key: cord-0689885-if715s0q
authors: YAMADA, YASUKO K.; TAKIMOTO, KAZUHIRO; YABE, MIKIKO; TAGUCHI, FUMIHIRO
title: Acquired Fusion Activity of a Murine Coronavirus MHV-2 Variant with Mutations in the Proteolytic Cleavage Site and the Signal Sequence of the S Protein
date: 1997-01-06
journal: Virology
DOI: 10.1006/viro.1996.8313
sha: 503e62e833ac68ac12fc1007d8acb8e22f1221d8
doc_id: 689885
cord_uid: if715s0q

Abstract The spike (S) protein of a nonfusogenic murine coronavirus, MHV-2, was compared to the S protein of a variant with fusion activity, MHV-2f. Two amino acids differed between the S proteins of these viruses; one was located in the signal sequence and the other was in the putative cleavage site. The amino acid at position 12 in the signal sequence was S in MHV-2 and C in MHV-2f. The amino acid sequence of the cleavage site of MHV-2 was HRARS, while that of MHV-2f was HRARR, showing one amino acid replacement at position 757. In DBT cells infected with MHV-2, the S protein was not cleaved, while the S protein of MHV-2f was cleaved. The S protein of MHV-2f expressed in a transient vaccinia virus expression system was cleaved and was fusogenic in contrast to the nonfusogenic activity of uncleaved MHV-2 S protein. Because the signal sequence is assumed to be removed from the mature S protein soon after synthesis, and because the S protein of MHV-2 was expressed on the cell surface in the same way as the S protein of MHV-2f, the difference in the signal sequence seemed to have had little effect on the transportation and the fusion activity of the S protein. These results showed that MHV-2 does not fuse cells due to the lack of cleavage of its S protein. This conclusion differs from studies on the activity of syncytium formation by the S proteins of fusogenic MHV-JHM and -A59 strains. Possible reasons for these differences in fusion activity are discussed.

though these fusion peptides characteristically have a single-stranded, positive-sense genomic RNA of about hydrophobic amino acid cluster, such a cluster is not 31 kb (1) (2) (3) (4) . The spike (S) protein of MHV is a transmemfound in the N terminus of the membrane-anchored S2 brane glycoprotein of about 180 kDa and is cleaved by subunit of coronavirus S protein (8) . Studies of mutants trypsin-like enzymes that are derived from the host cell MHV-JHMV and -A59 and an MHV-A59 variant, whose S into two 90-kDa subunits (5) . The N-terminal S1 subunit proteins were not cleaved due to amino acid replaceforms the bulbous head and the C-terminal S2 subunit ments in the cleavage site, suggested that cleavage of is anchored in the virion envelope (4) . The S protein the S protein may not be a prerequisite for syncytium mediates attachment to cells and induces syncytium in formation, although it facilitates that activity (9) (10) (11) (12) . fusogenic MHV strains (5, 6) . Whether or not S protein Most MHV strains produce cell-to-cell fusion on culcleavage is required for syncytium formation is still contured DBT cells forming syncytia. MHV-2 is the only strain troversial. The treatment with exogenous protease enwhich does not induce syncytia (13, 14) . However, we hanced syncytium formation by MHV (5) and the treatnoticed that fusion-type MHV-2 variants were present in ment with protease inhibitors causes a delay in the onset our MHV-2 stock virus and we isolated a fusion-type of fusion (7) . These data suggested that the cleavage of MHV-2 variant by plaque purification. We compared futhe S protein may be important for fusion activity in much sion-negative MHV-2 and its fusion-positive variant with the same way that protein cleavage acts in other fusorespect to cleavage of the S protein and transportation genic viruses, such as orthomyxoviruses, paramyxoviof the S protein to the cell surface. These comparisons ruses, and retroviruses; in those viruses, cleavage of the suggested that fusion activity of the MHV-2 depends surface glycoprotein exposes the fusion peptide on the upon cleavage of its S protein. N terminus of the membrane-anchored subunit (8) . Al-A fusion-positive variant of MHV-2 was obtained using plaque cloning. About 3-5 1 10 4 PFU of the MHV-2 stock virus were layered on DBT cells growing in a 15-cm dish, proximately 1 plaque of 1-2 1 10 6 parental plaques were mixed with 2.5 U Takara Ex Taq (Takara Biochemicals), 0.25 mM each dATP, dCTP, dTTP, and dGTP showed a fusion-type plaque. The fusion-type plaques were isolated and plaque-purified three times. The final (dNTPs), and 0.2 mM each sense primer, 5-CGCAAG-CTTCTAAACATGCTATTCGTGT-3 [the MHV-2 S gene plaque-purified isolate was designated MHV-2f. By using a panel of monoclonal antibodies against the S protein around the initiation codon (17) and its attached HindIII site], and complementary primer, 5-CGGGATCCAGGA-of JHMV, we confirmed that MHV-2f was closely related to parental MHV-2 and was not a fusion-type contaminant GAGGCTGTGATAGTCA-3 [the S gene around the stop codon (18) and its attached BamHI site], in a total volume from another MHV strain (data not shown).

To compare cytopathic effects and virus growth, DBT of 50 ml Ex Taq buffer. The mixture was amplified for 30 cycles of denaturation at 94Њ for 0.5 min, annealing at cells were inoculated with MHV-2 or -2f at a multiplicity of 1 to 3. Virus titers in the culture fluids were deter-60Њ for 1 min, and elongation at 72Њ for 3.5 min, and the amplified samples were electrophoresed on a 0.5% mined by plaque assay as previously reported (15) . No substantial difference in the growth of these viruses agarose gel. Bands of about 4 kb were clearly amplified from MHV-2 and -2f, and were purified from agarose was observed. The progeny viruses were detected from 6 hr postinfection (p.i.) after which their titers increased gel by the Prep-A-Gene DNA purification kit (Bio-Rad). Sequence analysis of PCR products was performed by gradually, reaching a plateau at 12 hr p.i. to 0.8 -3 1 10 7 PFU/ml (data not shown). Infected cells were also a dideoxy termination labeling method according to the manufacturer's instructions (Applied Biosystems; Model observed for cytopathic effect and expression of viral antigen. At 7 hr p.i., syncytium formation was observed, 373A-18 DNA sequencing system). Sequencing oligonucleotide primers were synthesized to fit every 300 to 400 and viral antigen was detected in the cytoplasm in MHV-2f-infected cells (Fig. 1A) . In contrast, parental bases of the MHV-2 S gene in both directions of genomic and complementary sense. The deduced amino acid se-MHV-2 had not induced syncytia by 24 hr, although viral antigen was detected in the cytoplasm of infected cells quence of MHV-2 showed that the MHV-2 S protein was composed of 1361 amino acids. The length of the S gene (Fig. 1B) . Viral antigen was not detected in uninfected DBT cells (Fig. 1C) .

is assumed to vary among the MHV strains. In MHV-JHMV, several types are reported; JHMV cl-2 is known The S protein is responsible for syncytium formation of MHV infected cells (5, 6, (9) (10) (11) (12) ; therefore, we com-to have a long S gene (18) . The amino acid sequence of MHV-2 was compared with the published amino acid pared the amino acid sequences of the S proteins of nonfusogenic MHV-2 and fusogenic MHV-2f. The full-sequence of JHMV cl-2 (schematically shown in Fig. 2 ). Fifteen amino acids were deleted from the corresponding length coding region of the S protein was amplified using PCR. At 16 hr p.i., total cellular RNA was extracted from region of cl-2. Among these deleted amino acids, 12 amino acids were located in a hypervariable region in the infected DBT cells and 10 mg of RNA was reverse transcribed into cDNA using oligo(dT) as a primer, as the S1 subunit where a nucleotide deletion occurs in many MHV strains (19, 20) . Three additional amino acids described previously (16) . The cDNA transcripts (5 ml) data showed that the difference in the fusogenicity of MHV-2 and -2f depended on the S protein.

The first amino acid replacement was located in the signal sequence, which is reported to be composed of 20 amino acids from the N terminus of the S protein and to be removed from the S protein soon after synthesis (22) . A possibility exists that the amino acid difference in the signal sequence can influence the transportation of the S protein to the cell surface. To determine the cellsurface expression of the MHV-2 S protein, live monolayers of MHV-2-and -2f-infected cells were reacted with S protein-specific monoclonal antibody at 4Њ for 1 hr, followed by FITC-labeled anti-mouse Ig's (TAGO). Granular positive dots were detected on the cell surface at 7 and 24 hr after MHV-2 infection, and by 7 hr after MHV- the syncytium, demonstrating that both S proteins were cited from published data (18, 23, 22) . The deletion is indicated by thin transported to the cell surface (data not shown).

lines.

The second amino acid replacement was located in the putative cleavage site. This mutation could influence the cleavage of the S protein, since a host cell-derived were deleted from the S2 subunit. The overall S genesequence homology between MHV-2 and cl-2 was proteolytic enzyme is reported to recognize the amino acid sequence RRARR in this position for JHMV and 82.23%, excluding the deleted sequence. Comparison of the nucleotide sequences of MHV-2 and -2f revealed only 2 nucleotide replacements (35, C to G and 2271, C to A). Both of these changes led to the replacement of the predicted amino acids. One was the 12th amino acid from the initiation codon which is located in the signal sequence (Fig. 3) . The amino acid at position 12 in MHV-2 was S and in MHV-2f was C. The other replacement was located in the basic amino acid cluster of the cleavage site (Fig. 3) . The amino acid at position 757 was S in MHV-2 and R in MHV-2f. To compare these two positions among various MHV strains, DNA fragments including the signal sequence region and the putative cleavage site were amplified. Sequence analysis of the PCR product showed that MHV-2 had unique amino acids at posi- of serine with arginine at position 757 appeared to influence the cleavage of the S protein. As expected, the MHV-2f S protein was cleaved, in contrast to the cleavage resistance of nonfusogenic MHV-2 S protein. The other mutation at position 12 was located in the signal sequence that is considered to be removed from the mature S protein soon after synthesis. Transportation of the S protein to the cell surface was detectable in both MHV-2-and -2f-infected cells, indicating that this mutation in the signal sequence had little effect on transporta- of the S1 subunit (24). The transmembrane region (12) and a predicted heptad repeat region of the S2 subunit (25) are also proposed to influence the fusion activity. RRAHR for A59 (23, 22) . To determine whether the S proteins of MHV-2 and -2f are cleaved, we performed These regions were identical in fusogenic MHV-2f and nonfusogenic MHV-2, and therefore are unlikely to have Western blotting analysis using cell lysates infected with MHV-2, -2f, and JHMV cl-2 (Fig. 4A) . Cell lysates collected affected the fusogenic activities of MHV-2 and -2f. In fusogenic strains JHMV and A59, cleavage of the S 18 hr p.i. were used in Western blotting, and the S protein was detected with S2-specific monoclonal antibody protein is not absolutely required, because the S proteins of these viruses with mutations in the putative cleavage (kindly provided by Dr. S. Siddell). Two major bands at 150 and 180 kDa were detected in the lysate of MHV-site can induce syncytium, yet are not cleaved (9) (10) (11) (12) . We describe here that the S protein of MHV-2 failed to 2-infected cells. The 150-kDa protein may represent a nonglycosylated form of the mature 180-kDa protein that induce fusion due to a lack of cleavage. Perhaps the S proteins of fusogenic MHV strains induce fusion without is observed in the MHV-A59 S protein (12) . In contrast, the 150-and 90-kDa bands were detected in the lysates cleavage of the S protein. In contrast, the MHV-2 S protein required cleavage for activation of its fusion ability. of MHV-2f-and JHMV cl-2-infected cells. Most of the mature 180-kDa S protein was cleaved to S1 and S2

Cleavage of the MHV-2 S protein may alter its conformation thereby activating the otherwise nonfusogenic char-subunits which were detected as the 90-kDa band. In MHV-2-infected cells, a 90-kDa band was not detected, acter of this protein. Such a process would be very similar to that of other fusogenic orthomyxo-, paramyxo-, and revealing that the MHV-2 S protein was not cleaved. Faint bands were seen just above the location of the S2 band retroviruses (8) . In these viruses, the newly appearing hydrophobic N terminal region of the membrane-an-in MHV-2-, -2f-, and cl-2-infected cells; these bands probably do not represent S protein cleavage products, be-chored subunit is postulated to work as a fusion peptide.

In the N terminus of the MHV-2 S2 subunit, however, a cause they are absent from the S proteins expressed in the vaccinia virus system, as shown in Fig. 4B . These similar fusion peptide with a stretch of apolar amino acids, containing mainly alanine and glycine, was not results showed that the replacement of amino acid 757 changed the cleavability of the MHV-2 S protein.

found. We speculated that the region affecting fusion activity is possibly located inside S2, as is postulated for We have investigated differences between nonfusogenic MHV-2 and its fusogenic variant MHV-2f, looking Semliki Forest, Sindbis, and Rous sarcoma viruses (8), because fusion activity is thought not to reside in the N specifically at differences in fusion activity and in the characteristics of the S protein, which is known to medi-terminus of the MHV S protein (14). The hypothesis that the MHV-2 S protein is inactive with respect to fusion ate fusion in MHV. Sequence analysis showed that two amino acids at positions 12 and 757 from the N terminus and that the S proteins of A59 and JHMV are actively fusogenic without cleavage could be tested by using chi-of the S protein differed between the two viruses. The mutation found at position 757 was located at the putative meric S proteins. Fusion chimeras of the S protein could also help identify the region responsible for fusion activity cleavage site, which is reported to be digested with a trypsin-like host cell-derived enzyme. The replacement on the S protein.

''Fields Virology

Proc. Natl

Van der