Sequence of the fusion protein gene of human parainfluenza type 2 virus and its 3′ intergenic region: Lack of small hydrophobic (SH) gene

VIROLOGY

178,2899292

(1990)

Sequence of the Fusion Protein Gene of Human Parainfluenza Type 2 Virus and Its 3’ intergenic Region: Lack of Small Hydrophobic (SH) Gene MITSUO KAwANo,*+ HISANORI BANDO,* SHINJI OHGIMOTO,* KUNIO KONDO,*‘t MASATO TSURUDOME,* MACHIKO NISHIO,* AND YASUHIKO /TO*.’ *Department of Microb/ology, Mie University School of Medicine, 2- 774 Edobashi, Tsu-Shi, Mie Prefecture and tfujlkura Research Center, Fujikura Kasei Co., Ltd., 3-20-7 Hasune, Itabashi, Tokyo, Japan

5 14,

Received October 30, 1989; accepted May 8, 1990 cDNA clones representing the fusion (F) gene of human parainfluenza virus type 2 (PIV-2) were isolated from cDNA libraries constructed from virus-specific mRNA and genomic RNA, and the complete nucleotide sequence of the F gene was determined. The F gene is 1854 nucleotides long and encodes one long open reading frame of 551 amino acids. The cleavage site for activation of the precursor F, protein is Thr-Arg-Gln-Lys-Arg. The F gene of PIV-2 is most closely related to those of simian virus 5 (SV5) and mumps virus (MuV). Interestingly, although the HN glycoprotein of PIV-2 shows no relatedness to the HA glycoprotein of measles virus (MV), a distinct homology is found in the F proteins of PIV-2 and MV. As concerns F proteins, paramyxoviruses can be divided into two subgroups; that is, PIV-2, SV5, and MuV belong to one group, and HPIV-1, SV, and PIV-3 belong to the other group. Newcastle disease virus (NDV) and MV are intermediate. Coding regions for small hydrophobic (SH) proteins have been found between the HN and F genes of SV5 and MuV, which are the viruses most closely related to PIV-2. However, such a gene could not be detected 0 1990AcademicPress. Inc. in two different strains of PIV-2.

Human parainfluenza type 2 virus (PIV-2) belongs to genus Paramyxovirus within the Paramyxoviridae and infects the respiratory tract of man occasionally, resulting In croup during early life (7). Although PIV-2 is one of important respiratory pathogens of man, little is known about its gene structures and primary amino acid sequences. Recently we have determined the nucleotide sequence of the PIV-2 HN gene and constructed a phylogenetic tree for HN proteins of all the paramyxoviruses that are infectious to human (2). However, a gene structure of another glycoprotein, F protein, remains unclear. In this study, we tried to analyze the F gene structure of PIV-2 and compared the structure with other paramyxoviruses. Virus propagation and purification of virus mRNA and nucleocapsid RNA were performed as previously reported (2, 3). The cDNA libraries derived from mRNA and genomic RNA were constructed according to the Okayama-Berg method (4) and to Gubler and Hoffmann (5) as described by Kawano eta/. (2). Sequencing was performed by the dideoxy chain termination method (6). We first screened the genomtc cDNA library using the 32P-labeled 5’terminal fragment of an HN cDNA clone (pKH272) reported previously (2). Two cDNA clones (pKH326G and pKH40G). which extended about 1000 nucleotides beyond the 5’-terminus of the

HN gene, were obtained. The partially sequenced data indicated that the deduced amino acid sequences showed apparent homologies with the F protein of SV 5 (data not shown), confirming that these cDNAs include a part of the PIV-2 F gene. The genomic cDNA library was further screened using the insert from pKH326G as a probe. Some clones extending into the M gene were obtained. The 5’-end of F mRNA was determined by primer extension. The complete nucleotide sequence of the PIV-2 F gene is shown in Fig. 1 in positive sense. The F gene is 1854 nucleotides long, excluding the poly(A) regron. Seven nucleottdes (5’AllTAAG-3’) preceding the six A residues at the terminus of F gene were considered to be a polyadenylation signal of PIV-2, because the same sequence was found in 3’-termini of HN (2) and NP genes (Yuasa et a/., manuscript submitted for publication) of PIV-2. The sequence contained a single large open reading frame with 1671 nucleotides, that is, 557 codons which started from AGG at nucleotides l-3 and terminated at TAA (nucleotides 1672- 1674). The first ATG codon is located at nucleotides 19-21 and the flanking sequence around the AUG codon is favorable for the actual initiation site of translation with eukaryotic ribosomes (7). Thus, a polypeptide with a molecular weight of 59,720 is deduced, although the molecular weight of PIV-2 F, glycoproteln has been estimated to be 72K on SDS-polyacrylamide gels (3). The discrepancy is probably explained by the glycosylation of the protein

’ To whom requests for reprmts should be addressed.

289

0042.6822:90 Copyright

$3.00

ii,’ ,990 byAcade!nfc

Press. Inc

290

SHORT COMMUNICATIONS -28

a5

1 CTAAACGTlCCACAATAAATCAACG~CAGGCCAAAATA~CAGCCATGCATCACCTGCATCCAATGATAGTATGCATC~G~ATGTACA~GGAA~GTAGG~CAGATG FVMYTG MHHLHPMIVCI

IVGSDA 205

CCATTG(TTGGAGATCAACTACTTAATITATAGGGGTCA~CAATCAAAGATAAGATCACTCATGTACTATACTGATGGTGGTGCTAGC~A~G~GTAAAA~GCTACCTAATC~CCCC IAGDQLLNIGVIQSKIRSLMYYTDGGASF

IVVKLLPNLPP 325

CAAGCAATGGAACATGCAACATCACCAGTCTAGATGCATATAATG~ACCCTA~AAG~ACTAACACCCCTGA~GAGAACCTGAGTA~A~CCACTG~ACAGATACCAAAACCC I E I-1 K I Sm[Cr-~SLDAY~[NLFKLLTPL . GCCAAAAACG~~GCAGGAGTAG~G~GGACTTCCTGCTGCA~AGGAGTAGCCACAGCCGCACAAATAACTGCAGCTGTAGCAATAGTGAAAGCTAATGCAAATGCTGCTGCGATAAACA QKRFAGVVVGLAALGVATAAQITAAVAIVKANANAAA

STVTDTKTR 44s I

N

N 565

ATCTTGCATCTTCAATTCAATCCACCAACAAGGCAGTATCCGATGTGATAGATGCATCAAGAACAA~GCAACCGCAG~CAAGCAA~CAGGATCGCATCAATGGAGCTA~G~AATG I P LASSIQSTNKAVSDVIDASRTIATAVQA

D

R

I

N

G

A

I

V

N

G 685

GGATAACATCTGCATCATGCCGTGCCCATGATGCACTCA~GGGTCAATA~AAATC~TATCTCACTGAGC~ACCACAATA~CATAATCAAATAACAAACCCTGCGCTGACACCAC ILNLYLTELTTIFHNQ ITSASCRAHDAL I G S

ITNPALTPL 805

TCTCCATCCAAGCmAAGAATCCTCCTCGGTAGCACCTTCCCAATTGTCATTGAGTCCAAACTCAACACAAACTTCAACACGGTCAAA SIQALRILLGSTLPIVIESKLNTNFNTAELLSSGLLTGQI 925 TAA~TCCATTTCCCCAATGTACATGCAAATGCTAA~CAAATCAATG~CCGACA~ATAATGCAACCCGGTGCGAAGGTAA~GATCTAA~GCTATCTCCGCAAACCATAAA~GC INVPTF IMQPGAKVIDL I s ISPYYMQWLIQ

IAISANHKLQ 1045

AAGAAGTGG~GTACAAGTTCCGAATACGAATAGGA~CTAGAGTATGCAAATGAACTACAAAA~ACCCAGCCAATGACTGTGTCGTGACACCGAACTCTGTA~TGTAGATACAATGAGGGTT EVVVQVPNR ILEYANELQNYPANDCVVTPNSVFCRYNEGS 1165 CCCCTATCCCTGAATCACAATATCAATGC~GAGGGGGAATC~AA~C~GCAC~ACCCCTA~ATCGGGAAC~C~AAGCGA~CGCA~GCTAATGGTGTGCTCTATGCCA IGNFLKRFAFANGVLYAN PIPESQYQCLRGNLNSCTFTPI 1285 ACTGCAAATCTTTGCTATGTAGGTGTGCCGACCCCCCCCATG~GTATCCCAGGATGATACCCAAGGCATCAGCATAATTGATATTAAGAGATGCTCTGAGATGATGC~GACAC~T IKRCSEMYLDTFS CKSLLCRCADPPHVVSQDDTClGISIID 1405 CATTTAGGATCACATCTACmCAATGCTACGTACGTGACAGAC~CTCAATGA~AATGCAAATA~GTACATCTAAGTCCTCTAGAT~GTCAAATCAAATCAA~CAATAAACAAAT FRITSTF[N~YVTDFSSINANNVHLSPLDLSNQ~NSI~~~ 1525 CTC~AAAAGTGCTGAGGATGGATTCCAGATAGCAAC~C~GCTAATCAAGCCAGGACAGCCAAGACAC~TATTCACTAAGTGCAATAGCA~AATACTATCAGTGA~AC~GG 4KSAEDl!ADSNFF?NQARTAKTLYSLSAI

A

L

I

LSVITLV 1645

~GTCGTGGGATTGCTGATGCCTACATCATCAAGCTGG~CTCAAATCCATCAA~CAGATCGCTAGCTGCTACAACAATG~CCACAGGGAAAATCCTGCC~C~CCAAGAATA V V G L L I A Y I IKLVSQIHQFRSLAATTWFHRENPAFFSKNN 1765 ACCAn;GAAACATATATGGGATATC~AAGAAATCTATCACAAGT~ATATATGTCCACAA~GACCC~AAGAACCAAC~CCAACGA~ATCCG~AAA~AAGTATAATAG~AA H G N 1 Y G I S 1860 AAA~AACATTAAGCCTCCAGATACCAATGAATATGAATATATCTC~AGAAAACCTGA~A~ATGTGATAGCGTAGTACAA~AAGAAAAAA

FIG. 1. The nucleotide sequence of the F gene in the positive sense and the predicted amino acid sequence. Potential asparagine-linked glycosylation sites are boxed. Underlines show major hydrophobic regions. (7) cleavage site for activation of the F, protein into mature Fl + 2. Dots indicate leucines and alanine in leucrne zipper-like motif.

in viva. Six potential asparagine-linked glycosylation sites (Asn-X-Ser/Thr) were identified in PIV-2 F, protein sequence (Fig. 1); four were found on F2 and two on Fl . Ito et al, (3) reported that when PIV-2-infected cells were labeled with [3H]glucosamine, the density of F2 was approximately twice as much as that of Fl , being consistent with the sequence data. Like the F proteins of other paramyxoviruses, the F protein of PIV-2 is highly hydrophobic and has three major hydrophobic regions which correspond to the signal, the amino terminus of Fl , and the membrane anchor (underlined in Fig. 1). The amino acid sequence deduced from the nucleotide sequence of the PIV-2 F gene was compared with those of other paramyxoviruses, SV5 (8) MuV (9) NDV (IO), PIV-3 (1 I), SV (12) HPIV-1 (73) and morbillivirus, MV (14) by homology dot matrix. PIV-2 showed extensive homology with SV5 and MuV, evident homology with NDV, and moderate homology with PIV-3, HPIV1, and SV. Interestingly, the PIV-2 F protein sequence was slightly more similar to MV than to HPIV-1, SV, and

PIV-3. In previous study (2) no sequence homology between PIV-2 HN and MV HA could be observed, whereas the homology between the PIV-2 HN and the HN of SV and PIV-3 was 21.7 and 22.9% respectively. To further analyze the relationships among these F proteins, the predicted amino acid sequence of PIV-2 F protein was aligned with those of seven other paramyxovirus F proteins (data not shown). The percentage of identifiable amino acid sequences between the F proteins was calculated (Table 1). As concerns F proteins, the paramyxoviruses can be divided into two subgroups: PIV-2, SV5, and MuV belong to one group; and HPIV-1, SV, and PIV-3 belong to the other group. NDV and MV are intermediate. No significant regions of amino acid sequence homology among all the viruses were found except at the amino terminus of Fl. However, a high conservation of cysteine residues is evident. Almost all the conserved Cys residues exist in the Fl domain, and only one is found in the F2 domain. The conserved Cys residue in the F2 domain must be involved in the disulfide bonding between the Fl and

291

SHORT COMMUNICATIONS TABLE 1 PERCENTAGE IDENTIP/ OF AMINO ACID SEQUENCES BETWEEN THE FUSION PROTEINS OF PARAMYXOVIRUSES

Virus

SV5

MuV

NDV

PIV-3

sv

HPIV-1

MV

PIV-2 sv5 MuV NDV PIV-3 sv HPIV-1

48.3

39.5 44.6

32.7 32.9 32.7

23.7 25.0 25.0 28.0

23.7 24.3 20.7 24.1 43.1

23.9 24.1 22.9 24.3 44.6 71.2

26.8 30.1 26.6 26.4 25.8 28.8 28.2

F2 domains, as suggested by Cote et al. (15). These findings suggest that the conformational structure is important for specific activities of the F protein. The position of potential glycosylation sites was conserved in the same subgroups of paramyxovirus, whereas potential glycosylation sites of MV showed no consetvation with paramyxoviruses. Furthermore, all the sites of MV were present in F2 domain. Hiebert eta/. (16) reported a previously unrecognized gene encoding a small hydrophobic (SH) protein of 44 amino acids, located between the F and HN genes of SV5. A gene encoding a similar small hydrophobic protein of 57 amino acids was also found between F and HN genes of MuV (17, 18). Since PIV-2 is most closely related with SV5 and MuV (79,20), it was expected that the SH-like gene would also exist in PIV-2. However, a sequence capable of encoding such a polypeptide was not detected between the terminal codon of the F gene and the initiation codon of the HN gene of PIV-2. Furthermore, a cDNA (about 1000 nucleotides) containing the noncoding region between the F and HN genes was unable to detect a small mRNA by Northern hybridization (data not shown). To confirm this conclusion, cDNA clones encoding a region between HN and F genes were isolated from another strain of PIV-2, strain 62-M786, and sequenced. The region of strain 62M786 showed 98% homology with strain Toshiba, and

TOSHIBA 62-M786

the SH-like gene could not be detected in the region (data not shown). Noncoding sequences of 3’.end of F and 5’-end of HN genes of PIV-2 are extraordinarily long as compared with MuV and SV5. Therefore, a sequence homologous to the SH gene was searched, with the aid of a computer, in these noncoding regions using SH gene sequence of MuV as a probe. Such a sequence was not found in the 3’-end of the F gene, while, surprisingly, one region (residues l-45) of MuV SH gene showed 80% identity with a region (residues 9-53) of PIV-2 HN gene and a stretch of 18 identical nucleotides was found in these sequences (Fig. 2). Furthermore, the intergenic sequence between the F and SH of MuV was identical with the sequence immediately downstream of the A tail on the PIV-2F gene except for 1 nucleotide. Alignment of the nucleotide sequence of the MuV HN gene with that of the PIV-2 HN gene indicates that the first nucleotide of MuV corresponds to nucleotide 85 of PIV-2 (Fig. 2). The PIV-2 coding frame begins at nucleotide 187 (2) and the sequence of PIV 2 showing homology with the MuV SH sequence does not contain any coding region of HN gene. This alignment was supported by alignment among PIV-2, MuV, and SV5. These findings indicate that the greater part of SH gene was deleted from PIV-2 during its evolution. The alternative possibility that a gene might have been inserted into MuV cannot be excluded. There is no homology between the SH proteins of MuV and SV5. Furthermore, the hydrophobic domain of SV5 SH protein is at the carboxyl terminus, whereas that of MuV SH protein is near the amino terminus. In addition, no SHlike genes have been found in paramyxoviruses other than SV5 and MuV. Therefore, the SH protein may not play an essential role in replication of paramyxoviruses. We failed to find any relatedness in the amino acid sequence of the HN/HA glycoprotein between PIV-2 and MV (2) whereas the other glycoprotein, the F protein, of PIV-2 showed a distinct homology with MV. Intriguingly, the homology between PIV-2 and MV is slightly higher than that between PIV-2 and some para-

C~AAAATAAGCACGAACCrrTAAGGTCTCGTAACGTCTCGTGACACCG-GG~CAG~CAAATATCGACCTCTAACCCAA~A . . . . . . . . . . . . . . . . . . . ..*............*.......... .,. . . . . . . . . . ..*............. C~AAAATAAGCA~GAA~CC~AAGGTGTCGTAACGT~~GTGACGC~G-GG~~AG~T~AAATAT~GA~~T~TAAT~~AA~A e......... . . . . . . . . . . . . . . . .. . . . . . . . . . . .

. . . . . . . .

ACACCCATTCITATATAAGAACACAGTATAA . . . . . . ..*.................*.... ACACCCATTCTTATATAAGAACACAGTATAA . . . . . . . .

. .

. .

HUV 31

GGCGGTTCGATATGCAGCACTGTACCAGCGATCCTGCTCTCGCTGGGG~GATCAATCACTCTAGAAAGATCCCCAGTT AGGACAAGACCCGATCCGTCACGCTAGAACAAGCTGCA~AAATGAAGCTGTGCTACCATGAGACATAAAGAAAAAA 316

FIG. 2. Alignment among SH gene of MuV and HN noncoding the intergenic sequence of MuV (30).

GC -

regions of PIV-2 (strains Toshiba and 62.M786) and MuV (18). Underlines show

292

SHORT COMMUNICATIONS

influenza viruses, PIV-3, HPIV-1, and SV. Virus receptors on cell membranes for parainfluenza virus are considered to be sialoconjugates, and sialidase activity is detected in the HN protein of parainfluenza viruses. On the contrary, the HA protein of MV does not bear sialidase activity and the cellular receptors for the HA protein contain no sialic acid (21). Such differences might exert different evolutionary restrictions on the polypeptides, resulting in structural divergence. The exact function of the F protein remains to be clarified; however, the F proteins of both parainfluenza virus and morbillivirus mediate fusion of lipid bilayers, which is essential for virus penetration and cell-to-cell spread of infection. Common evolutionary restriction may be present in fusion polypeptides due to their common function. Although primary structures of the amino acid sequence of parainfluenza virus F proteins show close relatedness to MV, no antigenic relationship was found between parainfluenza viruses and MV. This discrepancy may be partly explained by a lack of conservation of potential glycosylation sites between them. A leucine zipper motif, originally reported as a characteristic of a new category of DNA-binding proteins (22) was found in the F glycoproteins of paramyxoviruses (23). There is a similar motif in the region preced‘ing an anchorage domain of PIV-2, although alanine is substituted for the fourth leucine. The F proteins of paramyxoviruses appear to be dimers and/or tetramers. The SV F spike is composed of a noncovalent association of four homooligomers, each consisting of two peptides, Fl and F2, linked by a disulfide bond (24). The tetrameric form of the SV F protein in its native form consists of two identical dimers (24). Respiratory syncytial virus F proteins were reported to be dimers (25). The oligomer formation of paramyxovirus F proteins may be mediated at least in part by the leucine zippers. The F, protein of PIV-2 is cleaved by either cellular protease or exogenous protease to produce the mature Fl + 2 protein. The sequence at the C-terminus of F2 is Thr-Arg-Gln-Lys-Arg (underline shows basic amino acid) which is very similar to the connecting peptide of PIV-3 (11). Pathogenicity has been correlated with the cleavability of the fusion protein (26) and the cleavability is related to the sequence of the connecting peptide (27). The connecting peptide of SV5 is composed of five basic amino acids (Arg-Arg-Arg-ArgArg) and that of MuV consists of two dibasic amino acids (Arg-Arg-His-Lys-Arg). SV5 and MuV cannot be isolated with uncleaved F protein. In contrast, SV has only a single Arg residue at the cleavage site (12) and is often isolated with uncleaved F protein from cer-

tain cells (28, 29). Although PIV-2 is considered to be avirulent, the cleavage site sequence, unexpectedly, belongs to the virulent type.

REFERENCES 1. KINGSBURY,D. W., BRATS, M. A., CHOPPIN, P. W., HANSON, R. P., HOSAKA, Y., TER MEULEN, V., NORRBY,E., PLOWRIGHT,W., Rorr, R., and WUNNER, W. H., lntervirology 10, 137-l 52 (1978). 2. KAWANO, M., BANDO, H., YUASA, T., KONDO, K., TSURUDOME, M., KOMADA, H., NISHIO, M., and ITO, Y., L/iro/ogy 174, 308-313 (1990). 3. ITO, Y., TSURUDOME,M., and HISHIYAMA, M., Arch. Viral. 95,2 1 l224 (1987). 4. OKAYAMA, H., and BERG, P. A., Mol. Ce//. Biol, 2, 161-l 70 (1982). 5. GUBLER, U., and HOFFMANN, B. J., Gene 25, 263-269 (1983). 6. SANGER, F., NICKLEN, S., and COULSON, A. R.. Proc. Nat/. Acad. Sci. USA 74,5463-5467 (1977). 7. KOZAK, M., Cell44, 283-292 (1986). 8. PATERSON, R. G., HARRIS, T. J. R., and LAMB, R. A., Proc. Nat/. Acad.%. USA81,6706-6710(1984). 9. WAXHAM, M. N., SERVER,A. C., GOODMAN, H. M., and WOLINSKY. J. S.. Wology 159, 381-388 (1987). 10. CHAMBERS, P., MILLAR, N. S., and EMMERSON, P. T., 1. Gen. Viral. 67,2685-2694 (1986). 11. SPRIGGS,M. K., OLMSTED, R. A., VENKATESAN,S., COLIGAN, J. E., ~~~COLLINS, P. L., virologyl52, 241-251 (1986). 12. BLUMBERG, B. M., GIORGI, C., ROSE, K., and KOLAKOFSKY,D., J. Gen. L&o/. 66, 3 17-331 (1985). 13. MERSON, 1. R., HULL, R. A., ESTES,M. K., and KASEL,J. A., virology 167,97-105 (1988). 14. RICHARDSON,C., HULL, D., GREER, P., HASEL, K., BERKOVICH,A., ENGLUND, G., BELLINI, W., RIMA, B., and LPZZARINI,R., virology 155,508~523(1986). 15. COTE, M.-J., STOREY. D. G., KANG, C. Y., and DIMOCK, K., J. Gen. L&o/. 68, 1003-l 010 (1987). 16. HIEBERT. S. W., PATERSON, R. G., and LAMB, R. A., J. Vifol. 55, 744-751 (1985). 17. ELLIOTT, G. D., AFZAL, M. A., MARTIN, S. J., and RIMA, B. K., virus Res. 12, 61-75 (1989). 18. ELANGO, N., KOVAMEES,J., VARSANYI, T. M., and NORRBY, E., J. Vifol.63, 1413-1415(1989). 19. ITO, Y.. TSURUDOME, M., HISHIYAMA, M., and YAMADA, A., J. Gen. Viral. 68, 1289-l 297 (1987). 20. TSURUDOME, M., NISHIO, M., KOMADA, H., BANDO, H., and ITO, Y., lhology 171, 38-48 (1989). 21. KRAH. D. L.. Lbfology 172, 386-390 (1989). 22. LANDSCHULZ, W. H., JOHNSON, P. F., and MCKNIGHT, S. L., Science 240,1759-l 764 (1988). 23. BUCKLAND, R., and WILD, F.. Nature (London) 338,547 (1989). 24. SECHOY, O., PHILIPPOT,1. R., and BIENVENUE,A., /. Biol. Chem. 262, 11,519-l 1,523 (1987). 25. ARUMUGHAM, R. G., HILDRETH, S. W., and PARADISO, P. R., Arch. Viral. 106, 327-334 (1989). 26. NAGAI, Y., KLENK, H.-D., and Roar, R., Virology 72, 494-508 (1976). 27. TOYODA, T., SAKAGUCHI,T., IMAI, K., INOCENCIO,N. M., GOTOH, B., HAMAGUCHI, M., and NAGAI, Y., !/ifo/ogy 158, 242-247 (1987). 28. HONMA, M.. and OHUCHI, M../. viral. 12, 1457-1465 (1973). 29. SCHEID, A., and CHOPPIN, P. W., virology57, 475-490 (1974). 30. ELANGO. N., VARSANYI, T. M., KOVAMEES,J., and NORRBY, E., J. Gen. viral. 69, 2893-2900 (1988).