- Email: [email protected]
Structural comparison of the cleavage-activation site of the fusion glycoprotein between virulent and avirulent strains of newcastle disease virus
VIROLOGY
158, 242-247
(1987)
Structural Comparison of the Cleavage-Activation Site of the Fusion Glycoprotein between Virulent and Avirulent Strains of Newcastle Disease Virus TETSUYATOYODA, TAKEMASA SAKAGUCHI,KUNITOSHIIMAI, NOEL MENDOZA INOCENCIO, BIN GOTOH, MICHINARI HAMAGUCHI, AND YOSHIYUKINAGAI’ Research Institute for Disease Mechanism
and Control, Nagoya University,
Received November
School of Medicine,
Nagoya 466, Japan
17, 1986; accepted January 23, 1987
The nucleotide sequence of the mRNA encoding the fusion (FJ protein of a virulent strain of Newcastle disease virus was determined. A single open reading frame in the sequence encodes a protein of 553 amino acids with a calculated molecular weight of 59058. The amino acid sequence predicted several structural features involving the fusion-inducing hydrophobic stretch (residues 117-l 42) and the cleavage-activation site (residues 112-l 16) to generate the disulfide-linked F, and F2 subunits. The cleavage-activation site as well as a part of the fusion-inducing sequence were compared among a series of virulent and avirulent strains by the chain-termination method using a synthetic oligonucleotide primer. It was found that without exception, the cleavage-activation site of virulent strains consisted of two dibasic residues with an intervening glutamine, Arg-Arg-Gln-Arg-Arg, whereas the corresponding region of avirulent strains was made of a sequence with single basic residues scattered among uncharged residues, Gly-$-Gln -!$-Arg. On the basis of these observations and the previous results showing a strict correlation between the pathogenicity and the cleavability of the fusion protein of NDV (Y. Nagai. H-D. Klenk, and R. Rott, Virology, 72, 494-508, 1976), we propose the importance of the dibasic residues for efficient proteolytic activation of the fusion protein and for the pantropic property of NDV. Some strains were found to have Leu-lle-Gly as the N-terminus of F, , whereas others contained Phe-lle-Gly, indicating that Phe-X-Gly is not always conserved at F, N-terminus of paramyxovirus. Q 1997 Academic Press. Inc.
Extensive studies with prototypic paramyxoviruses such as Sendai virus, SV5, and Newcastle disease virus (NDV) have shown that viral fusion (F) glycoprotein is synthesized as an inactive precursor, FO, which is cleaved by a trypsinlike protease to yield two disulfidelinked subunits, F, and Fp, and that this proteolytic cleavage is essential for the fusion of the viral envelope with the target-cell membrane and hence for the initiation of virus infection (1-5). A hydrophobic amino acid sequence is generated at the N-terminus of F, upon proteolytic cleavage and it is this sequence that is involved directly in the fusion activity (6). Among paramyxoviruses, NDV is unique in that the significant variation in virulence within the same serotype has been described (7). There is evidence that the cleavage of NDV F,, plays a major role in tissue tropism and pathogenesis of birds (2, 8, 9). The F0 of naturally occurring virulent strains is cleaved in a wide variety of cells both in vivo and in culture, whereas that of naturally occurring avirulent strains is done so only in limited types of cells such as endodermal cells of chick embryo. Thus only virulent strains are properly processed and able to grow in cells of various origins in vivo, whereas only endodermal cells are permissive for avirulent strains. This pantropic property of virulent NDV that permits its F,, to be proteolytically activated in different cells en-
ables the viruses to disseminate throughout its host and cause extensive diseases (8). Here, we describe cDNA clones of NDV F,, mRNA of a virulent strain, present its complete nucleotide sequence and the deduced amino acid sequence, and compare the proteolytic cleavage site and a part of the fusogenic sequence between virulent and avirulent strains. We previously constructed a cDNA library to mRNAs of a virulent Miyadera strain, by insertion of the cDNAs into Psrl site of pBR322 by homopolymer tailing (10). One cDNA clone designated 38Ell of 1700 bp was shown to contain F0 mRNA sequence (10). The sequence determination of both strands of the cDNA was carried out by recloning the restriction endonuclease fragments of cDNA into Ml 3 vectors mpl0 and mpl 1 (1 I) followed by dideoxy nucleotide chain termination (12). 38El 1 clone, however, included neither 3’ nor 5’ terminus of F,, mRNA. Another F,-specific clone (M5B5) from the cDNA library constructed using a vector primer plasmid pSV7 186 (13) served for determination of the 3’terminal sequence. The sequencing strategy of these clones is shown in Fig. 1. As shown in Fig. 2 the sequence determined indicates that the clone M5B5 contains the exact 3’ terminus with the same basic structure, 5’-TAAG(A)n-3’, as that of the polyadenylation signal found in Sendai virus, parainfluenza virus 3, and VSV (14-7 7). The exact number of A residues
’ To whom requests for reprints should be addressed. 242 0042-6822187 Copyright All rights
$3.00
0 1987 by Academic Press. Inc. of reproduction I” any form reserved.
243
SHORT COMMUNICATIONS base
t
c
M5B5
FIG. 1. DNA restriction map and the strategy for sequencing cDNA clones (38El 1 and M5B5) of the NDV F0 mRNA. The restriction enzyme sites were those used to produce fragments for subcloning into Ml3 phages. The arrows indicate the direction and extent of sequence determination in the message sense (arrow above) and anti-message sense (arrow below). The positions of primers Pl (a) and P2()) are shown. These primers were used for sequencing the 5’ terminal region and the cleavage site between F, and Fz protein subunits, respectively.
in this putative polyadenylation signal should await sequencing of the genome RNA but have been shown to be 6 for NDV mRNAs elsewhere (18, 19). To search for the missing sequence at 5’terminus, we carried out primer extension analysis with a synthetic oligonucleotide primer (17 nucleotides) (Fig. 2) and mRNA template isolated from infected BHK cells. The results indicated that the cDNA clone lacks the 5’ terminal sequence of 58 bases. This missing sequence was determined by the dideoxy chain termination method, the first two nucleotides not being read by this method. The 5’ end sequence NNGGGTAGAAG of the mRNA was similar to that of the recently identified 5’terminus (ACGGGTAGAAG) of NDV mRNAs ( 18-2 I). The mRNA is 1786 nucleotide long (excluding poly(A)). The ATG triplet at nucleotide positions 47-49 is the only significant start of a long open reading frame which terminates at positions 1706-1708. The other two reading frames are blocked frequently. The F0 message encodes an F0 protein of 553 amino acids with a molecular weight of 59,058 in unmodified form. Following charged residues, there is a core of 17 hydrophobic or uncharged amino acids (residues 925) at the N-terminus. This region could serve as the signal sequence for insertion of the protein into the membrane of rough endoplasmic reticulum. According to the (-3, -1) rule, counting from the cleavage site between positions -1 and +1 (22) there are three potential cleavage sites, Ala, Gly, and Cys, at residues 22, 24, and 25, respectively. There is a prominent cluster of 27 hydrophobic or uncharged amino acids at the C-terminus (residues 500-526). This sequence could serve as the transmembrane anchorage domain. It is followed by 27 residues which are relatively hydrophilic and could represent the cytoplasmic tail of the glyco-
protein. The size of this cyctoplasmic tail is consistent with our finding that trypsin or chymotrypsin digestion of microsomes of infected cells reduces the size of F. by about 2500 Da (23). The hydrophobic and fusogenic sequence was identified between 1 17 and 142. The first 20 residues of the sequence was exactly the same as that of the F, N-terminus of the Hickman strain NDV which had been determined by Edman degradation (6) except for the occurrence of Ser instead of Gly at residue 124. Immediately preceding this hydrophobic sequence, there is the cleavage site (residues 1 12-l 16) that is likely to yield F, and F2 subunits. This site consists of two dibasic residues with an intervening Gln (Arg-Arg-GlnArg-Arg). Eight potential glycosylation sites (Asn-X-Ser/Thr) were detected. One (85-87) is in the F, subunit. Earlier studies with radioactive precursor-labeled virions showed that the glucosamine to amino acid-labeling ratio was similar in F, and F2 of NDV (5). Assuming that not only the signal sequence but also the proteaseactivation site are removed as discussed below, F2 could have about 85 amino acid residues, and F, 437 residues, giving an approximate ratio of 1 to 5. Thus, it can be assumed that five of the seven potential sites in F, are actually glycosylated. Probably the site 541543 is not utilized because it is present within the probable cytoplasmic tail domain. One of the two sites, 19 l193 or 192-l 94, may not be utilized either, if they are spatially too close to each other to accept an oligosaccharide chain of a usual molecular weight of 2000 to 3000 (24, 25). There is one Cys (76) in F2 except for that in the presumed signal sequence. This could be involved in the disulfide bond between F, and F2 subunits (7, 5). The other seven of the nine cysteine resi-
SHORT COMMUNICATIONS
244
PL:s’-*GG TCTAGA NNGGGTAGAA GAGTTTGGAT CCCGGTTGGC GCGGGCIAGG CGCAAG ATG.GCC.TCC.AG*.TCT WCt-AI.-Ser-Ar~-sCr - GIY Pro - -
n,y.dCr. A”.tr.II.-“LCtOrI. ICI TGGTC-5’
_
-_-
--_
-
-
-
-
-
_
-
-
-
*
-
-
_
-
-
_
-
-
_
-
“.I
-
-
PZ:S’-CTT
FIG. 2. The nucleotide sequence of F0 mRNA of Miyadera strain and its predicted amino acid sequence. Underline, N-terminal signal sequence and C-terminal anchor region; double underline, N-terminal region of F, subunit. Cystein residues are dotted. Potential glycosylation sites for N-linked carbohydrate are boxed. Sequences of synthetic oligonucleotides Pl and P2 are shown at the corresponding positions. The amino acid sequence of F0 of Australia-Victoria strain (21) is shown for comparison. Dashes indicate identity with Miyadera strain.
245
dues are clustered between 338 and 424. Similar clustering of Cys residues seems to occur in the F0 protein of other paramyxoviruses (26-29). These Cys residues could be involved in intrachain disulfide bonds within F,, possibly leading to bunching of the polypeptide. Two other cysteines are in the membrane anchorage domain and may not be utilized for the disulfide bond formation but for fatty acid acylation (30). During preparation of this manuscript, McGinnes and Morrison (21) reported the F0 protein sequence of a virulent NDV strain, Australia-Victoria (AuV). There is a very high degree of homology (94.5%) in nucleotide sequence between this and our strain shown here. A comparison of amino acid sequence between two strains is shown in Fig. 2. The two F,, proteins are the same in size, having 553 residues, and exhibit 93.5% homology. The positions of the cysteine residues are completely conserved. The potential glycosylation sites are also well conserved, but two of them of Miyadera strain (residues 192-l 94 and 497-499) appear to be absent in AuV strain. Presence of dibasic amino acid residues or clustering of several basic residues is characteristic of proteolytic processing of peptide hormones (31, 32) as well as a number of viral glycoproteins involving influenza fowl plaque virus HA and many paramyxovirus F. proteins (21, 28, 29, 33, 34). In all of these cases and in the case of F. protein of Miyadera strain shown here, the precursor proteins are readily cleaved intracellularly, presumably at the trans-Golgi membranes (23, 31, 35) and by the same cellular protease (27). On the other hand, the F. protein of avirulent NDV strains is not efficiently cleaved in many cell types. This resembles the situation of Sendai virus F. protein as well as hemagglutinin (HA) protein of human influenza viruses, and it has been shown that these proteins have only one arginine at the cleavage site (26, 27, 36-39). We have compared the cleavage sites of virulent and avirulent strains by the dideoxy chain termination methods using a synthetic oligonucleotide as the primer and 50 S genomic RNA as the template. The primer sequence was derived from the nucleotide se-
quence from 329 to 345 (Fig. 2). The results are summarized in Fig. 3. The F. protein of virulent strains Italian and Herts has the same sequence at the cleavage sites as does the Miyadera strain (Arg-Arg-Gln-Arg-Arg). The only difference in the cleavage site of another virulent strain, AuV, is that one arginine at position -2, counting from the cleavage site between positions -1 and +l , was substituted for another basic residue, lysine (21). On the other hand, the corresponding site of avirulent strains La Sota, DZ6, Ulster, and Queensland were different. Two basic residues at positions -2 and -5, which were present in virulent strains, were, without exception, substituted for either glycine or serine. Each of these changes was due to one single base substitution. Thus, the dibasic sequences were no longer present at the cleavage site of each of avirulent F. proteins. It is now important to elucidate whether either or both of the dibasic structures are essential for efficient proteolytic cleavage. Evidence available indicates that one pair of basic residues is the minimum but fairly sufficient requirement for proteolytic processing of hormones and viral proteins in eukaryotic cells (31, 32, 40-42). Thus, either of the dibasic sequences could be the target of the cellular processing enzyme. However, to yield the correct hydrophobic F, N-terminus, Arg-Arg at positions -1 and -2 is apparently important. In fact, the cleavage site of human parainfluenza type 3 virus F. protein consists of sequence Pro(-5)Arg-Thr-Lys-Arg(-1), and the Lys-Arg could permit an efficient cleavage (29). However, the cleavage might be accelerated by the nearby dibasic residues or clustering of basic residues (29). Alternatively, the clustered basic residues may utilize a different processing enzyme which may also be widespread in eukaroytic cells. There is an additional dibasic sequence, Arg-Arg (residues 100 and lOl), in Miyadera strain (Fig. 2), which is not involved in proteolytic activation since uncleavable avirulent strains (DZ6 and Ulster) have the same sequence at the same position (not shown; M. Kawakita, personal communication). A similar difference in amino acid sequence around
246
SHORT COMMUNICATIONS
H
AGGATACAAGACTCTGTGACTACATCCGGAGGAAGGAGACAGAGACGC'TTTATAGGTGCCATTATCGGTAGTGTAGCT
II
---c----c
----------TC-------------------------------------------C---------
I A L D Q
-----------------A--C---A------A-----
A----G-----C-------------------C---------
U (B) 10
5 I4
heIleGlyAlaIleIleGlySerValAla
H
---
I
____-_-_-_
A
-_--------
L
______-_-_
D
e"-
Q
e" -
U
eu -
-
-
-
-
-
-
-
-
-
-
-
- Gly -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FIG. 3. Comparison of nucleotide sequence (A) and amino acid sequence(B) of cleavage site among virulent and avirulent strains. Nucleotide sequence was determined by primer extension with P2 primer (Fig. 2) using the viral genome RNA as the template. Strains Miyadera (M), Herts (H), Italian (I), and Australia-Victoria (A) are virulent, whereas strains La Sota (L), Dz6 (D), Queensland (Q), and Ulster (U) are avirulent. Arrow, cleavage site. Dashes indicate identity with Miyadera strain. Basic amino acid residues are boxed. Data of Australia-Victoria strain were taken from McGinnes and Morrison (21).
the cleavage site of influenza HA protein was proposed to determine its cleavability and the host range of the virus (43, 44). Furthermore, it has been demonstrated that the proteolytic activation of the HA protein is not only a result of simple cleavage of the polypeptide by the endopeptidase but is followed by removal of the arginine residues by an exopeptidaselike carboxypeptidase B (44). Similar removal of the cleavage site of NDV F0 protein may occur, because the proteolytic activation is paralleled by an acidic shift in the isoelectric point of the fusion protein (45). The N-terminal region of F, proteins of paramyxoviruses so far sequenced has shared certain structural features. One of these is Phe-X-Gly as the N-terminus, and the others are overall hydrophobicity and the presence of spaced glycines. Furthermore, the oligopeptides which mimic the N-terminus, but not those unrelated to the N-terminal sequence, were found to inhibit the viral fusion activity and the virus spread from cell to cell possibly by competing for a cellular site(s) with F, (6, 46). This indicates the direct involvement of F, N-terminal region in fusion activity of paramyxoviruses. Furthermore, these studies have shown that an carbobenzoxy-o-Phe-Phe-Gly correoligopeptide, sponding to Sendai virus F, N-terminus (Phe-Phe-Gly),
inhibits not only Sendai virus infectivity but also that of SV5 and measles virus, whose F, N-termini are both Phe-Ala-Gly (6, 47) and that the steric configuration of the first two amino acids significantly affects the inhibitory effect. Thus, it appears that the first aromatic residue, Phe, is not only structurally conserved among paramyxoviruses but also functionally important for the viral fusion activity. In view of these considerations, it is quite unexpected that strains Dp6, Queensland, and Ulster have Leu instead of Phe at the F, N-terminus. These strains can also exhibit prominent fusion activity, and replicate very efficiently through multiple growth cycle, if their F0 is proteolytically activated by trypsin (2, 48, 49). Thus, the F, N-terminal Phe would not always be an essential requirement for paramyxovirus fusion activity. Probably, the comparative inhibition studies of different NDV strains with synthetic oligopeptides corresponding to the two distinct F, N-terminal sequences would be helpful for understanding the role of F, N-terminal region in the paramyxovirus-induced cell fusion and are now in progress. It has to be also noted that those strains found to have Leu (i-1) were previously shown to be unique in that their hemagglutinin-neuraminidase (HN) glycoprotein is formed from a larger precursor HN, (2, 47, 48).
SHORT COMMUNICATIONS
With the other strains having Phe (+l), no such HNo precursor has been detected and the HN appears to be the primary translation product (47). It remains to be elucidated whether these parallelisms are merely casual or have any evolutionary significance. In summary, the nucleotide sequence of the mRNA encoding the NDV fusion protein was determined. A single open reading frame in the sequence encodes a hydrophobic protein of 553 amino acids with a molecular weight of 59058, in which there is the fusion-inducing hydrophobic stretch of 29 amino acids of F, Nterminus. The F0 precursor protein of each of the virulent strains analyzed has a pair of dibasic residues with an intervening Gln at the proteolytic cleavage site. On the other hand, the corresponding region of avirulent strains contains two single basic residues scattered among uncharged residues and no longer has dibasic cleavage signals. This structural difference could account for the marked difference in cleavability between virulent and avirulent F,, proteins and thereby the difference in virulence. ACKNOWLEDGMENTS We thank Drs. M. Tanaka, K. Nakajima, K. Shigesada. M. Kawakita, and H. Shibuta for invaluable help in cloning experiments. This work was supported by a Grant-in-Aid for Scientific Research from Ministry of Education, Science and Culture, Japan, by lshida Foundation. and by Naito Foundation.
REFERENCES 1. NAGAI, Y., OGURA, H., and KLENK, H-D., Virology 69, 523-538 (1976). 2. NAGAI, Y., KLENK, H.-D., and Roar, R., Virology 72, 494-508 (1976). 3. HOMMA, M., and OHUCHI, M., /. Viral. 12, 1457-1465 (1973). 4. SCHEID, A., and CHOPPIN, P. W., Virology 57, 470-490 (1974). 5. SCHEID, A., and CHOPPIN, P. W., Virology 80, 54-66 (1977). 6. RICHARDSON,C. D., SCHEID,A., and CHOPPIN, P. W., Virology 105, 205-222 (1980). 7. WATERSON,A. P., PENNINGTON,T. M., and ALLAN, W. M., &it. Med. Bull. 23, 138-l 43 (1967). 8. NAGAI, Y., SHIMOKATA, K., YOSHIDA, T., HAMAGUCHI, M., IINUMA, M., MAENO, K., MATSUMOTO, T., KLENK, H-D., and Roar, R., 1. Gen. Viral. 45, 263-272 (1979). 9. GARTEN, W., BERK, W., NAGAI, Y., Ro-rr, R., and KLENK, H-D., J. Gen. Viral. 50, 135-147 (1980). 10. TOYODA, T., HAMAGUCHI, M., and NAGAI, Y., Arch. Viral., in press. 17. MESSING,J., In “Methods in Enzymology” (R. Wu, L. Grossman, and K. Moldave, Ed%). Vol. 101, pp. 20-78. Academic Press, New York/London, 1983. 12. SANGER,A. C. R., CHAMBERS,P., and DICKINSON,J. H., Proc. Nat/. Acad. Sci. USA 74, 5463-5467 (1977). 13. OKAYAMA, H., and BERG, P., Mol. Cell. Bio. 2, 161-170(1982). 14. GUPTA. C., and KINGSBURY,D. W., Virology 135, 518-523 (1982). 15. MORGAN, E. M., RE, G. G., and KINGSBURY,D. W., Virology 135, 279-287 (1982). 76. SCHANCHEZ,A., and BANERJEE,A. K., Virology 143,45-54 (1985). 17. ROSE,J. K., Ce// 19, 415-421 (1980). 18. MILLER, N. S., CHAMBERS, P., and EMMERSON,P. T., 1. Gen. Viral. 67, 1917-1927 (1986).
247
19. ISHIDA, N., TAIRA, H., OMATA, T., MIZUMOTO, K., HATTORI, S., IWASAKI,K., and KAWAKITA,M., NucleicAcids Res. 14, 6551-6564 (1986). 20. KURILLA, M. G., STONE, H. D., and KEENE, J. D., Virology 145, 203-212 (1985). 21. MCGINNES, L. W., and MORRISON,T. G., Virus Res. 5, 343-356 (1986). 22. VON HEIJINE,G.. Eur. J. Biochem. 133, 17-21 (1983). 23. YOSHIDA, T., NAKAYAMA,Y., NAGURA, H., TOYODA, T., NISHIKAWA. K., HAMAGUCHI, M., and NAGAI, Y., Virus Res. 4, 179-195 (1986). 24. KEIL, W., KLENK, H-D., and SCHWARZ,R. T., 1. Viral. 31, 253-256 (1979). 25. HORISBERGER,M. A., DE STARITZKY,C., and CONTENT, J., Arch. Viral. 64, 9-l 6 (1980). 26. SHIODA, T., IWASAKI,K., and SHIBUTA, H., Nucleic Acids Res. 14, 1545-1563(1986). 27. BLUMBERG,B. M., GIORGE, C., ROSE, K., and KOLAKOFSKY,D., J. Gen. Viral. 60, 317-331 (1985). 28. PATERSON,R. G., HARRIS,T. J. R., and LYME, R. A. C., Proc. Nat/. Acad. Sci. USA 81, 6706-6710 (1984). 29. SPRIGGS,M. K.. OLMSTED, R. A., VENKATESAN,S., COLIGAN, J. E., and COLLINS, P. L., Virology 152, 241-251 (1986). 30. ROSE,J., ADAMS, G., and GALLIONE,C., Proc. Nat/. Acad. Sci. USA 81, 2050-2054 (1984). 31. STREINER,D. F., QUINN, P. S., CHAN, S. J., MARSK, J., and TAGER, D. F., Ann. N. Y. Acad. Sci. 343, 1- 16 (1980). 32. UNDENFRIEND,S., and KILPATRICK,D. L., Arch. Biochem. Biophys. 221,309-323 (1983). 33. COLLINS, P. L., HUANG, Y. T., and WERTZ. G. W.. Proc. Nat/. Acad. Sci. USA 81,7687-7691 (1984). 34. PORTER,A. C., BARBER.C., CAREY, N. H., HALLWELL,R. A., THRELFALL, G., and EMTAGE, J. S., Nature (London) 282, 471-477 (1979). 35. MORRISON,T., WARD, L., and SEMERJIAN,A., J. Viral. 53,85 l-857 (1985). 36. WARD, C. W., and DOPHEIDE,T. A. A., Virology 103, 37-53 (1980). 37. GETHING,M. J., BYE,J., SKEHEL,J. J., and WATERFIELD,M. D., Nature (London) 287, 301-306 (1980). 38. WATERFIELD,M. D., ESPELIE,K., ELDER, K., and SKEHEL,J. J., Brit. Med. Bull. 35, 57-63 (1979). 39. MIN-JOE, W., VERHOGEN,M., DAVOS. R., SAMAN, E., FANG, R., HUYLESROCK,D., FIEERS,W., THRELFALL,G.. CAREY,N., and EMTAGE, S. C., Cell 19, 683-696 (1980). 40. ORO~~LAN, S., HENDERSON, L. E., COPELAND, T. D., SCHULTZ, A. M., and RABIN, E. M., ln “Biosynthesis, Modification and Processing of Cellular and Viral Proteins” (G. Koch and D. Richter, Eds.), pp. 219-232. Academic Press, New York, 1980. 41. GAROFF,H., FRISCHANF,A. M., SIMONS,K., LEHRACH,H., and DELIUS, H.. Nature (London) 288, 236-241 (1980). 42. RICE, C. M., and STRAUSS,1. H.. Proc. Nat/. Acad. Sci. USA 178, 2062-2066 (1981). 43. BOSCH, F. X., GARTEN, W., KLENK, H-D., and Roar, R., Virology 113, 725-735 (1981). 44. GARTEN.W., BOSCH, F. X., LINDER, D., Roar, R., and KLENK, H-D., Virology 115, 361-374 (1981). 45. KOHAMA, T.. GARTEN, W., and KLENK, H-D., Virology 111, 364376 (1981). 46. RICHARDSON,C. D.. and CHOPPIN, P. W., Virology 131, 518-532 (1983). 47. VARSANYI,T. M., JORNVALL,H., and NORREY, E., Virology 147, llO-117(1985). 48. NAGAI, Y., and KLENK, H-D., Virology 77, 125-l 34 (1977). 49. NAGAI, Y., YOSHIDA, T., HAMAGUCHI, M., NARUSE, H., IINUMA, M., MAENO, K., and MATSUMOTO,T., Microbial. lmmunol. 24, 173177 (1980).
158, 242-247
(1987)
Structural Comparison of the Cleavage-Activation Site of the Fusion Glycoprotein between Virulent and Avirulent Strains of Newcastle Disease Virus TETSUYATOYODA, TAKEMASA SAKAGUCHI,KUNITOSHIIMAI, NOEL MENDOZA INOCENCIO, BIN GOTOH, MICHINARI HAMAGUCHI, AND YOSHIYUKINAGAI’ Research Institute for Disease Mechanism
and Control, Nagoya University,
Received November
School of Medicine,
Nagoya 466, Japan
17, 1986; accepted January 23, 1987
The nucleotide sequence of the mRNA encoding the fusion (FJ protein of a virulent strain of Newcastle disease virus was determined. A single open reading frame in the sequence encodes a protein of 553 amino acids with a calculated molecular weight of 59058. The amino acid sequence predicted several structural features involving the fusion-inducing hydrophobic stretch (residues 117-l 42) and the cleavage-activation site (residues 112-l 16) to generate the disulfide-linked F, and F2 subunits. The cleavage-activation site as well as a part of the fusion-inducing sequence were compared among a series of virulent and avirulent strains by the chain-termination method using a synthetic oligonucleotide primer. It was found that without exception, the cleavage-activation site of virulent strains consisted of two dibasic residues with an intervening glutamine, Arg-Arg-Gln-Arg-Arg, whereas the corresponding region of avirulent strains was made of a sequence with single basic residues scattered among uncharged residues, Gly-$-Gln -!$-Arg. On the basis of these observations and the previous results showing a strict correlation between the pathogenicity and the cleavability of the fusion protein of NDV (Y. Nagai. H-D. Klenk, and R. Rott, Virology, 72, 494-508, 1976), we propose the importance of the dibasic residues for efficient proteolytic activation of the fusion protein and for the pantropic property of NDV. Some strains were found to have Leu-lle-Gly as the N-terminus of F, , whereas others contained Phe-lle-Gly, indicating that Phe-X-Gly is not always conserved at F, N-terminus of paramyxovirus. Q 1997 Academic Press. Inc.
Extensive studies with prototypic paramyxoviruses such as Sendai virus, SV5, and Newcastle disease virus (NDV) have shown that viral fusion (F) glycoprotein is synthesized as an inactive precursor, FO, which is cleaved by a trypsinlike protease to yield two disulfidelinked subunits, F, and Fp, and that this proteolytic cleavage is essential for the fusion of the viral envelope with the target-cell membrane and hence for the initiation of virus infection (1-5). A hydrophobic amino acid sequence is generated at the N-terminus of F, upon proteolytic cleavage and it is this sequence that is involved directly in the fusion activity (6). Among paramyxoviruses, NDV is unique in that the significant variation in virulence within the same serotype has been described (7). There is evidence that the cleavage of NDV F,, plays a major role in tissue tropism and pathogenesis of birds (2, 8, 9). The F0 of naturally occurring virulent strains is cleaved in a wide variety of cells both in vivo and in culture, whereas that of naturally occurring avirulent strains is done so only in limited types of cells such as endodermal cells of chick embryo. Thus only virulent strains are properly processed and able to grow in cells of various origins in vivo, whereas only endodermal cells are permissive for avirulent strains. This pantropic property of virulent NDV that permits its F,, to be proteolytically activated in different cells en-
ables the viruses to disseminate throughout its host and cause extensive diseases (8). Here, we describe cDNA clones of NDV F,, mRNA of a virulent strain, present its complete nucleotide sequence and the deduced amino acid sequence, and compare the proteolytic cleavage site and a part of the fusogenic sequence between virulent and avirulent strains. We previously constructed a cDNA library to mRNAs of a virulent Miyadera strain, by insertion of the cDNAs into Psrl site of pBR322 by homopolymer tailing (10). One cDNA clone designated 38Ell of 1700 bp was shown to contain F0 mRNA sequence (10). The sequence determination of both strands of the cDNA was carried out by recloning the restriction endonuclease fragments of cDNA into Ml 3 vectors mpl0 and mpl 1 (1 I) followed by dideoxy nucleotide chain termination (12). 38El 1 clone, however, included neither 3’ nor 5’ terminus of F,, mRNA. Another F,-specific clone (M5B5) from the cDNA library constructed using a vector primer plasmid pSV7 186 (13) served for determination of the 3’terminal sequence. The sequencing strategy of these clones is shown in Fig. 1. As shown in Fig. 2 the sequence determined indicates that the clone M5B5 contains the exact 3’ terminus with the same basic structure, 5’-TAAG(A)n-3’, as that of the polyadenylation signal found in Sendai virus, parainfluenza virus 3, and VSV (14-7 7). The exact number of A residues
’ To whom requests for reprints should be addressed. 242 0042-6822187 Copyright All rights
$3.00
0 1987 by Academic Press. Inc. of reproduction I” any form reserved.
243
SHORT COMMUNICATIONS base
t
c
M5B5
FIG. 1. DNA restriction map and the strategy for sequencing cDNA clones (38El 1 and M5B5) of the NDV F0 mRNA. The restriction enzyme sites were those used to produce fragments for subcloning into Ml3 phages. The arrows indicate the direction and extent of sequence determination in the message sense (arrow above) and anti-message sense (arrow below). The positions of primers Pl (a) and P2()) are shown. These primers were used for sequencing the 5’ terminal region and the cleavage site between F, and Fz protein subunits, respectively.
in this putative polyadenylation signal should await sequencing of the genome RNA but have been shown to be 6 for NDV mRNAs elsewhere (18, 19). To search for the missing sequence at 5’terminus, we carried out primer extension analysis with a synthetic oligonucleotide primer (17 nucleotides) (Fig. 2) and mRNA template isolated from infected BHK cells. The results indicated that the cDNA clone lacks the 5’ terminal sequence of 58 bases. This missing sequence was determined by the dideoxy chain termination method, the first two nucleotides not being read by this method. The 5’ end sequence NNGGGTAGAAG of the mRNA was similar to that of the recently identified 5’terminus (ACGGGTAGAAG) of NDV mRNAs ( 18-2 I). The mRNA is 1786 nucleotide long (excluding poly(A)). The ATG triplet at nucleotide positions 47-49 is the only significant start of a long open reading frame which terminates at positions 1706-1708. The other two reading frames are blocked frequently. The F0 message encodes an F0 protein of 553 amino acids with a molecular weight of 59,058 in unmodified form. Following charged residues, there is a core of 17 hydrophobic or uncharged amino acids (residues 925) at the N-terminus. This region could serve as the signal sequence for insertion of the protein into the membrane of rough endoplasmic reticulum. According to the (-3, -1) rule, counting from the cleavage site between positions -1 and +1 (22) there are three potential cleavage sites, Ala, Gly, and Cys, at residues 22, 24, and 25, respectively. There is a prominent cluster of 27 hydrophobic or uncharged amino acids at the C-terminus (residues 500-526). This sequence could serve as the transmembrane anchorage domain. It is followed by 27 residues which are relatively hydrophilic and could represent the cytoplasmic tail of the glyco-
protein. The size of this cyctoplasmic tail is consistent with our finding that trypsin or chymotrypsin digestion of microsomes of infected cells reduces the size of F. by about 2500 Da (23). The hydrophobic and fusogenic sequence was identified between 1 17 and 142. The first 20 residues of the sequence was exactly the same as that of the F, N-terminus of the Hickman strain NDV which had been determined by Edman degradation (6) except for the occurrence of Ser instead of Gly at residue 124. Immediately preceding this hydrophobic sequence, there is the cleavage site (residues 1 12-l 16) that is likely to yield F, and F2 subunits. This site consists of two dibasic residues with an intervening Gln (Arg-Arg-GlnArg-Arg). Eight potential glycosylation sites (Asn-X-Ser/Thr) were detected. One (85-87) is in the F, subunit. Earlier studies with radioactive precursor-labeled virions showed that the glucosamine to amino acid-labeling ratio was similar in F, and F2 of NDV (5). Assuming that not only the signal sequence but also the proteaseactivation site are removed as discussed below, F2 could have about 85 amino acid residues, and F, 437 residues, giving an approximate ratio of 1 to 5. Thus, it can be assumed that five of the seven potential sites in F, are actually glycosylated. Probably the site 541543 is not utilized because it is present within the probable cytoplasmic tail domain. One of the two sites, 19 l193 or 192-l 94, may not be utilized either, if they are spatially too close to each other to accept an oligosaccharide chain of a usual molecular weight of 2000 to 3000 (24, 25). There is one Cys (76) in F2 except for that in the presumed signal sequence. This could be involved in the disulfide bond between F, and F2 subunits (7, 5). The other seven of the nine cysteine resi-
SHORT COMMUNICATIONS
244
PL:s’-*GG TCTAGA NNGGGTAGAA GAGTTTGGAT CCCGGTTGGC GCGGGCIAGG CGCAAG ATG.GCC.TCC.AG*.TCT WCt-AI.-Ser-Ar~-sCr - GIY Pro - -
n,y.dCr. A”.tr.II.-“LCtOrI. ICI TGGTC-5’
_
-_-
--_
-
-
-
-
-
_
-
-
-
*
-
-
_
-
-
_
-
-
_
-
“.I
-
-
PZ:S’-CTT
FIG. 2. The nucleotide sequence of F0 mRNA of Miyadera strain and its predicted amino acid sequence. Underline, N-terminal signal sequence and C-terminal anchor region; double underline, N-terminal region of F, subunit. Cystein residues are dotted. Potential glycosylation sites for N-linked carbohydrate are boxed. Sequences of synthetic oligonucleotides Pl and P2 are shown at the corresponding positions. The amino acid sequence of F0 of Australia-Victoria strain (21) is shown for comparison. Dashes indicate identity with Miyadera strain.
245
dues are clustered between 338 and 424. Similar clustering of Cys residues seems to occur in the F0 protein of other paramyxoviruses (26-29). These Cys residues could be involved in intrachain disulfide bonds within F,, possibly leading to bunching of the polypeptide. Two other cysteines are in the membrane anchorage domain and may not be utilized for the disulfide bond formation but for fatty acid acylation (30). During preparation of this manuscript, McGinnes and Morrison (21) reported the F0 protein sequence of a virulent NDV strain, Australia-Victoria (AuV). There is a very high degree of homology (94.5%) in nucleotide sequence between this and our strain shown here. A comparison of amino acid sequence between two strains is shown in Fig. 2. The two F,, proteins are the same in size, having 553 residues, and exhibit 93.5% homology. The positions of the cysteine residues are completely conserved. The potential glycosylation sites are also well conserved, but two of them of Miyadera strain (residues 192-l 94 and 497-499) appear to be absent in AuV strain. Presence of dibasic amino acid residues or clustering of several basic residues is characteristic of proteolytic processing of peptide hormones (31, 32) as well as a number of viral glycoproteins involving influenza fowl plaque virus HA and many paramyxovirus F. proteins (21, 28, 29, 33, 34). In all of these cases and in the case of F. protein of Miyadera strain shown here, the precursor proteins are readily cleaved intracellularly, presumably at the trans-Golgi membranes (23, 31, 35) and by the same cellular protease (27). On the other hand, the F. protein of avirulent NDV strains is not efficiently cleaved in many cell types. This resembles the situation of Sendai virus F. protein as well as hemagglutinin (HA) protein of human influenza viruses, and it has been shown that these proteins have only one arginine at the cleavage site (26, 27, 36-39). We have compared the cleavage sites of virulent and avirulent strains by the dideoxy chain termination methods using a synthetic oligonucleotide as the primer and 50 S genomic RNA as the template. The primer sequence was derived from the nucleotide se-
quence from 329 to 345 (Fig. 2). The results are summarized in Fig. 3. The F. protein of virulent strains Italian and Herts has the same sequence at the cleavage sites as does the Miyadera strain (Arg-Arg-Gln-Arg-Arg). The only difference in the cleavage site of another virulent strain, AuV, is that one arginine at position -2, counting from the cleavage site between positions -1 and +l , was substituted for another basic residue, lysine (21). On the other hand, the corresponding site of avirulent strains La Sota, DZ6, Ulster, and Queensland were different. Two basic residues at positions -2 and -5, which were present in virulent strains, were, without exception, substituted for either glycine or serine. Each of these changes was due to one single base substitution. Thus, the dibasic sequences were no longer present at the cleavage site of each of avirulent F. proteins. It is now important to elucidate whether either or both of the dibasic structures are essential for efficient proteolytic cleavage. Evidence available indicates that one pair of basic residues is the minimum but fairly sufficient requirement for proteolytic processing of hormones and viral proteins in eukaryotic cells (31, 32, 40-42). Thus, either of the dibasic sequences could be the target of the cellular processing enzyme. However, to yield the correct hydrophobic F, N-terminus, Arg-Arg at positions -1 and -2 is apparently important. In fact, the cleavage site of human parainfluenza type 3 virus F. protein consists of sequence Pro(-5)Arg-Thr-Lys-Arg(-1), and the Lys-Arg could permit an efficient cleavage (29). However, the cleavage might be accelerated by the nearby dibasic residues or clustering of basic residues (29). Alternatively, the clustered basic residues may utilize a different processing enzyme which may also be widespread in eukaroytic cells. There is an additional dibasic sequence, Arg-Arg (residues 100 and lOl), in Miyadera strain (Fig. 2), which is not involved in proteolytic activation since uncleavable avirulent strains (DZ6 and Ulster) have the same sequence at the same position (not shown; M. Kawakita, personal communication). A similar difference in amino acid sequence around
246
SHORT COMMUNICATIONS
H
AGGATACAAGACTCTGTGACTACATCCGGAGGAAGGAGACAGAGACGC'TTTATAGGTGCCATTATCGGTAGTGTAGCT
II
---c----c
----------TC-------------------------------------------C---------
I A L D Q
-----------------A--C---A------A-----
A----G-----C-------------------C---------
U (B) 10
5 I4
heIleGlyAlaIleIleGlySerValAla
H
---
I
____-_-_-_
A
-_--------
L
______-_-_
D
e"-
Q
e" -
U
eu -
-
-
-
-
-
-
-
-
-
-
-
- Gly -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FIG. 3. Comparison of nucleotide sequence (A) and amino acid sequence(B) of cleavage site among virulent and avirulent strains. Nucleotide sequence was determined by primer extension with P2 primer (Fig. 2) using the viral genome RNA as the template. Strains Miyadera (M), Herts (H), Italian (I), and Australia-Victoria (A) are virulent, whereas strains La Sota (L), Dz6 (D), Queensland (Q), and Ulster (U) are avirulent. Arrow, cleavage site. Dashes indicate identity with Miyadera strain. Basic amino acid residues are boxed. Data of Australia-Victoria strain were taken from McGinnes and Morrison (21).
the cleavage site of influenza HA protein was proposed to determine its cleavability and the host range of the virus (43, 44). Furthermore, it has been demonstrated that the proteolytic activation of the HA protein is not only a result of simple cleavage of the polypeptide by the endopeptidase but is followed by removal of the arginine residues by an exopeptidaselike carboxypeptidase B (44). Similar removal of the cleavage site of NDV F0 protein may occur, because the proteolytic activation is paralleled by an acidic shift in the isoelectric point of the fusion protein (45). The N-terminal region of F, proteins of paramyxoviruses so far sequenced has shared certain structural features. One of these is Phe-X-Gly as the N-terminus, and the others are overall hydrophobicity and the presence of spaced glycines. Furthermore, the oligopeptides which mimic the N-terminus, but not those unrelated to the N-terminal sequence, were found to inhibit the viral fusion activity and the virus spread from cell to cell possibly by competing for a cellular site(s) with F, (6, 46). This indicates the direct involvement of F, N-terminal region in fusion activity of paramyxoviruses. Furthermore, these studies have shown that an carbobenzoxy-o-Phe-Phe-Gly correoligopeptide, sponding to Sendai virus F, N-terminus (Phe-Phe-Gly),
inhibits not only Sendai virus infectivity but also that of SV5 and measles virus, whose F, N-termini are both Phe-Ala-Gly (6, 47) and that the steric configuration of the first two amino acids significantly affects the inhibitory effect. Thus, it appears that the first aromatic residue, Phe, is not only structurally conserved among paramyxoviruses but also functionally important for the viral fusion activity. In view of these considerations, it is quite unexpected that strains Dp6, Queensland, and Ulster have Leu instead of Phe at the F, N-terminus. These strains can also exhibit prominent fusion activity, and replicate very efficiently through multiple growth cycle, if their F0 is proteolytically activated by trypsin (2, 48, 49). Thus, the F, N-terminal Phe would not always be an essential requirement for paramyxovirus fusion activity. Probably, the comparative inhibition studies of different NDV strains with synthetic oligopeptides corresponding to the two distinct F, N-terminal sequences would be helpful for understanding the role of F, N-terminal region in the paramyxovirus-induced cell fusion and are now in progress. It has to be also noted that those strains found to have Leu (i-1) were previously shown to be unique in that their hemagglutinin-neuraminidase (HN) glycoprotein is formed from a larger precursor HN, (2, 47, 48).
SHORT COMMUNICATIONS
With the other strains having Phe (+l), no such HNo precursor has been detected and the HN appears to be the primary translation product (47). It remains to be elucidated whether these parallelisms are merely casual or have any evolutionary significance. In summary, the nucleotide sequence of the mRNA encoding the NDV fusion protein was determined. A single open reading frame in the sequence encodes a hydrophobic protein of 553 amino acids with a molecular weight of 59058, in which there is the fusion-inducing hydrophobic stretch of 29 amino acids of F, Nterminus. The F0 precursor protein of each of the virulent strains analyzed has a pair of dibasic residues with an intervening Gln at the proteolytic cleavage site. On the other hand, the corresponding region of avirulent strains contains two single basic residues scattered among uncharged residues and no longer has dibasic cleavage signals. This structural difference could account for the marked difference in cleavability between virulent and avirulent F,, proteins and thereby the difference in virulence. ACKNOWLEDGMENTS We thank Drs. M. Tanaka, K. Nakajima, K. Shigesada. M. Kawakita, and H. Shibuta for invaluable help in cloning experiments. This work was supported by a Grant-in-Aid for Scientific Research from Ministry of Education, Science and Culture, Japan, by lshida Foundation. and by Naito Foundation.
REFERENCES 1. NAGAI, Y., OGURA, H., and KLENK, H-D., Virology 69, 523-538 (1976). 2. NAGAI, Y., KLENK, H.-D., and Roar, R., Virology 72, 494-508 (1976). 3. HOMMA, M., and OHUCHI, M., /. Viral. 12, 1457-1465 (1973). 4. SCHEID, A., and CHOPPIN, P. W., Virology 57, 470-490 (1974). 5. SCHEID, A., and CHOPPIN, P. W., Virology 80, 54-66 (1977). 6. RICHARDSON,C. D., SCHEID,A., and CHOPPIN, P. W., Virology 105, 205-222 (1980). 7. WATERSON,A. P., PENNINGTON,T. M., and ALLAN, W. M., &it. Med. Bull. 23, 138-l 43 (1967). 8. NAGAI, Y., SHIMOKATA, K., YOSHIDA, T., HAMAGUCHI, M., IINUMA, M., MAENO, K., MATSUMOTO, T., KLENK, H-D., and Roar, R., 1. Gen. Viral. 45, 263-272 (1979). 9. GARTEN, W., BERK, W., NAGAI, Y., Ro-rr, R., and KLENK, H-D., J. Gen. Viral. 50, 135-147 (1980). 10. TOYODA, T., HAMAGUCHI, M., and NAGAI, Y., Arch. Viral., in press. 17. MESSING,J., In “Methods in Enzymology” (R. Wu, L. Grossman, and K. Moldave, Ed%). Vol. 101, pp. 20-78. Academic Press, New York/London, 1983. 12. SANGER,A. C. R., CHAMBERS,P., and DICKINSON,J. H., Proc. Nat/. Acad. Sci. USA 74, 5463-5467 (1977). 13. OKAYAMA, H., and BERG, P., Mol. Cell. Bio. 2, 161-170(1982). 14. GUPTA. C., and KINGSBURY,D. W., Virology 135, 518-523 (1982). 15. MORGAN, E. M., RE, G. G., and KINGSBURY,D. W., Virology 135, 279-287 (1982). 76. SCHANCHEZ,A., and BANERJEE,A. K., Virology 143,45-54 (1985). 17. ROSE,J. K., Ce// 19, 415-421 (1980). 18. MILLER, N. S., CHAMBERS, P., and EMMERSON,P. T., 1. Gen. Viral. 67, 1917-1927 (1986).
247
19. ISHIDA, N., TAIRA, H., OMATA, T., MIZUMOTO, K., HATTORI, S., IWASAKI,K., and KAWAKITA,M., NucleicAcids Res. 14, 6551-6564 (1986). 20. KURILLA, M. G., STONE, H. D., and KEENE, J. D., Virology 145, 203-212 (1985). 21. MCGINNES, L. W., and MORRISON,T. G., Virus Res. 5, 343-356 (1986). 22. VON HEIJINE,G.. Eur. J. Biochem. 133, 17-21 (1983). 23. YOSHIDA, T., NAKAYAMA,Y., NAGURA, H., TOYODA, T., NISHIKAWA. K., HAMAGUCHI, M., and NAGAI, Y., Virus Res. 4, 179-195 (1986). 24. KEIL, W., KLENK, H-D., and SCHWARZ,R. T., 1. Viral. 31, 253-256 (1979). 25. HORISBERGER,M. A., DE STARITZKY,C., and CONTENT, J., Arch. Viral. 64, 9-l 6 (1980). 26. SHIODA, T., IWASAKI,K., and SHIBUTA, H., Nucleic Acids Res. 14, 1545-1563(1986). 27. BLUMBERG,B. M., GIORGE, C., ROSE, K., and KOLAKOFSKY,D., J. Gen. Viral. 60, 317-331 (1985). 28. PATERSON,R. G., HARRIS,T. J. R., and LYME, R. A. C., Proc. Nat/. Acad. Sci. USA 81, 6706-6710 (1984). 29. SPRIGGS,M. K.. OLMSTED, R. A., VENKATESAN,S., COLIGAN, J. E., and COLLINS, P. L., Virology 152, 241-251 (1986). 30. ROSE,J., ADAMS, G., and GALLIONE,C., Proc. Nat/. Acad. Sci. USA 81, 2050-2054 (1984). 31. STREINER,D. F., QUINN, P. S., CHAN, S. J., MARSK, J., and TAGER, D. F., Ann. N. Y. Acad. Sci. 343, 1- 16 (1980). 32. UNDENFRIEND,S., and KILPATRICK,D. L., Arch. Biochem. Biophys. 221,309-323 (1983). 33. COLLINS, P. L., HUANG, Y. T., and WERTZ. G. W.. Proc. Nat/. Acad. Sci. USA 81,7687-7691 (1984). 34. PORTER,A. C., BARBER.C., CAREY, N. H., HALLWELL,R. A., THRELFALL, G., and EMTAGE, J. S., Nature (London) 282, 471-477 (1979). 35. MORRISON,T., WARD, L., and SEMERJIAN,A., J. Viral. 53,85 l-857 (1985). 36. WARD, C. W., and DOPHEIDE,T. A. A., Virology 103, 37-53 (1980). 37. GETHING,M. J., BYE,J., SKEHEL,J. J., and WATERFIELD,M. D., Nature (London) 287, 301-306 (1980). 38. WATERFIELD,M. D., ESPELIE,K., ELDER, K., and SKEHEL,J. J., Brit. Med. Bull. 35, 57-63 (1979). 39. MIN-JOE, W., VERHOGEN,M., DAVOS. R., SAMAN, E., FANG, R., HUYLESROCK,D., FIEERS,W., THRELFALL,G.. CAREY,N., and EMTAGE, S. C., Cell 19, 683-696 (1980). 40. ORO~~LAN, S., HENDERSON, L. E., COPELAND, T. D., SCHULTZ, A. M., and RABIN, E. M., ln “Biosynthesis, Modification and Processing of Cellular and Viral Proteins” (G. Koch and D. Richter, Eds.), pp. 219-232. Academic Press, New York, 1980. 41. GAROFF,H., FRISCHANF,A. M., SIMONS,K., LEHRACH,H., and DELIUS, H.. Nature (London) 288, 236-241 (1980). 42. RICE, C. M., and STRAUSS,1. H.. Proc. Nat/. Acad. Sci. USA 178, 2062-2066 (1981). 43. BOSCH, F. X., GARTEN, W., KLENK, H-D., and Roar, R., Virology 113, 725-735 (1981). 44. GARTEN.W., BOSCH, F. X., LINDER, D., Roar, R., and KLENK, H-D., Virology 115, 361-374 (1981). 45. KOHAMA, T.. GARTEN, W., and KLENK, H-D., Virology 111, 364376 (1981). 46. RICHARDSON,C. D.. and CHOPPIN, P. W., Virology 131, 518-532 (1983). 47. VARSANYI,T. M., JORNVALL,H., and NORREY, E., Virology 147, llO-117(1985). 48. NAGAI, Y., and KLENK, H-D., Virology 77, 125-l 34 (1977). 49. NAGAI, Y., YOSHIDA, T., HAMAGUCHI, M., NARUSE, H., IINUMA, M., MAENO, K., and MATSUMOTO,T., Microbial. lmmunol. 24, 173177 (1980).