key: cord-0877436-3v5ebbio
authors: Yoshikura, H.; Tejima, S.
title: Role of protease in mouse hepatitis virus-induced cell fusion Studies with a cold-sensitive mutant isolated from a persistent infection
date: 1981-09-30
journal: Virology
DOI: 10.1016/0042-6822(81)90178-1
sha: 79f669ea3c9705106a2add7bca7f7f9382c5fd31
doc_id: 877436
cord_uid: 3v5ebbio

Abstract A plaque mutant was isolated from Kirsten mouse sarcoma virus-transformed BALB/c cells persistently infected with a mouse hepatitis virus strain MHV-S. While the wild type produced large plaques consisting of fused cells (fusion type) both at 39 and at 32°, the mutant produced small fusion-type plaques at 39°, and at 32°, only after longer incubation, plaques consisting of round dead cells (non-fusion type) were obtained. The mutant grew equally well at both temperatures. Thus, the mutant was cold sensitive in inducing cell fusion, but not in replication or in ultimate cell killing. The cold sensitivity was overcome by the addition of trypsin (0.04 to 0.05%) to the culture medium, but not by treating the virions with trypsin. SDS-polyacrylamide gel electrophoresis of the virion proteins failed to detect differences between the wild type and the mutant. In intracellular viral proteins immunoprecipitated with anti-MHV-S rabbit serum, a protein of 68,000 daltons (68K protein) which was present in the wild type-infected cells was absent in the mutant-infected cells, but trypsin treatment or incubation at 39° of the mutant-infected cells failed to induce the 68K protein. After six to seven undiluted passages, the mutant segregated varieties of mutants which were partially or totally reverted to the wild type in phenotype, and also those whose cell fusion induction was absent even at 39°. All these mutants failed to induce the 68K protein in the infected cells. Thus, there was no linkage between the presence of 68K protein and the fusion induction. Trypsin treatment of the infected cells enabled MHV-S to form fusion plaques on otherwise resistant cells, and MHV-2, a producer of non-fusion-type plaques, to form fusion-type plaques. Protease appears to play an important role in MHV-induced cell fusion in general.

Mouse hepatitis virus group, a coronavirus, is one of the widespread endemic viruses in laboratory mouse colonies (Rowe et al, 1963 , McIntosh, 1974 Fujiwara et aL, 1976) . Strain MHV-S isolated by Rowe et al. 1963) induces plaques in SRCDFl DBT cells (Hirano et al., 1974) which develop by fusion of the infected cells. We previously reported that Kirsten mouse sarcoma vi-maintained at 37" for nearly 2 years. Recently, we found that MHV-S released by the culture was a cold-sensitive plaque mutant. Characterization of this mutant is detailed in this report. In addition, our experiments suggested an involvement of proteolytic activities in plaque formation by MHV in general. MATERIALS AND METHODS rus-transformed BALB/c mouse cells (Ki-BALB cells) were able to be dually infected Vim. MHV-S and MHV-2 were kindly with MHV-S and mouse leukemia virus provided by Dr. K. Fujiwara of this insti- (Yoshikura and Taguchi, 1979) . The Ki-tute. Under ordinary conditions, MHV-S BALB cells producing both viruses were produced fusion-type plaques, and MHV-2 produced plaques consisting of round dead cells (<on-fusion type). The viruses cells in the manner described by Hirano luted in the virus standard buffer, layered et al. (1974) . over a 20-60s (w/v) linear sucrose gra-Cells. Ki-BALB cells (Aaronson and dient, and centrifuged in a Spinco SW27 Weaver, 1971) were obtained from DR. Y. rotor for 20 hr at 24,000 rpm (4"). Gradient Ikawa of the Cancer Institute of Japan. fractions with a buoyant density of about The cells were persistently infected with 1.17 g/cm3 were collected for analysis of MHV-S and FN-2, a helper component of virion proteins. Friend leukemia virus, and maintained as Radioisotope labeling. Cells were labeled persistent infections for nearly 2 years. for 3 hr with 50 &i/ml of [35S]methionine The cells were incubated at 37" and sub-(1140 Ci/mmol, Radiochemical Center) cultured when they became confluent, dissolved in methionine-free medium supabout twice a week. For titrating and plemented with 2.5% dialyzed calf serum. propagating MHV, SRCDFi DBT cells At the end of the labeling period, the cells which were not producing type-C virions were lysed, scraped into extraction buffer were used. The FV131 cell line was a ddY (1% Triton X-100; 0.5% deoxycholate; 0.1% mouse-derived lung cell line persistently SDS; 0.01 M NaHzP04-NazHPOd, pH 7.5; infected with Friend leukemia virus and 0.1 MNaCl), and centrifuged at 30,000 rpm was of untransformed morphology (Yosh-for 30 min in a Spinco SW50.1 rotor. The ikura et aZ., 1968).

supernatnat was used as cell lysate. Titration.

SRCDFl DBT cell cultures Immunoprecipitation and SDS-polggrown in 30-mm petri dishes were con-acrylamide gel ekctrophoresis.

The lysate tacted with the virus in a volume of 0.1 ml in a volume of 0.5 ml was mixed with 10 for 1 hr. After the inoculum was removed, ~1 of normal rabbit serum and 50 ~1 of 10% the cultures were overlaid with a medium formalin-fixed S. aureus. After 16 hr of containing 0.8% methyl cellulose, 4000 cps incubation, the mixture was centrifuged (Wako Pure Chemical Industries, Ltd.). at 2500 rpm for 2 min, and the supernatant Plaques were scored 24 to 40 hr after in-was incubated with 5 ~1 of anti-MHV-S fection in formalin-fixed and crystal vi-rabbit serum (gift of Dr. F. Taguchi) for olet-stained cultures. The culture medium 2 hr. S. aurem in a volume of 50 ~1 was consisted of Eagle's MEM (Nissui Co.) sup-added and incubated for a further 15 min. plemented with 7.5% calf serum heated at The pellets were washed three times with 56" for 30 min. extraction buffer and suspended for SDS-Chemicals. Trypsn (1:250) and trypsin polyacrylamide gel electrophoresis in 1% inhibitors lyophilized from soybean were SDS, 1% 2-mercaptoethanol, 50 n&f Tris, purchased from Difco laboratories and pH 6.8, and 10% glycerol. SDS-polyacrylfrom Sigma Chemical Company, respec-amide gels wer.e prepared and run as detively. Other protease inhibitors, antipain, scribed by Laemmli (Laemmli, 1970 i., about 0.1) was clarified by were found to produce exclusively MHVcentrifugation at 7000 rpm for 10 min. The S, forming small plaques when titrated in virions were precipitated by ammonium SRCDFl DBT cells at 37". No wild-type sulfate (313 g/liter). The precipitate was large plaques were produced. As persissuspended in virus standard buffer (50 tent infection was frequently established mMTris-HCl, pH 7.5; 100 mMNaC1; 1 mM by temperature-sensitive mutants (Preble EDTA) and sedimented to an interface of and Younger, 1973; Rima and Martin, 1976 ; 60% sucrose (w/v) (15 ml) and 20% su-Walker, 1964), we titrated the supernatant crose (w/v) (5 ml). The viral band, col-at 32,37, and 39". At 32", plaques were not lected from above to avoid pellet, was di-produced at all, while at 39", plaque for-mation was as efficient as at 37". Thus, persistently infected cultures appeared to produce an MHV-S mutant whose plaque formation was cold sensitive. Before continuing experiments, we cloned the viruses by two successive plaque purifications; Sp-11, a mutant, and 6v-1, a wild-type clone, were used in further experiments.

Plaque fmatims of &I-11 and 6v-1 at varous temperatures.

While the plaquing efficiency of the wild-type 6v-1 was comparable at 39 and at 32" (11 X lo5 and 12 x lo5 PFU/ml, respectively), Sp-11, having a titer of 1.2 X lo4 PFU/ml at 39" and of 3.5 X 104 PFU/ml at 37", formed no plaques at all at 32". In this assay, the plaques were scored 24 hr after infection. If the Sp-11-infected cultures were kept longer at 32", the cells remained apparently healthy for the first 4 days (viable cells of cultures infected at a m.o.i. of 0.1 and maintained in liquid medium remained constant in number in this period), but at 5 to 6 days, non-fusion-type plaques consisting of round dead cells appeared. Cell fusion was never induced by Sp-11 at 32". Growth of Sp-11 as well as 6v-1 was comparable at 39 and 32" (Fig. 1) . Thus, the cold sensitivity of the mutant affected only fusion activity and not ultimate cell killing or viral replication.

Effect of protease on Sp-11 plaque fcrrmath.

Trypsinization of Sp-11-infected cells kept at 32' resulted in rapid cell fusion followed by cell lysis. The cold sensitivity of Sp-11-induced cell fusion appeared to be overcome by proteases. After. adsorption, the cultures were overlaid with methyl cellulose medium containing trypsin (Difco, 1:250) at various concentrations. As shown in Table 1 , trypsin at a concentration of 0.05% completely restored the plaquing ability of Sp-11 at 32". At 39", Sp-11 in the presence of trypsin gave plaques as large as 6v-1 (Fig. 2) . The experiment suggested that fusion induction by Sp-11 was dependent upon proteoiytic activity in the culture. For confirmation, effects of soybean trypsin inhibitor were examined. Plaque induction by trypsin (0.04%) was effectively inhibited by the trypsin inhibitor.

In addition, cell fusion by 6v-1 in the absence of trypsin was reduced by the agent (Table 2) . Other protease inhibitors (Yuasa et d, 19'75) , such as pepstatin, chymostatin, and leupeptin, reduced the cell fusion; antipain, elestatinal, and /3-glucaro-a-lactam were poor in the activity (data not shown). Intracellular virus-coded proteins. The effect of trypsin was examined in an infectious center assay of Sp-11-infected cells kept at 32". The infected cells were treated with trypsin (0.5% for 5 min), washed twice, incubated with soybean trypsin inhibitor at a concentration of 100 pg/ml for 5 min, washed once, and plated onto the untreated uninfected cell monolayer; massive cell fusion was observed even at 32". Trypsin treatment of only uninfected cells failed to induce cell fusion. Thus, the continuous presence of trypsin was unnecessary for fusion induction, and once surface protein(s) of the Sp-ll-producing cells was cleaved by the protease, the infected cells were capable of fusing with untreated surrounding uninfected cells. Therefore, we examined the effect of trypsin on MHV-coded protein in the infected cells.

The cells were infected with Sp-11 or 6v-1 at a m.o.i. of 0.1. Sp-11-infected cells were incubated for 16 or 24 hr (1) at 32" in the absence of trypsin, (2) at 32" in the presence of trypsin (0.04%), or (3) at 39" in the absence of trypsin. The Gv-l-infected cells were incubated for 16 hr at 39 or at 32" in the absence of trypsin. The cells were then labeled with [??,Imethionine for 3 hr under the respective conditions. The cell lysates were immunoprecipitated with anti-MHV-S serum or with control normal rabbit serum. As shown in Fig. 3 , all the preparations precipitated with the antiserum showed two major bands, A and C, which appeared to correspond to 130,000-and 60,000-dalton proteins in JHM or A59 MHV-infected cells (Bond et aZ., 1979) . Band B of 68,000 daltons (68K protein), appearing in 6v-linfected cells, was absent in Sp-11-infected cells. However, incubation at 39" (slot 5) or treatment with trypsin (slots 2 and 4) of Sp-11-infected cells did not induce the band.

Studies with variants derived from Sp-11. After six to seven undiluted passages of Sp-11 in SRCDFr DBT cells, fusion-type plaques were produced sporadically even at 32". Plaques were isolated for further characterization.

In Table 4 , the properties of some representative isolates are listed. T4a, T4c, T3e, and T3f were plaque- (1) Infected with Sp-11 for 24 hr at 32' without trypsin.

(2) Infected with Sp-11 for 24 hr at 32" with trypsin.

(3) Infected with Sp-11 for 16 hr at 32" without trypsin.

(4) Infected with Sp-11 for 16 hr at 32" with trypsin (0.04%).

(5) Infected with Sp-11 for 16 hr at 39". (6) Infected with 6v-1 for 16 hr at 32". (7) Infected with 6v-1 for 16 hr at 39'. Samples were electrophoresed on 12% polyacrylamide gels. ' Plaque phenotype was determined 48 hr after infection. nf, non-fusion-type plaque; sf, small fusiontype plaque; LF, large fusion-type plaque; np, no plaques detected. If non-fusion-type plaques were minute, such plaques could not be detected in the assay condition.

* Plaque titer obtained at 32"/plaque titer obtained at 39". isolated at 37" in the presence of 0.05% trypsin, and L2b, L2c, and L2d at 32" in the absence of trypsin. T4a and T4c were devoid of cell fusion-inducing activity even at 39", while T3e and T3f were close to the parental Sp-11. Lp2b and Lp2d produced small fusion plaques at 32" and relatively large plaques (diameter was about half of that of the wild-type plaques) at 39". Though Lp2b plated equally well at both temperatures, Lp2d plated at a reduced efficiency at 32". Lp2c was completely reverted to the wild type in phenotype. intracellular viral proteins were analyzed as described above (Fig. 4) . None of the variant-infected cells had the 68K protein which was present in the wild type. Thus, the role of the 68K protein in the cell fusion was excluded in this experiment also. In this gel, band C in Fig. 3 was separated into two bands, Cl and C2. Interestingly, T3e-infected cells were free of detectable band Cl.

Vi&n proteins. Proteins in purified virions were compared by SDS-polyacrylamide gel electrophoresis (Fig. 5) . When the culture fluids were harvested, the wild type-infected cultures were full of fused cells, while the Sp-11-infected cultures were apparently healthy. The virus was purified either from culture fluids collected every 30 min with the autoharvester (Bellco Glass Inc.) for 16 hr (Fig. 5A) or from culture fluids contacted with the infected cells for 16 hr after infection (Fig.  5B ). Five to six major bands, 1 (EOK), 2 (lOOK), 3 (90K), 4 (68K), 5 (55K), and 6 (35K), were detected in both wild type and mutant. Minor bands between bands 2 and 3 were variable from one preparation to another. From molecular weight, bands 1, 3, 4, 5, and 6 may correspond to GP180, GP90, GP60, VP50, and GP38 in A59 strain (Sturman, 1977) . Band '7 may correspond to GP23 (Sturman, 1977) or a 20,000-dalton protein (Bond et aZ., 1979) in A59 strain.

Trypsin digestion of purified virions resulted in reduction of bands 1, 2, and 6 to generate more intense band 3 and band D. Bands 4 and 5 were relatively unaffected by the digestion. The trypsin digestion pattern was the same in both wild type and mutant. Conversion of GP180 to GP90 by trypsin has been reported by Sturman and Holmes (1977) . These studies with virion proteins failed to detect any difference between the mutant and the wild type both in polypeptide composition and in susceptibility of the polypeptides to trypsin. Involvement of proteolysis in MHV-induced cell fmiqn in other cases. FV131 cells normally do not develop cytopathic changes after MHV-S infection. However, in the presence of 0.05% trypsin, the cells infected with 6v-1 or with Sp-11 developed fusion-type plaques to give fairly high titers. Viral replication was comparable in trypsin-treated and untreated cultures (Table 5) .

MHV-2 induces non-fusion-type plaques consisting of round dead cells in SRCDFl DBT cells (Hirano et al., 1974) . The presence of 0.05% trypsin in the medium induced the rapid appearance of fusion-type plaques at 37". Purified virions were treated with trypsin for 5 min at 3'7', mixed with an equal volume of 1% SDS, 1% 2-mercaptoethanol, 50 mMTris, pH 6.8, and 20% glycerol, boiled immediately, and electrophoresed on 12% polyacrylamide gels. The gels were stained with Coomassie brilliant blue R. (A) Virions were purified from culture fluids harvested at 30-min intervals with the autoharvester from 6 to 16 hr after infection (at 32"). Slot A: Sp-11 treated with 0.01% trypsin (3x Crystallized, Sigma). Slot B Untreated Sp-11. Slot C: Untreated 6v-1. Slot D: 6v-1 treated with 0.01% trypsin. T, trypsin, Molecular weight markers are indicated at the right (kilodaltons). (B) Virions were purified from culture fluids contacted with the infected cells for 16 hr after infection. The purified virions of the wild type harvested at 37" were kindly supplied to us by Dr. F. Taguchi. Sp-11 virions were harvested at 32". Slot A: Untreated Sp-11. Slot B: Sp-11 treated with 0.065% trypsin. Slot C: Sp-11 treated with 0.05% trypsin. Slot D: Untreated wild type. Slot E: Wild type treated with 0.005% trypsin. Slot F: Wild type treated with 0.05% trypsin. Slot G: Trypsin. Slot G: Molecular weight markers. Bovine serum albumin (68K), chicken serum albumin (45K), and soybean trypsin inhibitor (21.5K). non-fusion-type plaques of MHV-2 to fusion type.

The role of protease in MHV plaque formation appears to be different from the similar role played by trypsin in Sendai virus infection (Homma and Ohuchi, 1973; Scheid and Choppin, 1974; Ishida and Homma, 1978) . In the later 'case, proteolytic cleavage of viral structural F glycoprotein caused virions infectious and capable of inducing cell fusion. In MHV case, the virions harvested from fused cell cultures and those harvested from unfused cultures, i.e., wild-type 6v-1 and mutant Sp-11 both harvested at 32", were almost indistinguishable in SDS-polyacrylamide gels, and protease treatment of the virions failed to recover the cell fusion activity. Another important difference was that fusion induction was independent of virus growth (Fig. 1) and protease did not affect MHV infectivity (Table 3) or replication (Table 5) .

Biological data indicated that proteolysis of the surface components of Sp-llinfected cells was required for cell fusion. Though the 68K protein which was present in the wild type-infected cells was absent in the mutant-infected cells, incubation at 39 or at 32" with trypsin failed to induce the 68K protein in the Sp-11-infected cells.

In addition, 68K protein was absent in the cells infected with the phenotypically reverted Lp2c. Thus, the involvement of intracellular 68K protein in the cell fusion was excluded. Tryptic digestion of 6v-1 and Sp-11 virions resulted in cleavage of bands 1,2, and 6 to generate bands 3 and D. But the polypeptide composition as well as its tryptic digestion pattern were quite similar in both wild type and mutant. This appears to exclude the possibility that proteolytic cleavage of viral polypeptides was responsible for cell fusion induction by the wild type, though it is still possible that for cell fusion to occur, only a few proteolysed MHV are enough, and in the wild type-infected cells such proteolysis could actually be brought about by a low level of cellular protease in an undetectably minor fraction of the virions on the cell surface. Another possibility is that cellular protein(s) may be interfering with MHV-mediated cell fusion, and proteolysis of such cellular materials may result in the fusion.

The effect of protease on fusion induction was not restricted to Sp-11. MHV-S was induced to form fusion-type plaques by trypsin in otherwise resistant cells, and MHV-2, a producer of non-fusion-type plaques, was induced to form fusion-type plaques by trypsin. The fact that MHV usually replicates and produces plaques better in transformed cells (Chany and Robb-Maridor, 1968; Sturman and Takemoto, 1972 ) may be related to this phenomenon, as transformed cells show an increased level of protease (Goldberg, 1974) . The observation not only has practical value, as fusion-type plaques are more easily scored than non-fusion-type plaques, and the development of plaques was more rapid, but also is important when we consider MHV pathogenesis, i.e., local proteolytic activity may affect the pathology of the organs affected.

One of the possible mechanisms of cold sensitivity of Sp-11 may be that the protease level of SRCDFi DBT cells at 32" was much lower than at 39", and Sp-11 was somehow more strongly dependent upon protease than the wild type. MHV-2 or the

Charenvelope: Tryptic peptide analysis. Virology 77, acterization of murine sarcoma virus (Kirsten) 650-660. transformation of mouse and human cells

Enhancsarcoma virus-transformed nonproducer mouse ing effect of the sarcoma virus in the replication cells persistently infected with mouse hepatitis of mouse hepatitis virus in vitro

Effect of protease inhibitors on focus Carrier state of antibody and viruses in a mouse formation by murine sarcoma virus

Sp-11 variants T4a and T4c could be, in and mouse hepatitis viruses. Lab. Anim. Sci. 26,  this sense, still more dependent upon pro- 153-159. tease, as it failed to induce cell fusion at GOLDBERG, A. R. (1974) . Increased protease levels in 37-39" without the help of trypsin. these possibilities will be tested by specific Mouse hepatitis virus infection as a highly conantisera which are now being prepared. tagious, prevalent, enteric infection of mice. Proc. Sot. Exp. Biol. Med. 112.161-165. SCHEID, A., and CHOPPIN, W. M. (1974) . Identification ACKNOWLEDGMENTS of biological activities of paramyxovirus glycoproteins. Activation of cell fusion, hemolysis, and in-Our thanks are due to Dr. F. Taguchi for the antifectivity by proteolytic cleavage of an inactive pre-MHV-S rabbit serum, preparation of purified wildcursor protein of Sendai virus. Virology 57, 475type virions, and advice. This work was supported in 490. part by a grant from the Japanese Ministry of Ed- STURMAN, L. (1977) . Characterization of a coronaviucation.rus. I. Structural proteins: Effects of preparative conditions on the migration of protein in polyacrylamide gels. Virob 77.637-649.