs from a group directed to amino acids 17-72; 5 moAbs from two groups directed to amino acids 32 to 72). The antigenic determinant recognized by the remaining 1 moAb was shown to be located in the amino acid region from 50 to 171. These moAbs should be useful for further studies on the biological functions of the M2 protein of rabies virus. 0 1992 Academic Press, Inc.
In contrast, little is known about the biological functions of the M2 protein of rabies virus. With vesicular stomatitis virus (VSV), a representative member of the genus Vesiculovirus of the Rhabdoviridae family, however, various studies have been undertaken to elucidate the functions of the M protein using monoclonal antibodies (moAb) and temperature-sensitive mutants which are defective in viral assembly and have lesions in the M protein gene (Clinton et al., 1978; De et al., 1982; Pal eta/,, 1985a,b; Shipley et a/., 1988; Kaptur et al., 1991). These studies revealed that the M protein plays a regulatory role in viral transcription and has an important structural role in viral assembly. A more recent study demonstrated another function of the protein involved in the cytopathogenesis of virus-infected cells (Blonde1 et al., 1990). Since moAbs are useful for the study of the analyses of antigenic properties and the functions of the protein, an attempt has been made to produce moAbs against rabies virus M2 protein. We describe here the preparation and characterization of moAbs made against the M2 protein of rabies virus and the mapping of the antigenie determinants on the protein recognized by moAbs.
INTRODUCTION Rabies is a disease of the central nervous system that is of importance to humans and animals and is caused by rabies virus which is a member of the genus Lyssavirus of the Rhabdoviridae family. Rabies virus is a bullet-shaped enveloped virus consisting of a single, negative-stranded RNA genome and five virus-coded proteins. Among these, nucleoprotein (N), phosphoprotein (Ml), and the large protein (L) are tightly associated with the genomic RNA and comprise a transcriptionally active nucleocapsid. The remaining two virus-coded proteins are membrane associated: the externally oriented transmembrane glycoprotein (G) and the peripheral matrix protein (M2). Biological and antigenic properties of the G protein have been extensively studied and it appears to be firmly established that rabies virus G protein forms spike projections on the external surface of the virus membrane, plays a important role in the establishment of viral infection (Dietzschold era/., 1978, 1987) and serves as the major antigen for the induction of virus neutralizing antibody (Wiktor et al., 1973, 1984; Cox et al., 1977). Recently, G protein has also been regarded as an important factor responsible for determining the virulence of the virus (Dietzschold et a/., 1983; Seif et al., 1985; Tuffereau et al., 1989).
MATERIALS
’ To whom reprint requests should be addressed. 0042-6822/92
$3.00
Copyright 0 1992 by Academic Press. Inc. All rights of reproduction in any form reserved.
AND METHODS
Mice Inbred BALB/c male mice, 7 to 10 weeks of age, were purchased from a commercial animal laboratory. 472
EPITOPE MAPPING
OF ANTI-RABIES
Viruses and cells The Nishigahara (RCEH) strain (Mifune et a/., 1986) of rabies virus was used for the production of moAbs. Other strains (CVS and HEP-Flury) of rabies virus propagated in baby hamster kidney (BHK-21) cells were also used for the characterization of moAbs. BHK-21 cells were grown in Eagle MEM supplemented with 10% tryptose-phosphate broth, 8% bovine serum, and antibiotics. Murine plasmocytoma cells, P3/NS-l/l-Ag4-1, were grown in Iscove’s modified Dulbecco’s MEM supplemented with 10% fetal calf serum and antibiotics. Preparation of moAbs against the M2 protein of rabies virus After SDS-polyacrylamide gel electrophoresis (PAGE) (Laemmli, 1970) of the concentrated RCEH virus in a 10% gel under unreduced conditions, a part of the gel containing the M2 protein was excised. The M2 protein was extracted from the gel slice by electroelution, followed by dialysis against deionized water for 3 days, freeze-dried for concentration, and thereafter resuspended in phosphate-buffered saline (PBS, pH 7.4). Mice were immunized subcutaneously with 100 pg of the M2 protein suspended with an equal volume (0.25 ml) of Freund’s complete adjuvant twice at 2-week intervals, followed by an additional immunization with the same dose of M2 protein by intravenous inoculation. Three days later, the mice were sacrificed and the spleens were removed. Using a singlecell suspension of the spleen and P3/NSl/l-Ag4-1 cells, hybridoma cells were prepared by essentially the same method as described elsewhere (Wiktor and Koprowski, 1978). Clones producing antibody were detected by enzyme-linked immunosorbent assay (ELISA) and indirect fluorescent antibody staining using RCEH virus-infected BHK cells. Specificity of the antibodies against the M2 protein was confirmed by immunoprecipitation with 35S-methionine-labeled viral antigens. lmmunofluorescent
staining
BHK-21 cells infected with different strains of rabies virus at a multiplicity of infection of 0.05 PFU/cell were harvested at 72 hr postinfection and fixed with cold acetone. These cells were stained with moAbs for 30 min at 37”, washed with PBS twice, and further stained with FITC-labeled anti-mouse immunoglobulin (or IgG) goat IgG (Cappel, Cochranville, PA) for 30 min at 37”. Cell-ELISA Enzyme immunoassay using RCEH virus-infected BHK-21 cells for the screening of moAbs was per-
M2 PROTEIN moAbs
1473
formed by the method described by Mannen el al. (1987). lmmunoprecipitation Ascitic fluid (50 ~1)containing moAbs was added to labeled viral antigens or in vitro translation products produced from various sizes of cDNAs of the M2 gene in immunoprecipitation buffer (0.5% Triton X-l 00, 0.5% DOC, 0.005 M NaCl in 0.025 MTris-HCI, pH 8.0; Saleh eta/., 1979). The reaction mixture was incubated for 1 hr at 4”. Immune complexes were collected on Protein A-Cellulofine (Seikagaku Corp., Tokyo, Japan) after a 30-min incubation at 4” and washed three times in washing buffer (0.1% Triton X-l 00, 0.1 M NaCI, 0.05 M Tris-HCI, pH 8.0). Pellets were resuspended in PAGE sample buffer and boiled for 5 min to elute the bound immune complex. Samples were electrophoresed in 10% SDS-polyacrylamide gels, processed for fluorography, and exposed to Kodak X-Omat RP film at -80”. Determination of the isotypes of moAbs lmmunoglobulin isotypes of each ascitic fluid were identified using the immunoglobulin identification kit supplied by Amersham (Tokyo, Japan). Competitive binding assay IgG was purified using a Protein A-Cellulofine column as described by Eyetal. (1978) and then iodinated with lodogen (Pierce Chemical Co. Rockford, IL) and Nalz51(spec. act.) loo-250 mCi/ml, Nordion International Inc., Canada) as described by Volk et a/. (1982). ELISA microtiter wells (Costar Corp., Cambridge, MA) coated with 40 pg of concentrated RCEH virions disrupted with Tween 20 (1O/o)were exposed to increasing concentrations of competing unlabeled moAb (0.05 to 25 pg) and then to 5 X 1O4cpm/well of 1251-labeledmoAb. After a 2-hr incubation at 37”, the wells were washed seven times with PBS containing 0.05% Tween 20, and the radioactivity of each well was counted separately in a Aloka gamma counter (Aloka Co. Ltd., Tokyo, Japan). Western blot analysis RCEH virions concentrated and purified by ultracentrifugation and high-performance liquid chromatography (Shimazu Corp., Kyoto, Japan) were electrophoresed in 10% SDS-polyacrylamide gel under reduced condition. The electrophoresed proteins were electro-
474
HIRAMATSU ET AL.
phoretically transferred to nitrocellulose sheets (Sartorius Corp., Tokyo, Japan). Nitrocellulose sheets were soaked in PBS containing 0.05% Tween 20 and 3% bovine serum albumin for 1 hr at room temperature, reacted with moAbs, and colored using the ABC kit (Vector Laboratories, Inc., Burlingame, CA). Construction of full-length the M2 gene
and truncated
cDNAs of
A cDNA (RC3-8) which contains a part of the preceding Ml gene and a part of the intergenic gene of the M2 and G genes was first cloned from RCEH-infected BHK-21 cells as described previously (Mannen et a/., 1991). A full-length cDNA of the M2 gene (M2-cDNA) lacking extra genes was amplified by polymerase chain reaction (PCR) using synthetic oligonucleotide sense and antisense primers described below and RC3-8 cDNA as a template. These primers (Fig. 1, Pl and P2) were synthesized to contain the Sal1 site just upstream of the authentic ATG codon or downstream of the stop TAA codon, respectively. Amplified cDNAs thus obtained were excised by SalI digestion and inserted into the Salt site of the wild-type pBluescript II SK(-) plasmid. The orientation of the insert cDNA was checked by the cutting patterns of the plasmid by appropriate restriction enzymes. The BamHl cDNA fragment was constructed as follows. The pBluescript II SK(-) plasmid containing fulllength M2-cDNA was digested with BarnHI, and a smaller size cDNA fragment was removed after electrophoresis in a 1% agarose gel. The remaining larger cDNA fragment was treated with the Klenow fragment of DNA polymerase to generate blunt-ended cDNA, followed by addition of MURFI linker (Pharmacia, Uppsala, Sweden) containing the lVhel site and the TAG stop codon, digested with IVhel, and then self-ligated. When this BamHl cDNA fragment of M2-cDNA was analyzed by nucleotide sequencing, extra sequences encoding Ala and Ser were found to be supplemented just downstream of the 72nd amino acid, Leu, and before the TAG stop codon. The 17-202 cDNA fragment was obtained by SalI digestion of PCR-amplified cDNA with sense primer 4 and antisense primer 2 and RC3-8 cDNA as a template. Primer 4 (Fig. 1, P4) was synthesized to include the SalI site just upstream of the newly made ATG start codon (this site corresponds to the 16th amino acid and Met was substituted for the authentic Thr). Therefore, after in vitro translation of this 17-202 cDNAfragment the peptide is deduced to consist of authentic amino acids 17 to 202. Other truncated cDNA fragments were constructed in a similar manner. The schematic size, the location of
those cDNAs in relation to full-length M2-cDNA, and the nucleotide sequences of synthetic oligonucleotide primers used in the experiments are shown in Fig. 1. The sequences of synthetic oligonucleotide primers were designed to have two to four mismatches in the 5’ end of the start or stop codons with respect to the template DNA. Nucleotide sequences of a part of the truncated cDNAs (around the start and stop codons) were determined by the dideoxy termination method to ensure that the sequences were as designed.
In vitro transcription
and translation
Each cDNA inserted into pBluescript II SK(-) was linearized by digestion with an appropriate restriction enzyme, followed by a 30-min incubation at 37” with added proteinase K (1 pg/50 ~1 of total volume) to remove RNase and enzymes. Template cDNA, which was precipitated by phenol-chloroform and ethanol, was resuspended in RNase-free distilled water and the RNA was synthesized in the presence of 0.8 mlVI each ATP, GTP, CTP, UTP, 40 mM Tris (pH 8.0), 50 mM NaCI, 8 mM MgCI,, 2 mM spermidine, RNasin (1 unit/ PI), and T3 or T7 DNA-dependent RNA poiymerase for 30 min at 37”. In vitro transcription products were then incubated with RNase-free DNase I (10 units) for 5 min at 37”. In vitro translations were performed in a rabbit reticulocyte system as described by the supplier (Amersham, Tokyo, Japan) in the presence of in vitro transcription product (20 pg/50 ~1 of total volume), 0.1 mM methionine(-) amino acid mixture (Amersham, Tokyo, Japan), and 1 &i/PI of 35S-protein labeling mixture (spec. act. for methionine 1 186 Ci/mmol, DuPont, NEN Research Products, Boston, MA) at 30” for 1 hr. RESULTS AND DISCUSSION Production, moAbs
isotype,
and specificity
of anti-M2
Hybridoma cells producing antibodies to the M2 protein were screened by fluorescent antibody staining and cell-ELISA of the culture fluid of hybridoma cells. After sequential cloning of hybridoma cells producing antibodies, 21 hybridoma clones were finally established. Examination for their isotypes using their culture fluid revealed that of these clones 17 moAbs are of IgGl and the remainings are of IgG2a (Table 1). Ascitic fluid produced in BALB/c mice was then loaded onto a Protein A-Cellulofine column and IgG was purified. These monoclonal IgGs were demonstrated to have a capacity to bind only to the M2 protein of rabies virus. Binding to any other protein was below the background
EPITOPE MAPPING
OF ANTI-RABIES
M2 PROTEIN moAbs
475
N2 gene
4 (A) IiTG, 1
100 I
(EcoRl) ,
200
(BarHl) ,
300
400
500 I
(Xhol) ,
6tj9
RC3-8 Pl' Full-length # l-202
cDNA
4 TM
Pl I Xhol cDNA x 1-171
H
i P3
:
(L;u)-Alder-TAG :.:.:::::::: . .. . . .. WRFl linker
Bamtil cDNA # l-72 j
EcoRl(-) cDNA # 50-202 :
P4-
(ATG) I
ATG
17-202 cDNA # 17-202
-P2 :
t-b
P5-f=-
9-202 cDNA # 9-202 P6wds-%. 32-202 CDNA # 32-202
(6)
l
**
Pl:
5'-AACA$C~C;GTCGACATGAACATTC-3'
P4:
Y-GCA$GAAG%AC&&AAAA-3'
P2:
5'-GGAT6TCG4CTGA~l-KTAGAAG-3'
P5:
5'-ACAGTCGACGTAAGATAAAW+3'
P3:
5'-TGl-KTCGAGllC$&ATA-3'
P6:
5'-CT6AGiCGAC$TGTGGCT-3'
FIG. 1. Schematic illustration of the construction and the interrelation in terms of location between the full-length and the truncated cDNAs of the M2 gene of rabies virus (A). Figures with # beneath the name of the cDNA indicate the deduced amino acid number from the N-terminus encoded by the cDNA. Figures above the box indicate the nucleotide sequence number of the gene. When in parentheses, the genetic code, the site of the restriction enzyme, and the name of the amino acid indicate, respectively, those present in the authentic gene, while those not in parentheses indicate those which were newly made. 4,3 , and r indicate Sall, EcoRI, and Xhol sites, respectively. In (B), the sequences of the synthetic oligonucleotide primers used for the construction of the cDNAs are shown. In these sequences, start and stop codons are double-underlined and the sequences for restriction enzymes are underlined. The nucleotides with an asterisk indicate the sites intended for mismatch with the template DNA.
levels in immunoprecipitation reactions (Data shown) and Western blot analysis (Fig. 2).
not
Competitive binding assay of moAbs MoAbs were next examined by competitive binding assay to determine if they recognize the same or different antigenic determinants on the M2 protein. Each IgG was iodinated with Na’251 and the test was performed to assay for the capacity of increasing amounts of unlabeled antibody to inhibit binding of another antibody labeled with Na1251to the M2 protein of disrupted virions. Figure 3 shows a typical result of such a competitive binding assay, in which increasing amounts of moAbs 1, 3, and 9 were tested for their capacity to inhibit binding of 1251-labeled moAb 9 to the M2 protein. It is obvious from the result that moAbs 3 and 9 competed more than 90% with ‘251-labeled moAb 9, however, no competition was observed between moAbs 1 and 9, suggesting that moAbs 3 and 9 share the same
or adjacent antigenic determinants and the former two moAbs and moAb 1 bind to two distinct antigenic determinants. MoAbs exhibiting more than 90% competition with each other were considered to recognize the same epitope, whereas moAbs exhibiting less than 10% competition against other moAbs were classified
TABLE 1 ANTIGENIC DETERMINANTSOF THE M2 PROTEINOF RABIESVIRUS IDENTIFIEDBY COMPETITIVEBINDINGASSAY MoAb clone No. and isotype Group 1 2 3 4 U (unclassified)
IgGl
IgG2a
1, 12, 15, 16, 17, 20 3, 4, 9, 11 7, 13, 19, 21 2 10,14
8,. 18 5
6
476
HIRAMATSU
ET AL TABLE 2 REACTIVITYOFMOABSTOTHECELLSINFECTEDWITHDIFFERENTSTRAINS OF RABIESVIRUS Reactivity of the cells infected with Group
123456 FIG. 2. Western blot analysis of the binding of moAbs to the M2 protein. Purified virions were electrophoresed in 10% SDS-polyacryamide gel. The proteins were transferred to nitrocellulose sheets and reacted with moAbs as described in the text. Lane 1, viral proteins stained with Coomassie brilliant blue; lane 2, viral proteins tested with moAb 8 from group 1; lane 3, viral proteins tested with moAb 5 from group 2; lane 4, with moAb 7 from group 3; lane 5, with moAb 2 from group 4; lane 6, with moAb 6 from the unclassified group.
into other epitope groups. MoAbs showing competition from less than 90% to more than 10% with moAbs of other groups were designated as the unclassified group in this experiment. Based on the competitive binding assay, the moAbs were classified into four distinct epitope groups, 1, 2, 3, and 4, and the clones that correspond to these groups are shown in Table 1. The
.
* 0.05 0.2
0.8
3.2 unlabeled
MAb (pg)
FIG. 3. Competitive binding to the disrupted virions of unlabeled moAbs 1 1 3 1 9, and ‘251-labeled moAb 9. Microtiter wells coated with 40 pg of concentrated RCEH virions disrupted with 1% Tween 20 were exposed to increasing concentrations of competing unlabeled moAbs and then to ‘251-labeled moAb 9 (5 X 1 O4cpm/well). After 2 hr at 37”, the wells were washed seven times with PBS containing 0.05% Tween 20. and the radioactivity of each well was separately counted in a gamma counter. 0, ‘%moAb 9 with unlabeled moAb 9; l , 1*51-moAb 9 with unlabeled moAb 3; A, ‘?moAb 9 with unlabeled moAb 1.
MoAb No.
RCEH
HEP
cvs
1-l
15, 18, 20
+
-
-
1-2 2
1, 8, 12, 16, 17 3, 4, 5, 9, 11
+ +
+
+ -
3 4 U-l u-2
7, 13, 19, 21 2 10,14 6
+ + + +
+ + + -
+ + + -
Note. Reactivity of moAbs was examined by indirect fluorescent antibody staining. + indicates that virus-infected cells were stained with moAb and - indicates negative reaction.
representative moAbs from each group and all from the unclassified group reacted with the M2 protein after Western blotting and SDS-polyacrylamide gel electrophoresis of the virions under reduced conditions (Fig. 2). Reactivity of moAbs against the ceils infected different strains of rabies virus
with
To further confirm the group classification of moAbs by competitive binding assay and to detect the possible existence of other determinants on the M2 protein, reactivity of moAbs to the cells infected with the CVS and HEP-Flury strains in addition to the parent RCEH virus was examined by indirect fluorescent antibody staining. As shown in Table 2, reactivity patterns were considerably variable from group to group. MoAbs from group 1 did not bind to the cells infected with the HEP-Fluty strain and were classified into two subgroups based on their reactivity to CVS-infected cells. MoAbs of group 2 reacted with RCEH- and HEP-infected cells but not with the cells infected with CVS. MoAbs in groups 3 and 4 reacted with all strains of rabies virus examined. Also, three moAbs in the unclassified group were classified into two groups by their reactivity patterns to these virus strains. These results might suggest that the moAbs established in this study consist of moAbs recognizing seven different antigenic determinants on the M2 protein. Mapping of the antigenic determinants by means of an immunoprecipitation test with in vitro translation products To determine the antigenic determinants recognized by moAbs, several truncated cDNAs of the M2 gene
EPITOPE MAPPING
A
-.. -15
12345678
moAb --10-+-a-5
OF ANTI-RABIES
no. -
123456
FIG. 4. Mapping of the antigenic determinants of moAbs by immunoprecipitation with in vitro translation products of truncated cDNAs of the M2 gene. After reaction of in vitro translation products of cDNAs with moAbs for 1 hr at 4”, immune complexes were collected on Protein A-Cellulofine by a 30.min incubation at 4”. Pellets resuspended in PAGE sample buffer were boiled for 5 min to elute the immune complexes. Samples were electrophoresed in 10% SDS-polyacrylamide gels and the gels were processed forfluorography. (A) Lane 1, 35S-labeled virions; lanes 2, 3, 4, moAb 15 (group l-l) tested with the products of full-length, BarnHI, and 9-202 cDNAs, respectively; lanes 5,6, moAb 10 (group U-1) tested with the products of the 9-202 cDNA and the 17-202 cDNAs, respectively; lanes 7, 8, moAb 5 (group 2) tested with the products of the 17-202 cDNA and the 32-202 cDNA, respectively. (B) Lane 1, ?Yabeled virions; lanes 2, 3, moAb 7 (group 3) tested with the products of the 32-202 and EcoRl (-) cDNAs, respectively; lanes 4, 5, 6, moAb 2 (group 4) tested with the products of the EcoRl (-), BarnHI, and Xhol cDNAs, respectively. G, N, and M2 indicate the glycoprotein, nucleoprotein, and matrix protein, respectively, of rabies virus.
were constructed as described under Materials and Methods and immunoprecipitation tests were performed using in vitro translation products from the truncated cDNAs. Typical results and the summarized results are shown in Fig. 4 and Table 3, respectively. All of the representative clones from seven different groups precipitated an in vitro translation product from full-length cDNA as shown in Fig. 4A, lane 2, and the product migrated in a gel to the same position as that of the authentic M2 protein from rabiesvirions (Fig. 4A, lane 1). MoAb (No. 15) from group 1-l precipitated an in vitro translation product from the BarnHI cDNA fragment (Fig. 4A, lane 3) however, it did not react with that from the 9-202 cDNA fragment (Fig. 4A, lane 4). MoAb (No. 10) from group U-l and moAb (No. 5) from group 2 precipitated the translation products derived from the 9-202 cDNA fragment (Fig. 4A, lane 5) and from the BarnHI cDNA fragment (not shown in figure) but did not precipitate the product of the 32-202 cDNA fragment (Fig. 4A, lane 8 for moAb no. 5). However, the latter moAb reacted with the product of the 17-202 cDNA fragment (Fig. 4A, lane 7) whereas the former did not (Fig. 4A, lane 6). MoAb (No. 7) from group 3
M2 PROTEIN moAbs
477
precipitated the product by the 32-202 cDNA fragment (Fig. 4B, lane 2) however, it did not precipitate a product from the EcoRl (-) cDNA fragment (Fig. 4B, lane 3). Finally, moAb (No. 2) of group 4 precipitated the products from the EcoRl (-) cDNA fragment (Fig. 4B, lane 4) and the Xhol cDNA fragment (Fig. 4B, lane 6) but did not precipitate a product from the BarnHI cDNA fragment (Fig. 4B, lane 5). Attention should be paid in the interpretation of the location of antigenic determinants on the basis of differential reactivity with overlapping peptides derived from truncated cDNAs. If a peptide is shortened, its configuration and consequently its reactivity with a given moAbs might change. Therefore, only positive results will be meaningful and negative results should not be taken. Additionally, if there are many positive results with the given overlapping peptides, the minimal sequences that still show reactivity should be taken. On the basis of this principle, an antigenic determinant of moAbs from group 1 is suggested to be located in a region between amino acid 1 and 72, group 2 from 17 to 72, group 3 from 32 to 72, group 4 from 50 to 171, group U-l from 9 to 72, and group U-2 from 32 to 72. It would be possible to speculate more defined locations of these antigenic determinants if we could take the negative results, for instance, an antigenic determinant of moAbs from group 1 is located between amino acid residues 1 and 8, because the moAbs reacted with peptide 1-72 and did not react with peptide 9-202. However, one should provide direct evidence that the moAb certainly binds to synthetic oligopeptide l-8. MoAbs obtained were initially classified into four distinct epitope groups and an unclassified group by competitive binding assay and further divided into seven groups by examining their reactivity against the cells infected with the other strains of rabies virus. Furthermore, it was possible to map the antigenic determinants by immunoprecipitation reaction with in vitro translation products derived from truncated cDNAs of the M2 gene. By these analyses, 20 moAbs belonging to six groups have been suggested to recognize a determinant in the region of amino acids 1 to 72 from the N-terminus. This might reflect the finding that the Nterminal region of the M2 protein of rabies virus is essentially hydrophilic when analyzed from the deduced amino acid sequences (Rayssiguier et a/., 1986; Hiramatsu et a/., unpublished data). The results of several studies with VSV have pointed to the region of up to the 43rd amino acid from the N-terminus of the M protein as being responsible for transcription inhibition activity (Pal et a/., 1985b; Og-
478
HIRAMATSU
ET AL.
TABLE 3 CAPABILITYOF MOABS TO PRECIPITATEIN VITROTRANSLATIONPRODUCTSBYTRUNCATEDcDNAs OF THE M2 GENE With the product from truncated
Group
MoAb No.
Full-length (l-202)
Xho 1 (1-171)
BamHl (l-72)
EcoRl (-) (50-202)
l-2 1-2 2 3 4 U-l U-2
15*, 18, 20 1, 8*, 12, 16, 17 3, 4, 5*, 9, 11 7*, 13, 19, 21 2* 10*, 14 6*
+ + + + + + +
nd nd nd nd + nd nd
+ + + + -
nd -
+ +
+ -
cDNA fragment of 32-202 (32-202)
17-202 (17-202)
9-202) (g-202)
-
-
+ + -
+ + + -
+
+
+ nd nd + nd
nd nd -
Location of antigenic determinant indicated by amino acid No. from N-terminus l-72 l-72 17-72 32-72 50-171 9-72 32-72
Note. Asterisks indicate the number of moAbs used in the experiments as a representative clone of the group. The numbers in parentheses under the name of the truncated cDNA fragment indicate the number of amino acids of the peptides encoded by the cDNAs. + indicates that the moAb reacted with a peptide. nd; not tested.
den el al., 1986; Kaptur et al., 1991). To our knowledge, only one moAb (RM 3-9-l 6) against the M2 protein of rabies virus (HEP strain) has been prepared so far (Honda and Kawai, unpublished data). This moAb has been shown to exhibit almost the same reactivity pattern with our overlapping truncated peptides as that of moAbs from the U-l group, suggesting that the antigenie determinant lies between amino acids 9 and 72. MoAbs obtained in this study represent the largest panel of moAbs against the M2 protein and should be useful for further studies on the possible regulatory function of viral transcription and other functions of the M2 protein of rabies virus. ACKNOWLEDGMENTS This work was supported in part by research grants from the Japan-united States Cooperative Medical Science Program and from Yakult-Honsha Co. Ltd, Tokyo, Japan.
REFERENCES BLONDEL, D., HARMISON, G. G., and SCHUBERT, M. (1990). Role of matrix protein in cytopathogenesis of vesicular stomatitis virus. /. Viral. 64, 1716-1725. CLINTON, G. M., LITTLE, S. P., HAGEN, F. S., and HUANG, A. S. (1978). The matrix (M) protein of vesicular stomatitis virus regulates transcription. Cell 15, 1455-l 462. Cox, J. H., DIETZSCHOLD, B., and SCHNEIDER, L. G. (1977). Rabies virus glycoprotein. II. Biological and serological characterization. Infect. Immun. 16, 754-759. DE, B. P., THORNTON, G. B., LUK, D., and BANERJEE,A. K. (1982). Purified matrix protein of vesicular stomatitis virus blocks viral transcription in vitro. Proc. Natl. Acad. Sci. USA 79, 7 137-7 141. DIETZSCHOLD, B., Cox, J. H., SCHNEIDER, L. G., WIKTOR, T. J., and KOPROWSKI,H. (1978). Isolation and purification of a polymeric form of the glycoprotein of rabies virus. J. Gen. viral. 40, 131-l 39. DIETZSCHOLD, B., TOLLIS, M., LAFON, M., WUNNER, W. H., and KOPROWSKI,H. (1987). Mechanisms of rabies virus neutralization by
glycoprotein-specific monoclonal antibodies. virology 161, 2936. DIETZSCHOLD,B., WUNNER, W. H., WIKTOR, T. J., LOPES,A. D., LAFON, M., SMITH, C. L., and KOPROWSKI,H. (1983). Characterization of an antigenic determinant of the glycoprotein that correlates with pathogenicity of rabies virus. Proc. Nat/. Acad. Sci. USA 80, 7074. EY, P. L., PROWSE,S. J., and JENKIN, C. R. (1978). Isolation of pure IgGl , IgG2a and IgG2b immunoglobulins from mouse serum using protein A-Sepharose. immunochemistry 15, 429-436. KAPTUR, P. E., RHODES, R. B., and LYLES, D. S. (1991). Sequences of the vesicular stomatitis virus matrix protein involved in binding to nucleocapsids. 1. Viral. 65, 1057-l 065. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-685. MANNEN, K., HIRAMATSU, K., MIFUNE, K., and SAKAMOTO, S. (1991). Conserved nucleotide sequence of rabies virus cDNA encoding the nucleoprotein. Virus Genes 5, 69-73. MANNEN, K., MIFUNE, K., REID-SANDEN, F. L., SMITH, J. S., YAGER, P. A., SUMNER, J. W., FISHBEIN,D. B., TONG, T. C., and BAER, G. M. (1987). Microneutralization test for rabiesvirus based on an enzyme immunoassay. J. Clin. Microbial. 25, 2440-2442. MIFUNE, K., MANNEN, K., MINAMOTO, N., and ARAI, Y. T. (1986). Further studies on an improved haemagglutination inhibition test with higher sensitivity for rabies virus antibody. Bulletin W. H. 0. 64, 133-137. OGDEN, J. R., PAL, R., and WAGNER, R. R. (1986). Mapping regions of the matrix proteins of vesicular stomatitis virus which bind to ribonucleocapsids, liposomes, and monoclonal antibodies. J. Viral. 58,860-868. PAL, R., GRINNELL, B. W., SNYDER, R. M., WIENER, J. R., VOLK, W. A., and WAGNER, R. R. (1985a). Monoclonal antibodies to the M protein of vesicular stomatitis virus (Indiana serotype) and to a cDNA M gene expression product. /. Vifol. 55, 298-306. PAL, R., GRINNELL, B. W., SNYDER, R. M., and WAGNER, R. R. (198513). Regulation of viral transcription by the matrix protein of vesicular stomatitis virus probed by monoclonal antibodies and temperature-sensitive mutants. /. Viral. 56, 386-394. RAYSSIGUIER,C., CIOE, L., WITHERS, E., WUNNER, W. H., and CURTIS, P. J. (1986). Cloning of rabies virus matrix protein mRNA and determination of its amino acid sequence. Virus Res. 5, 177-l 90.
EPITOPE MAPPING
OF ANTI-RABIES
SALEH, F., GARD, G. P., and COMPANS, R. W. (1979). Synthesis of tacaribe viral proteins. Virology 93, 369-376. SEIF, I., COULON, P., ROLLIN, P. E., and FLAMAND, A. (1985). Rabies virulence: Effect on pathogenicity and sequence characterization of rabies virus mutations affecting antigenic site Ill of the glycoprotein. J. Viral. 53, 926-934. SHIPLEY,J. B., PAL, R., and WAGNER, R. R. (1988). Antigenicity, function, and cornformation of synthetic oligopeptides corresponding to amino-terminal sequences of wild-type and mutant matrix proteins of vesicular stomatitis virus. 1. Viol. 62, 2569-2577. TUFFEREAU,C., LEBLOIS, H., B~N~JEAN, J., COULON, P., LAFAY, F., and FLAMAND, A. (1989). Arginine or lysine in position 333 of ERA and CVS glycoprotein is necessary for rabies virulence in adult mice. virology 172, 206-212. VOLK, W. A., SNYDER, R. M., BENJAMIN, D. C., and WAGNER, R. R.
M2 PROTEIN moAbs
479
(1982). Monoclonal antibodies to the glycoprotein of vesicular stomatitis virus: Comparative neutralizing activity. 1. Viral. 42, 220227. WIKTOR, T. J., GYORGY, E., SCHLUMBERGER,H. D., SOKOL, F., and KoPROWSKI,H. (1973). Antigenic proper-ties of rabies virus components. J. Immunol. 110, 269-276. WIKTOR, T. J., and KOPROWSKI,H. (1978). Monoclonal antibodies against rabies virus produced by somatic cell hybridization: Detection of antigenic variants. Proc. Nat/. Acad. Sci, USA 75, 39383942. WIKTOR, T. J., MACFARLAN, R. I., REAGAN, K. .I., DIETZSCHOLD,B., CURTIS, P. J., WUNNER, W. H., KIENY, M.-P., LATHE, R., LECOCQ, J.-P., MACKETT, M., Moss, B., and KOPROWSKI,H. (1984). Protection from rabies by a vaccinia virus recombinant containing the rabies glycoprotein gene. Proc. Nat/. Acad. Sci. USA 81, 7194-7 198.
187,472-479
(1992)
Mapping
of the Antigenic Determinants Recognized by Monoclonal against the M2 Protein of Rabies Virus
Antibodies
KAZUFUMI HIRAMATSU,* KUMATO MIFUNE,*s’ KAZUAKI MANNEN,* AKIRA NISHIZONO,* HIROSHI KAWANO,* YUJI ITO,” AND AKIHIKO KAWAI *Department
of Microbiology, Medical College of Oita, ‘Hazama-cho, Oita 879-55, Japan; and Department of Molecular Faculty of Pharmaceutical Sciences, Kyoto University, Shimoadachicho, Sakyo-ku, Kyoto 606, /apan Received September
3, 199 1; accepted
December
Microbiology,
9, 199 1
Twenty-one hybridomas producing monoclonal antibodies (moAbs) against the M2 protein of the Nishigahara (RCEH) strain of rabies virus were prepared using the SDS-polyacrylamide gel-purified M2 protein as the immunogen. All moAbs reacted with the protein after Western blotting of rabies virus. By combinations of competitive binding assays, examination of the reactivity of moAbs to the cells infected with parent RCEH and two other strains, CVS and HEP-Flury, and immunoprecipitation with in vitro translation products derived from full-length and truncated cDNAs of the M2 gene, these moAbs could be classified into seven epitope groups. Of these, 20 moAbs belonging to six epitope groups were suggested to recognize an antigenic determinant in the amino-terminal region, from the 1st to the 72nd amino acid of the protein (8 moAbs from two groups directed to amino acids 1 to 72; 2 moAbs from a group directed to amino acids 9 to 72; 5 moAbs from a group directed to amino acids 17-72; 5 moAbs from two groups directed to amino acids 32 to 72). The antigenic determinant recognized by the remaining 1 moAb was shown to be located in the amino acid region from 50 to 171. These moAbs should be useful for further studies on the biological functions of the M2 protein of rabies virus. 0 1992 Academic Press, Inc.
In contrast, little is known about the biological functions of the M2 protein of rabies virus. With vesicular stomatitis virus (VSV), a representative member of the genus Vesiculovirus of the Rhabdoviridae family, however, various studies have been undertaken to elucidate the functions of the M protein using monoclonal antibodies (moAb) and temperature-sensitive mutants which are defective in viral assembly and have lesions in the M protein gene (Clinton et al., 1978; De et al., 1982; Pal eta/,, 1985a,b; Shipley et a/., 1988; Kaptur et al., 1991). These studies revealed that the M protein plays a regulatory role in viral transcription and has an important structural role in viral assembly. A more recent study demonstrated another function of the protein involved in the cytopathogenesis of virus-infected cells (Blonde1 et al., 1990). Since moAbs are useful for the study of the analyses of antigenic properties and the functions of the protein, an attempt has been made to produce moAbs against rabies virus M2 protein. We describe here the preparation and characterization of moAbs made against the M2 protein of rabies virus and the mapping of the antigenie determinants on the protein recognized by moAbs.
INTRODUCTION Rabies is a disease of the central nervous system that is of importance to humans and animals and is caused by rabies virus which is a member of the genus Lyssavirus of the Rhabdoviridae family. Rabies virus is a bullet-shaped enveloped virus consisting of a single, negative-stranded RNA genome and five virus-coded proteins. Among these, nucleoprotein (N), phosphoprotein (Ml), and the large protein (L) are tightly associated with the genomic RNA and comprise a transcriptionally active nucleocapsid. The remaining two virus-coded proteins are membrane associated: the externally oriented transmembrane glycoprotein (G) and the peripheral matrix protein (M2). Biological and antigenic properties of the G protein have been extensively studied and it appears to be firmly established that rabies virus G protein forms spike projections on the external surface of the virus membrane, plays a important role in the establishment of viral infection (Dietzschold era/., 1978, 1987) and serves as the major antigen for the induction of virus neutralizing antibody (Wiktor et al., 1973, 1984; Cox et al., 1977). Recently, G protein has also been regarded as an important factor responsible for determining the virulence of the virus (Dietzschold et a/., 1983; Seif et al., 1985; Tuffereau et al., 1989).
MATERIALS
’ To whom reprint requests should be addressed. 0042-6822/92
$3.00
Copyright 0 1992 by Academic Press. Inc. All rights of reproduction in any form reserved.
AND METHODS
Mice Inbred BALB/c male mice, 7 to 10 weeks of age, were purchased from a commercial animal laboratory. 472
EPITOPE MAPPING
OF ANTI-RABIES
Viruses and cells The Nishigahara (RCEH) strain (Mifune et a/., 1986) of rabies virus was used for the production of moAbs. Other strains (CVS and HEP-Flury) of rabies virus propagated in baby hamster kidney (BHK-21) cells were also used for the characterization of moAbs. BHK-21 cells were grown in Eagle MEM supplemented with 10% tryptose-phosphate broth, 8% bovine serum, and antibiotics. Murine plasmocytoma cells, P3/NS-l/l-Ag4-1, were grown in Iscove’s modified Dulbecco’s MEM supplemented with 10% fetal calf serum and antibiotics. Preparation of moAbs against the M2 protein of rabies virus After SDS-polyacrylamide gel electrophoresis (PAGE) (Laemmli, 1970) of the concentrated RCEH virus in a 10% gel under unreduced conditions, a part of the gel containing the M2 protein was excised. The M2 protein was extracted from the gel slice by electroelution, followed by dialysis against deionized water for 3 days, freeze-dried for concentration, and thereafter resuspended in phosphate-buffered saline (PBS, pH 7.4). Mice were immunized subcutaneously with 100 pg of the M2 protein suspended with an equal volume (0.25 ml) of Freund’s complete adjuvant twice at 2-week intervals, followed by an additional immunization with the same dose of M2 protein by intravenous inoculation. Three days later, the mice were sacrificed and the spleens were removed. Using a singlecell suspension of the spleen and P3/NSl/l-Ag4-1 cells, hybridoma cells were prepared by essentially the same method as described elsewhere (Wiktor and Koprowski, 1978). Clones producing antibody were detected by enzyme-linked immunosorbent assay (ELISA) and indirect fluorescent antibody staining using RCEH virus-infected BHK cells. Specificity of the antibodies against the M2 protein was confirmed by immunoprecipitation with 35S-methionine-labeled viral antigens. lmmunofluorescent
staining
BHK-21 cells infected with different strains of rabies virus at a multiplicity of infection of 0.05 PFU/cell were harvested at 72 hr postinfection and fixed with cold acetone. These cells were stained with moAbs for 30 min at 37”, washed with PBS twice, and further stained with FITC-labeled anti-mouse immunoglobulin (or IgG) goat IgG (Cappel, Cochranville, PA) for 30 min at 37”. Cell-ELISA Enzyme immunoassay using RCEH virus-infected BHK-21 cells for the screening of moAbs was per-
M2 PROTEIN moAbs
1473
formed by the method described by Mannen el al. (1987). lmmunoprecipitation Ascitic fluid (50 ~1)containing moAbs was added to labeled viral antigens or in vitro translation products produced from various sizes of cDNAs of the M2 gene in immunoprecipitation buffer (0.5% Triton X-l 00, 0.5% DOC, 0.005 M NaCl in 0.025 MTris-HCI, pH 8.0; Saleh eta/., 1979). The reaction mixture was incubated for 1 hr at 4”. Immune complexes were collected on Protein A-Cellulofine (Seikagaku Corp., Tokyo, Japan) after a 30-min incubation at 4” and washed three times in washing buffer (0.1% Triton X-l 00, 0.1 M NaCI, 0.05 M Tris-HCI, pH 8.0). Pellets were resuspended in PAGE sample buffer and boiled for 5 min to elute the bound immune complex. Samples were electrophoresed in 10% SDS-polyacrylamide gels, processed for fluorography, and exposed to Kodak X-Omat RP film at -80”. Determination of the isotypes of moAbs lmmunoglobulin isotypes of each ascitic fluid were identified using the immunoglobulin identification kit supplied by Amersham (Tokyo, Japan). Competitive binding assay IgG was purified using a Protein A-Cellulofine column as described by Eyetal. (1978) and then iodinated with lodogen (Pierce Chemical Co. Rockford, IL) and Nalz51(spec. act.) loo-250 mCi/ml, Nordion International Inc., Canada) as described by Volk et a/. (1982). ELISA microtiter wells (Costar Corp., Cambridge, MA) coated with 40 pg of concentrated RCEH virions disrupted with Tween 20 (1O/o)were exposed to increasing concentrations of competing unlabeled moAb (0.05 to 25 pg) and then to 5 X 1O4cpm/well of 1251-labeledmoAb. After a 2-hr incubation at 37”, the wells were washed seven times with PBS containing 0.05% Tween 20, and the radioactivity of each well was counted separately in a Aloka gamma counter (Aloka Co. Ltd., Tokyo, Japan). Western blot analysis RCEH virions concentrated and purified by ultracentrifugation and high-performance liquid chromatography (Shimazu Corp., Kyoto, Japan) were electrophoresed in 10% SDS-polyacrylamide gel under reduced condition. The electrophoresed proteins were electro-
474
HIRAMATSU ET AL.
phoretically transferred to nitrocellulose sheets (Sartorius Corp., Tokyo, Japan). Nitrocellulose sheets were soaked in PBS containing 0.05% Tween 20 and 3% bovine serum albumin for 1 hr at room temperature, reacted with moAbs, and colored using the ABC kit (Vector Laboratories, Inc., Burlingame, CA). Construction of full-length the M2 gene
and truncated
cDNAs of
A cDNA (RC3-8) which contains a part of the preceding Ml gene and a part of the intergenic gene of the M2 and G genes was first cloned from RCEH-infected BHK-21 cells as described previously (Mannen et a/., 1991). A full-length cDNA of the M2 gene (M2-cDNA) lacking extra genes was amplified by polymerase chain reaction (PCR) using synthetic oligonucleotide sense and antisense primers described below and RC3-8 cDNA as a template. These primers (Fig. 1, Pl and P2) were synthesized to contain the Sal1 site just upstream of the authentic ATG codon or downstream of the stop TAA codon, respectively. Amplified cDNAs thus obtained were excised by SalI digestion and inserted into the Salt site of the wild-type pBluescript II SK(-) plasmid. The orientation of the insert cDNA was checked by the cutting patterns of the plasmid by appropriate restriction enzymes. The BamHl cDNA fragment was constructed as follows. The pBluescript II SK(-) plasmid containing fulllength M2-cDNA was digested with BarnHI, and a smaller size cDNA fragment was removed after electrophoresis in a 1% agarose gel. The remaining larger cDNA fragment was treated with the Klenow fragment of DNA polymerase to generate blunt-ended cDNA, followed by addition of MURFI linker (Pharmacia, Uppsala, Sweden) containing the lVhel site and the TAG stop codon, digested with IVhel, and then self-ligated. When this BamHl cDNA fragment of M2-cDNA was analyzed by nucleotide sequencing, extra sequences encoding Ala and Ser were found to be supplemented just downstream of the 72nd amino acid, Leu, and before the TAG stop codon. The 17-202 cDNA fragment was obtained by SalI digestion of PCR-amplified cDNA with sense primer 4 and antisense primer 2 and RC3-8 cDNA as a template. Primer 4 (Fig. 1, P4) was synthesized to include the SalI site just upstream of the newly made ATG start codon (this site corresponds to the 16th amino acid and Met was substituted for the authentic Thr). Therefore, after in vitro translation of this 17-202 cDNAfragment the peptide is deduced to consist of authentic amino acids 17 to 202. Other truncated cDNA fragments were constructed in a similar manner. The schematic size, the location of
those cDNAs in relation to full-length M2-cDNA, and the nucleotide sequences of synthetic oligonucleotide primers used in the experiments are shown in Fig. 1. The sequences of synthetic oligonucleotide primers were designed to have two to four mismatches in the 5’ end of the start or stop codons with respect to the template DNA. Nucleotide sequences of a part of the truncated cDNAs (around the start and stop codons) were determined by the dideoxy termination method to ensure that the sequences were as designed.
In vitro transcription
and translation
Each cDNA inserted into pBluescript II SK(-) was linearized by digestion with an appropriate restriction enzyme, followed by a 30-min incubation at 37” with added proteinase K (1 pg/50 ~1 of total volume) to remove RNase and enzymes. Template cDNA, which was precipitated by phenol-chloroform and ethanol, was resuspended in RNase-free distilled water and the RNA was synthesized in the presence of 0.8 mlVI each ATP, GTP, CTP, UTP, 40 mM Tris (pH 8.0), 50 mM NaCI, 8 mM MgCI,, 2 mM spermidine, RNasin (1 unit/ PI), and T3 or T7 DNA-dependent RNA poiymerase for 30 min at 37”. In vitro transcription products were then incubated with RNase-free DNase I (10 units) for 5 min at 37”. In vitro translations were performed in a rabbit reticulocyte system as described by the supplier (Amersham, Tokyo, Japan) in the presence of in vitro transcription product (20 pg/50 ~1 of total volume), 0.1 mM methionine(-) amino acid mixture (Amersham, Tokyo, Japan), and 1 &i/PI of 35S-protein labeling mixture (spec. act. for methionine 1 186 Ci/mmol, DuPont, NEN Research Products, Boston, MA) at 30” for 1 hr. RESULTS AND DISCUSSION Production, moAbs
isotype,
and specificity
of anti-M2
Hybridoma cells producing antibodies to the M2 protein were screened by fluorescent antibody staining and cell-ELISA of the culture fluid of hybridoma cells. After sequential cloning of hybridoma cells producing antibodies, 21 hybridoma clones were finally established. Examination for their isotypes using their culture fluid revealed that of these clones 17 moAbs are of IgGl and the remainings are of IgG2a (Table 1). Ascitic fluid produced in BALB/c mice was then loaded onto a Protein A-Cellulofine column and IgG was purified. These monoclonal IgGs were demonstrated to have a capacity to bind only to the M2 protein of rabies virus. Binding to any other protein was below the background
EPITOPE MAPPING
OF ANTI-RABIES
M2 PROTEIN moAbs
475
N2 gene
4 (A) IiTG, 1
100 I
(EcoRl) ,
200
(BarHl) ,
300
400
500 I
(Xhol) ,
6tj9
RC3-8 Pl' Full-length # l-202
cDNA
4 TM
Pl I Xhol cDNA x 1-171
H
i P3
:
(L;u)-Alder-TAG :.:.:::::::: . .. . . .. WRFl linker
Bamtil cDNA # l-72 j
EcoRl(-) cDNA # 50-202 :
P4-
(ATG) I
ATG
17-202 cDNA # 17-202
-P2 :
t-b
P5-f=-
9-202 cDNA # 9-202 P6wds-%. 32-202 CDNA # 32-202
(6)
l
**
Pl:
5'-AACA$C~C;GTCGACATGAACATTC-3'
P4:
Y-GCA$GAAG%AC&&AAAA-3'
P2:
5'-GGAT6TCG4CTGA~l-KTAGAAG-3'
P5:
5'-ACAGTCGACGTAAGATAAAW+3'
P3:
5'-TGl-KTCGAGllC$&ATA-3'
P6:
5'-CT6AGiCGAC$TGTGGCT-3'
FIG. 1. Schematic illustration of the construction and the interrelation in terms of location between the full-length and the truncated cDNAs of the M2 gene of rabies virus (A). Figures with # beneath the name of the cDNA indicate the deduced amino acid number from the N-terminus encoded by the cDNA. Figures above the box indicate the nucleotide sequence number of the gene. When in parentheses, the genetic code, the site of the restriction enzyme, and the name of the amino acid indicate, respectively, those present in the authentic gene, while those not in parentheses indicate those which were newly made. 4,3 , and r indicate Sall, EcoRI, and Xhol sites, respectively. In (B), the sequences of the synthetic oligonucleotide primers used for the construction of the cDNAs are shown. In these sequences, start and stop codons are double-underlined and the sequences for restriction enzymes are underlined. The nucleotides with an asterisk indicate the sites intended for mismatch with the template DNA.
levels in immunoprecipitation reactions (Data shown) and Western blot analysis (Fig. 2).
not
Competitive binding assay of moAbs MoAbs were next examined by competitive binding assay to determine if they recognize the same or different antigenic determinants on the M2 protein. Each IgG was iodinated with Na’251 and the test was performed to assay for the capacity of increasing amounts of unlabeled antibody to inhibit binding of another antibody labeled with Na1251to the M2 protein of disrupted virions. Figure 3 shows a typical result of such a competitive binding assay, in which increasing amounts of moAbs 1, 3, and 9 were tested for their capacity to inhibit binding of 1251-labeled moAb 9 to the M2 protein. It is obvious from the result that moAbs 3 and 9 competed more than 90% with ‘251-labeled moAb 9, however, no competition was observed between moAbs 1 and 9, suggesting that moAbs 3 and 9 share the same
or adjacent antigenic determinants and the former two moAbs and moAb 1 bind to two distinct antigenic determinants. MoAbs exhibiting more than 90% competition with each other were considered to recognize the same epitope, whereas moAbs exhibiting less than 10% competition against other moAbs were classified
TABLE 1 ANTIGENIC DETERMINANTSOF THE M2 PROTEINOF RABIESVIRUS IDENTIFIEDBY COMPETITIVEBINDINGASSAY MoAb clone No. and isotype Group 1 2 3 4 U (unclassified)
IgGl
IgG2a
1, 12, 15, 16, 17, 20 3, 4, 9, 11 7, 13, 19, 21 2 10,14
8,. 18 5
6
476
HIRAMATSU
ET AL TABLE 2 REACTIVITYOFMOABSTOTHECELLSINFECTEDWITHDIFFERENTSTRAINS OF RABIESVIRUS Reactivity of the cells infected with Group
123456 FIG. 2. Western blot analysis of the binding of moAbs to the M2 protein. Purified virions were electrophoresed in 10% SDS-polyacryamide gel. The proteins were transferred to nitrocellulose sheets and reacted with moAbs as described in the text. Lane 1, viral proteins stained with Coomassie brilliant blue; lane 2, viral proteins tested with moAb 8 from group 1; lane 3, viral proteins tested with moAb 5 from group 2; lane 4, with moAb 7 from group 3; lane 5, with moAb 2 from group 4; lane 6, with moAb 6 from the unclassified group.
into other epitope groups. MoAbs showing competition from less than 90% to more than 10% with moAbs of other groups were designated as the unclassified group in this experiment. Based on the competitive binding assay, the moAbs were classified into four distinct epitope groups, 1, 2, 3, and 4, and the clones that correspond to these groups are shown in Table 1. The
.
* 0.05 0.2
0.8
3.2 unlabeled
MAb (pg)
FIG. 3. Competitive binding to the disrupted virions of unlabeled moAbs 1 1 3 1 9, and ‘251-labeled moAb 9. Microtiter wells coated with 40 pg of concentrated RCEH virions disrupted with 1% Tween 20 were exposed to increasing concentrations of competing unlabeled moAbs and then to ‘251-labeled moAb 9 (5 X 1 O4cpm/well). After 2 hr at 37”, the wells were washed seven times with PBS containing 0.05% Tween 20. and the radioactivity of each well was separately counted in a gamma counter. 0, ‘%moAb 9 with unlabeled moAb 9; l , 1*51-moAb 9 with unlabeled moAb 3; A, ‘?moAb 9 with unlabeled moAb 1.
MoAb No.
RCEH
HEP
cvs
1-l
15, 18, 20
+
-
-
1-2 2
1, 8, 12, 16, 17 3, 4, 5, 9, 11
+ +
+
+ -
3 4 U-l u-2
7, 13, 19, 21 2 10,14 6
+ + + +
+ + + -
+ + + -
Note. Reactivity of moAbs was examined by indirect fluorescent antibody staining. + indicates that virus-infected cells were stained with moAb and - indicates negative reaction.
representative moAbs from each group and all from the unclassified group reacted with the M2 protein after Western blotting and SDS-polyacrylamide gel electrophoresis of the virions under reduced conditions (Fig. 2). Reactivity of moAbs against the ceils infected different strains of rabies virus
with
To further confirm the group classification of moAbs by competitive binding assay and to detect the possible existence of other determinants on the M2 protein, reactivity of moAbs to the cells infected with the CVS and HEP-Flury strains in addition to the parent RCEH virus was examined by indirect fluorescent antibody staining. As shown in Table 2, reactivity patterns were considerably variable from group to group. MoAbs from group 1 did not bind to the cells infected with the HEP-Fluty strain and were classified into two subgroups based on their reactivity to CVS-infected cells. MoAbs of group 2 reacted with RCEH- and HEP-infected cells but not with the cells infected with CVS. MoAbs in groups 3 and 4 reacted with all strains of rabies virus examined. Also, three moAbs in the unclassified group were classified into two groups by their reactivity patterns to these virus strains. These results might suggest that the moAbs established in this study consist of moAbs recognizing seven different antigenic determinants on the M2 protein. Mapping of the antigenic determinants by means of an immunoprecipitation test with in vitro translation products To determine the antigenic determinants recognized by moAbs, several truncated cDNAs of the M2 gene
EPITOPE MAPPING
A
-.. -15
12345678
moAb --10-+-a-5
OF ANTI-RABIES
no. -
123456
FIG. 4. Mapping of the antigenic determinants of moAbs by immunoprecipitation with in vitro translation products of truncated cDNAs of the M2 gene. After reaction of in vitro translation products of cDNAs with moAbs for 1 hr at 4”, immune complexes were collected on Protein A-Cellulofine by a 30.min incubation at 4”. Pellets resuspended in PAGE sample buffer were boiled for 5 min to elute the immune complexes. Samples were electrophoresed in 10% SDS-polyacrylamide gels and the gels were processed forfluorography. (A) Lane 1, 35S-labeled virions; lanes 2, 3, 4, moAb 15 (group l-l) tested with the products of full-length, BarnHI, and 9-202 cDNAs, respectively; lanes 5,6, moAb 10 (group U-1) tested with the products of the 9-202 cDNA and the 17-202 cDNAs, respectively; lanes 7, 8, moAb 5 (group 2) tested with the products of the 17-202 cDNA and the 32-202 cDNA, respectively. (B) Lane 1, ?Yabeled virions; lanes 2, 3, moAb 7 (group 3) tested with the products of the 32-202 and EcoRl (-) cDNAs, respectively; lanes 4, 5, 6, moAb 2 (group 4) tested with the products of the EcoRl (-), BarnHI, and Xhol cDNAs, respectively. G, N, and M2 indicate the glycoprotein, nucleoprotein, and matrix protein, respectively, of rabies virus.
were constructed as described under Materials and Methods and immunoprecipitation tests were performed using in vitro translation products from the truncated cDNAs. Typical results and the summarized results are shown in Fig. 4 and Table 3, respectively. All of the representative clones from seven different groups precipitated an in vitro translation product from full-length cDNA as shown in Fig. 4A, lane 2, and the product migrated in a gel to the same position as that of the authentic M2 protein from rabiesvirions (Fig. 4A, lane 1). MoAb (No. 15) from group 1-l precipitated an in vitro translation product from the BarnHI cDNA fragment (Fig. 4A, lane 3) however, it did not react with that from the 9-202 cDNA fragment (Fig. 4A, lane 4). MoAb (No. 10) from group U-l and moAb (No. 5) from group 2 precipitated the translation products derived from the 9-202 cDNA fragment (Fig. 4A, lane 5) and from the BarnHI cDNA fragment (not shown in figure) but did not precipitate the product of the 32-202 cDNA fragment (Fig. 4A, lane 8 for moAb no. 5). However, the latter moAb reacted with the product of the 17-202 cDNA fragment (Fig. 4A, lane 7) whereas the former did not (Fig. 4A, lane 6). MoAb (No. 7) from group 3
M2 PROTEIN moAbs
477
precipitated the product by the 32-202 cDNA fragment (Fig. 4B, lane 2) however, it did not precipitate a product from the EcoRl (-) cDNA fragment (Fig. 4B, lane 3). Finally, moAb (No. 2) of group 4 precipitated the products from the EcoRl (-) cDNA fragment (Fig. 4B, lane 4) and the Xhol cDNA fragment (Fig. 4B, lane 6) but did not precipitate a product from the BarnHI cDNA fragment (Fig. 4B, lane 5). Attention should be paid in the interpretation of the location of antigenic determinants on the basis of differential reactivity with overlapping peptides derived from truncated cDNAs. If a peptide is shortened, its configuration and consequently its reactivity with a given moAbs might change. Therefore, only positive results will be meaningful and negative results should not be taken. Additionally, if there are many positive results with the given overlapping peptides, the minimal sequences that still show reactivity should be taken. On the basis of this principle, an antigenic determinant of moAbs from group 1 is suggested to be located in a region between amino acid 1 and 72, group 2 from 17 to 72, group 3 from 32 to 72, group 4 from 50 to 171, group U-l from 9 to 72, and group U-2 from 32 to 72. It would be possible to speculate more defined locations of these antigenic determinants if we could take the negative results, for instance, an antigenic determinant of moAbs from group 1 is located between amino acid residues 1 and 8, because the moAbs reacted with peptide 1-72 and did not react with peptide 9-202. However, one should provide direct evidence that the moAb certainly binds to synthetic oligopeptide l-8. MoAbs obtained were initially classified into four distinct epitope groups and an unclassified group by competitive binding assay and further divided into seven groups by examining their reactivity against the cells infected with the other strains of rabies virus. Furthermore, it was possible to map the antigenic determinants by immunoprecipitation reaction with in vitro translation products derived from truncated cDNAs of the M2 gene. By these analyses, 20 moAbs belonging to six groups have been suggested to recognize a determinant in the region of amino acids 1 to 72 from the N-terminus. This might reflect the finding that the Nterminal region of the M2 protein of rabies virus is essentially hydrophilic when analyzed from the deduced amino acid sequences (Rayssiguier et a/., 1986; Hiramatsu et a/., unpublished data). The results of several studies with VSV have pointed to the region of up to the 43rd amino acid from the N-terminus of the M protein as being responsible for transcription inhibition activity (Pal et a/., 1985b; Og-
478
HIRAMATSU
ET AL.
TABLE 3 CAPABILITYOF MOABS TO PRECIPITATEIN VITROTRANSLATIONPRODUCTSBYTRUNCATEDcDNAs OF THE M2 GENE With the product from truncated
Group
MoAb No.
Full-length (l-202)
Xho 1 (1-171)
BamHl (l-72)
EcoRl (-) (50-202)
l-2 1-2 2 3 4 U-l U-2
15*, 18, 20 1, 8*, 12, 16, 17 3, 4, 5*, 9, 11 7*, 13, 19, 21 2* 10*, 14 6*
+ + + + + + +
nd nd nd nd + nd nd
+ + + + -
nd -
+ +
+ -
cDNA fragment of 32-202 (32-202)
17-202 (17-202)
9-202) (g-202)
-
-
+ + -
+ + + -
+
+
+ nd nd + nd
nd nd -
Location of antigenic determinant indicated by amino acid No. from N-terminus l-72 l-72 17-72 32-72 50-171 9-72 32-72
Note. Asterisks indicate the number of moAbs used in the experiments as a representative clone of the group. The numbers in parentheses under the name of the truncated cDNA fragment indicate the number of amino acids of the peptides encoded by the cDNAs. + indicates that the moAb reacted with a peptide. nd; not tested.
den el al., 1986; Kaptur et al., 1991). To our knowledge, only one moAb (RM 3-9-l 6) against the M2 protein of rabies virus (HEP strain) has been prepared so far (Honda and Kawai, unpublished data). This moAb has been shown to exhibit almost the same reactivity pattern with our overlapping truncated peptides as that of moAbs from the U-l group, suggesting that the antigenie determinant lies between amino acids 9 and 72. MoAbs obtained in this study represent the largest panel of moAbs against the M2 protein and should be useful for further studies on the possible regulatory function of viral transcription and other functions of the M2 protein of rabies virus. ACKNOWLEDGMENTS This work was supported in part by research grants from the Japan-united States Cooperative Medical Science Program and from Yakult-Honsha Co. Ltd, Tokyo, Japan.
REFERENCES BLONDEL, D., HARMISON, G. G., and SCHUBERT, M. (1990). Role of matrix protein in cytopathogenesis of vesicular stomatitis virus. /. Viral. 64, 1716-1725. CLINTON, G. M., LITTLE, S. P., HAGEN, F. S., and HUANG, A. S. (1978). The matrix (M) protein of vesicular stomatitis virus regulates transcription. Cell 15, 1455-l 462. Cox, J. H., DIETZSCHOLD, B., and SCHNEIDER, L. G. (1977). Rabies virus glycoprotein. II. Biological and serological characterization. Infect. Immun. 16, 754-759. DE, B. P., THORNTON, G. B., LUK, D., and BANERJEE,A. K. (1982). Purified matrix protein of vesicular stomatitis virus blocks viral transcription in vitro. Proc. Natl. Acad. Sci. USA 79, 7 137-7 141. DIETZSCHOLD, B., Cox, J. H., SCHNEIDER, L. G., WIKTOR, T. J., and KOPROWSKI,H. (1978). Isolation and purification of a polymeric form of the glycoprotein of rabies virus. J. Gen. viral. 40, 131-l 39. DIETZSCHOLD, B., TOLLIS, M., LAFON, M., WUNNER, W. H., and KOPROWSKI,H. (1987). Mechanisms of rabies virus neutralization by
glycoprotein-specific monoclonal antibodies. virology 161, 2936. DIETZSCHOLD,B., WUNNER, W. H., WIKTOR, T. J., LOPES,A. D., LAFON, M., SMITH, C. L., and KOPROWSKI,H. (1983). Characterization of an antigenic determinant of the glycoprotein that correlates with pathogenicity of rabies virus. Proc. Nat/. Acad. Sci. USA 80, 7074. EY, P. L., PROWSE,S. J., and JENKIN, C. R. (1978). Isolation of pure IgGl , IgG2a and IgG2b immunoglobulins from mouse serum using protein A-Sepharose. immunochemistry 15, 429-436. KAPTUR, P. E., RHODES, R. B., and LYLES, D. S. (1991). Sequences of the vesicular stomatitis virus matrix protein involved in binding to nucleocapsids. 1. Viral. 65, 1057-l 065. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-685. MANNEN, K., HIRAMATSU, K., MIFUNE, K., and SAKAMOTO, S. (1991). Conserved nucleotide sequence of rabies virus cDNA encoding the nucleoprotein. Virus Genes 5, 69-73. MANNEN, K., MIFUNE, K., REID-SANDEN, F. L., SMITH, J. S., YAGER, P. A., SUMNER, J. W., FISHBEIN,D. B., TONG, T. C., and BAER, G. M. (1987). Microneutralization test for rabiesvirus based on an enzyme immunoassay. J. Clin. Microbial. 25, 2440-2442. MIFUNE, K., MANNEN, K., MINAMOTO, N., and ARAI, Y. T. (1986). Further studies on an improved haemagglutination inhibition test with higher sensitivity for rabies virus antibody. Bulletin W. H. 0. 64, 133-137. OGDEN, J. R., PAL, R., and WAGNER, R. R. (1986). Mapping regions of the matrix proteins of vesicular stomatitis virus which bind to ribonucleocapsids, liposomes, and monoclonal antibodies. J. Viral. 58,860-868. PAL, R., GRINNELL, B. W., SNYDER, R. M., WIENER, J. R., VOLK, W. A., and WAGNER, R. R. (1985a). Monoclonal antibodies to the M protein of vesicular stomatitis virus (Indiana serotype) and to a cDNA M gene expression product. /. Vifol. 55, 298-306. PAL, R., GRINNELL, B. W., SNYDER, R. M., and WAGNER, R. R. (198513). Regulation of viral transcription by the matrix protein of vesicular stomatitis virus probed by monoclonal antibodies and temperature-sensitive mutants. /. Viral. 56, 386-394. RAYSSIGUIER,C., CIOE, L., WITHERS, E., WUNNER, W. H., and CURTIS, P. J. (1986). Cloning of rabies virus matrix protein mRNA and determination of its amino acid sequence. Virus Res. 5, 177-l 90.
EPITOPE MAPPING
OF ANTI-RABIES
SALEH, F., GARD, G. P., and COMPANS, R. W. (1979). Synthesis of tacaribe viral proteins. Virology 93, 369-376. SEIF, I., COULON, P., ROLLIN, P. E., and FLAMAND, A. (1985). Rabies virulence: Effect on pathogenicity and sequence characterization of rabies virus mutations affecting antigenic site Ill of the glycoprotein. J. Viral. 53, 926-934. SHIPLEY,J. B., PAL, R., and WAGNER, R. R. (1988). Antigenicity, function, and cornformation of synthetic oligopeptides corresponding to amino-terminal sequences of wild-type and mutant matrix proteins of vesicular stomatitis virus. 1. Viol. 62, 2569-2577. TUFFEREAU,C., LEBLOIS, H., B~N~JEAN, J., COULON, P., LAFAY, F., and FLAMAND, A. (1989). Arginine or lysine in position 333 of ERA and CVS glycoprotein is necessary for rabies virulence in adult mice. virology 172, 206-212. VOLK, W. A., SNYDER, R. M., BENJAMIN, D. C., and WAGNER, R. R.
M2 PROTEIN moAbs
479
(1982). Monoclonal antibodies to the glycoprotein of vesicular stomatitis virus: Comparative neutralizing activity. 1. Viral. 42, 220227. WIKTOR, T. J., GYORGY, E., SCHLUMBERGER,H. D., SOKOL, F., and KoPROWSKI,H. (1973). Antigenic proper-ties of rabies virus components. J. Immunol. 110, 269-276. WIKTOR, T. J., and KOPROWSKI,H. (1978). Monoclonal antibodies against rabies virus produced by somatic cell hybridization: Detection of antigenic variants. Proc. Nat/. Acad. Sci, USA 75, 39383942. WIKTOR, T. J., MACFARLAN, R. I., REAGAN, K. .I., DIETZSCHOLD,B., CURTIS, P. J., WUNNER, W. H., KIENY, M.-P., LATHE, R., LECOCQ, J.-P., MACKETT, M., Moss, B., and KOPROWSKI,H. (1984). Protection from rabies by a vaccinia virus recombinant containing the rabies glycoprotein gene. Proc. Nat/. Acad. Sci. USA 81, 7194-7 198.
