Pathway of rubella virus infectious entry into Vero cells

Transcription

1 Journal of' General Virology (1996), 77, Printed in Great Britain 303 Pathway of rubella virus infectious entry into Vero cells Rosanna Petruzziello, 1 Nicola Orsi, 1 Stefania Macchia, 1 Sabrina Rieti, 1 Teryl K. Frey 2 and Paola Mastromarino 1. 1 Istituto Pasteur-Fondazione Cenci Bolognetti, Institute of Microbiology, School of Medicine, University 'La Sapienza ', Piazzale Aldo Moro 5, Rome, Italy and 2 Department of Biology, Georgia State University, Atlanta, Georgia , USA The mechanism and the kinetics of rubella virus (RV) penetration into Vero cells were studied. By using pronase or acid treatment to inactivate virus which had adsorbed to cell membrane but had not been internalized, it was found that a period of 7 h was required in order for all of the adsorbed virus to enter the host cells. Lysosomotropic agents (monensin, methylamine, ammonium chloride and chloroquine) were used to study the mechanism by which RV penetrates host cells. Virus replication was inhibited if treatment of cells with these compounds was performed for at least 9 h after infection. However, if extracellular adsorbed virions were eliminated by acid treatment following removal of the lysosomotropic compounds, RV replication was completely inhibited by treatment with these drugs for any time period after adsorption. This indicated that the prolonged period of treatment with these compounds necessary to inhibit virus replication is due to the slow rate of RV internalization. None of the compounds had any effect on infection initiated by transfection of RV RNA, confirming that these drugs were exerting their inhibitory activity at penetration. The inhibition of RV replication by lysosomotropic compounds indicates that RV penetrates host cells by the endosomal pathway. Introduction After adsorption to the cell surface, which is a specific receptor-mediated event, virions penetrate host cells by two major pathways (Marsh & Helenius, 1989). Following the first pathway, viruses enter cells by a mechanism closely related to endocytosis. The recepto~ virus complex is internalized in clathriu-coated pits and delivered to endosomes. The acidic interior of the endosome triggers a conformational change in the viral fusion protein. Exposed hydrophobic sequences induce fusion of the viral surface with the endosomal membrane releasing the genome into the cytoplasm (Helenius et al., 1980; Marsh & Helenius, 1980). Following the second pathway, viruses interact with the cell surface and fusion between the virus envelope and the plasma membrane leads to the penetration of the genome directly into the cell cytoplasm (Marsh, 1984). Little is known about the mechanism by which rubella virus (RV) enters host cells. Members of the other togavirus genus, the alphaviruses, enter by the endocytotic pathway. Several lines of evidence suggest that * Author for correspondence. Fax RV fuses with cell membranes in an acidic environment suggesting that RV also enters cells by this pathway. RV produces haemolysis and fusion of erythrocytes in a mildly acidic ph (Kobayashi, 1978; Vfifin~inen & Kfifiri/iinen, 1980); a brief treatment at a ph below 6.0 causes a conformational change in the viral envelope glycoproteins E1 and E2, resulting in the acquisition of liposome-binding activity of virions and inducing fusion of RV-infected cells (Katow & Sugiura, 1988). RV capsids also undergo a solubility change below ph 5.5, becoming hydrophobic in nature (Mauracher et al., 1991). However the entry pathway of RV has not yet been identified. In this report we have analysed the kinetics of internalization of RV into host cells and the mechanism by which RV enters host cells. For this latter purpose we studied the effect on virus replication of lysosomotropic compounds (such as monensin, methylamine, ammonium chloride and chloroquine) which are able to increase the ph of intracellular acidic compartments. Methods Cells and virus. Vero cells were grown at 37 C in a 5% CO 2 atmosphere in Eagle's MEM supplemented with 6 % fetal bovine serum SGM

2 304 R. Petruzziello and others (FBS), 2 mm-glutamine, 100 IU/ml penicillin G and 100 pg/ml streptomycin. Rubella virus, Therien strain, was seeded in MEM containing 2 % FBS (maintenance medium) and titrated by plaque assay in Vero cells as previously described (Mastromarino et al, 1990). Indirect immunofluorescence assay. Vero cells were grown in microtissue culture chamber slides (LabTek Products) at a concentration of 5.2 x 104 cells/well. After 24 h, cells were mock-infected or infected with 1 : 3.3 dilutions of RV (1 h, 4 C). After 48 h incubation at 37 C the cells were washed three times with Dulbecco's PBS, fixed in acetone and incubated with a 1 : 10 dilution of human high-titre antirubella virus serum (Institut Pasteur). After incubation for 1 h at 37 C, the slides were washed with PBS and incubated with a 1 : 5 dilution of FITC-conjugated goat anti-human antibody (Behring). The slides were then washed three times with PBS ph 7-4, mounted in 90% (v/v) glycerol-carbonate buffer ph 9.4 and examined by fluorescence microscopy. Internalization assay. RV entry into Vero cells grown in 24-well tissue culture plates was assayed either by protease sensitivity of membranebound virions or by the protection of internalized virus from acid inactivation. The time course of RV penetration after pronase treatment was measured by an infectious centre assay. Vero cell monolayers were infected at a m.o.i, of 0.1 p.f.u/cell for 1 h at 4 C to allow attachment but not internalization of virions. Immediately after adsorption (time 0) or at different times after the temperature shift to 37 C, infected cell monolayers, washed five times in ice-cold PBS, were treated for 90 min at 4 C on ice with PBS or with 1 mg/ml of pronase in PBS to remove externally bound virions. The cells were scraped off, washed three times in maintenance MEM by low-speed centrifugation and resuspended in 1 ml of medium. Serial 1:3-3 dilutions of cells in MEM were added to 35 mm dishes of confluent Vero cells. After incubation for 1 h at 37 C, the cultures were carefully overlaid with plaque assay agar (0.4% Oxoid no. 1 in MEM containing 1% FBS and 0.1% DEAE-dextran). Plaques were stained with 0.1% crystal violet in 10% formalin after incubation at 37 C for 5 days. Protection of internalized virus from acid inactivation was assayed using a modified protocol described by Rosenthal et al. (1984). Vero cells, kept at 4 C on crushed ice in maintenance MEM for 20 min, were infected with approximately 103 p.f.u./well of RV at 4 C for 1 h to allow attachment but not internalization of virions. After the 1 h adsorption period, the cells (washed three times with ice-cold PBS) were added to pre-warmed (37 C) maintenance medium and incubated at 37 C. Immediately prior to and at various times after the temperature shift to 37 C, 1 ml of PBS or PBS ph 3.0 was added to wells for 3 min. Following several washes in medium to neutralize the acid, a plaque assay agar was added and the plates were incubated until plaques formed. Extraction of viral RNA. The RNA for transfection experiments was extracted from Vero cells 4 days post-infection (p.i.). After solubilization with proteinase K (Boehringer Mannheim) total RNA was extracted with phenol and chloroform using the methods of Sawicki et al. (1981). The extracted RNA was precipitated from the aqueous phase by the addition of sodium acetate to 100 mm and ethanol to 70 %. RNA extracted from 107 cells was dissolved in 250 lal of TE buffer (10 mm-tris HC1 ph 8.0, 1 mm-edta) and stored in small aliquots at 70 C. Transfection of Vero cells. Confluent monolayers in 35 mm tissue culture dishes were transfected with an amount of RV-infected cell RNA that had been determined to produce approximately 50 plaques per well. The RNA (1.5-3 lal) was added to 120 gl of ice-cold OPTI- MEM-I reduced serum medium (Gibco) containing 4.7 pg of Lipofectin reagent (Gibco BRL) gently mixed and maintained on ice for 10 min. The mixture was then added to Vero cells which had been washed twice with ice-cold OPTI-MEM and the cells incubated at room temperature on a rocking platform. After 1 h, the cells were washed twice with OPTI-MEM, overlaid with MEM containing 0.4 % agar (Oxoid) and incubated for 4 days at 37 C. Plaques were stained with 0.1% crystal violet solution. Lysosomotropic agents. Stock solutions of 0'5 M-methylamine, 0.01 M-chloroquine and 1 M-ammonium chloride were made in PBS. Monensin was dissolved in ethanol at a concentration of 0-01 M. The stock solutions were stored at -20 C. All the chemicals were obtained from Sigma. Results Inhibition of rubella virus infection by lysosomotropic agents The entry pathway of RV into Vero cells has been studied using acidotropic weak bases and carboxylic ionophores which raise the ph of acidic organelles and disrupt endosomal functions (Maxfield, 1982). By preventing the low ph-induced fusion of viral and cell membranes which occurs in the endosome, these compounds are able to block the entry of certain viruses into the cytosol (Gollins & Porterfield, 1986; Helenius et al., 1982; Matlin et al., 1981, 1982). In a first series of experiments, medium containing different concentrations of four lysosomotropic drugs (monensin, methylamine, ammonium chloride and chloroquine) were added to Vero cells after the viral binding step (1 h at 4 C) and maintained for the complete virus replication cycle (48 h). The effect of the drugs on virus antigen synthesis, assessed by indirect immunofluorescence assay, is shown in Table 1. With all of the compounds, a dose-dependent inhibition of RV multiplication was observed. In each case, the maximum concentration used, which was the maximum non-cytotoxic concentration, inhibited virus multiplication completely. The presence of the drugs at the concentrations used did not affect the ph of the culture medium. Fig. 1 shows the effect of the drugs on the early (0-24 h) or late (24~48 h) phases of the RV multiplication cycle using a concentration of each drug commonly employed to raise the ph of intracellular compartments. All the compounds showed an inhibitory effect on RV antigen synthesis when present during the first part of the virus replication cycle, whereas all were ineffective during the later phases of virus multiplication. The time course of the inhibitory activity of the lysosomotropic compounds on the early steps of viral infection was evaluated by experiments in which the compounds were added to the cells immediately after virus adsorption (1 h at 4 C) and maintained for different times (Fig. 2). The presence of the compounds during the first 3 h p.i. produced an approximately 50 %

3 Entry of rubella virus into Vero cells 305 Table 1. Effect of lysosomotropic compounds on R V replication* Compound Concentration (gm) RV antigen-positive cells (%) None 50 Monensin Methylamine Ammonium chloride Chloroquine O * Each compound at the indicated concentration was added to RV-infected Vero cells immediately after adsorption (1 h at 4 C). The cells were then incubated at 37 C for 48 h in the continuous presence of the compound and the percentage of infected cells was determined by indirect immunofluorescence. 5O 40 m t "~ 40 I I I I :~ 30 '~ 30 e~ 2O 20 ~> 10 >~ 10 Monensin Methylamine Ammonium (3"3 pm) (5 mm) chloride (20 m~) Chloroquine (50 gm) Fig. 1. Effect of time addition/removal of lysosomotropic compounds on RV multiplication. The compounds were present in the infected cell medium for different periods of time after adsorption (1 h at 4 C). Bar shading: (.) 0-24h; ([]) 24-48h; (D) 0-48h. Infection was performed with a viral input giving 50% RV antigen-positive cells in the absence of lysosomotropic compounds. The percentage of infected cells was determined by indirect immunofluorescence. decrease in the number of RV antigen-positive cells. An almost complete inhibition of RV replication was achieved only when the drugs were present in the culture medium during the first 9 h p.i. These results are in contrast with those obtained with other togaviruses. The penetration of the alphaviruses Semliki Forest virus (SFV) and Sindbis virus (Helenius et al., 1982; Talbot & m Time of removal (h) Fig. 2. Influence of time of removal of compounds on RV replication. The compunds, 3.3 pm-monensin (A), 5 mm-methylamine (O), 20 mmammonium chloride (11) or 50 gm-chloroquine (A) were added to the medium immediately after adsorption (1 h at 4 C) and removed 0, 3, 6 or 9 h later. The percentage of RV antigen-positive cells was determined at 48 h p.i. by indirect immunofluorescence. Vance, 1982) were completely inhibited by lysosomotropic compounds 1 h or less after viral adsorption. Therefore, either the drugs used have a different mechanism of inhibition of RV replication, or the kinetics of RV penetration are slower than those of other togaviruses. Kinetics of RV entry into Vero cells Since the kinetics of RV entry are as yet unknown, it was necessary to determine the rate at which RV penetrates cells. Pronase sensitivity of membrane-bound rubella

4 306 R. Petruzziello and others loo I 1 I I [] 100 I I I I O g 80 g g 60 ~." 40 e~ > =~ 20 I I I I Time after adsorption (h) Fig. 3. Kinetics of RV entry into Vero cells measured as the rate of loss of pronase (I--1) or acid (A) sensitivity (see Methods for details). The results are plotted as the percentage of control p.f.u, under the same experimental conditions without pronase or acid treatment (100%). virions was used to locate virus particles at different times after temperature shift to 37 C during RV penetration into host cells. The adsorption step was performed for 1 h at 4 C on ice to prevent penetration of the virus into the cells and to allow synchronization of subsequent steps in penetration. After removal of unbound virus, pronase treatment was performed prior to or at various times after incubation at 37 C. Internalized virus (resistant to protease) was quantified by an infectious centre assay which allows the detection of productive infection. The results shown in Fig. 3, indicate that 1 h after the temperature shift to 37 C, 50% of the virus bound to the cell surface was internalized, while 3 h after the temperature shift 65 % of the virions were insensitive to pronase and a period of 7 h was required in order for all the extracellular virus to be internalized. Thus, the kinetics of entry appeared to be biphasic with the most rapid rate occurring during the initial 1 h, followed by a much slower rate until internalization was complete. RV internalization was further studied by protection of internalized virus from acid inactivation following the method of Rosenthal et al. (1984). The assay used was based on the observation that RV infectivity is highly sensitive to low ph. We demonstrated that the exposure of the virus to ph 6.0 or 5.0 for 1 h, followed by neutralization to ph 7.4 before the infection of cell monolayers, caused a 300-fold and a 1000-fold drop in infectivity respectively, whereas treatment at ph 3"0 produced a complete inactivation of virus infectivity. Preliminary experiments were performed to verify the effect of ph 3.0 on cell morphology and viability. Treatment for 3 min did not produce any cytotoxic effect whereas the incubation of RV in the same experimental conditions caused a reduction of virus infectivity by more than 99 %. ~ o ~ o /. '~' _t I ~ I ~ W i ~I Time after adsorption (h) Fig. 4. Inhibition of RV penetration into Vero cells by lysosomotropic compounds. Compounds were added to Vero cells following adsorption with RV (1 h at 4 C) and then the cultures were shifted to 37 C. At various times, medium containing the compounds was removed, extracellular virus was inactivated with ph 3.0 buffer, and the culture was overlaid with plaque assay agar. The numbers of plaques per well are expressed as a percentage of the plaques per well obtained in the absence of the compounds (100%). Symbols: A, control; A, 3.3 gmmonensin; 0, 5 mm-methylamine; m, 20 mm-nh4c1; V, 50[/Mchloroquine. RV entry was studied by using acid ph treatment to inactivate extracellular membrane-bound virions (internalized virus is resistant to inactivation by acid) at different times following viral attachment to cell membrane. The amount of internalized virus was quantified by plaque assay. After adsorption (1 h at 4 C) extracellular virions were inactivated by treatment with PBS ph 3-0 immediately before (time 0) or at different times after the temperature shift to 37 C. Acid treatment of a cell monolayer before virus adsorption was performed to ensure that the susceptibility of cells to infection by RV was not altered. As shown in Fig. 3, the kinetics of RV penetration showed a pattern similar to the one obtained with pronase treatment. In all the subsequent experiments acid inactivation of the virus remaining on the cell surface was preferred to pronase treatment because it requires a minimum amount of time for handling and allows a good evaluation of the kinetics of virus uptake. Effect of lysosomotropic compounds on R V penetration Given that the kinetics of RV internalization are prolonged, the effect of lysosomotropic compounds on virus penetration was re-evaluated using acid inactivation to remove extracellular virions following treatment with the compounds. During incubation of infected cells with lysosomotropic agents virions internalized in the endosome fail to release the genome into the cell cytoplasm and are lost, whereas the acid-sensitive fraction is still on the cell surface and is not affected by the compounds. As shown in Fig. 4, the presence of the

5 Entry of rubella virus into Vero cells 307 Table 2. Effect of lysosomotropic compounds on R V-RNA transfection Virus recovered* (p.f.u./well) and treatment timing Treatment PreincubatioM Transfection:~ Transfection and virus replication None Monensin (3-3 gm) Methylamine (5000 gm) Ammonium chloride (20000 gm) Chloroquine (50 lam) * Data represent the mean of three independent experiments; each experiment was conducted in duplicate. The SD were < 5 % for all values. I" Vero cell monolayers were incubated with the compounds in OPTI-MEM for 1 h at 37 C. After two washes in OPTI-MEM, cells were transfected with RV RNA by lipofectin-mediated transfection in the absence of the compounds and then overlaid with MEM containing 0.4% agar. :~ Lysosomotropic compounds were added during the 1 h transfection period. Cell monolayers were transfected with RV RNA in the presence of the compounds, twice washed with OPTI- MEM and then incubated at 37 C for 7 h with the compounds in OPTI-MEM. At the end of this period, cells were washed twice with OPTI-MEM and then overlaid with MEM containing agar. drugs through 1, 3, 7 and 9 h p.i. totally inhibited the production of viral plaques. Transfection of R V RNA into Vero cells in the presence of lysosomotropic compounds To further study the effect of lysosomotropic agents on RV replication a different experimental approach was followed. Since the penetration step is considered to be the target of these drugs, this step was circumvented by transfecting cells with viral RNA in the presence or absence of lysosomotropic compounds. RV RNA extracted from infected Vero cells was introduced into cells by Lipofectin reagent-mediated transfection. Experiments were performed treating cells with the compounds for 1 h before or during transfection and continuing through 7 h p.i. All these different conditions were performed as no data are available on the kinetics of RV RNA transfection. Since at least 7 h of treatment of RVinfected cells with these compounds is required to inhibit RV multiplication (Fig. 2), cells transfected with RV RNA were treated with these compounds for 7 h to evaluate the effect on RV multiplication following transfection. Results obtained, reported in Table 2, demonstrated that in all the different experimental conditions the infectivity of viral RNA was not affected by the lysosomotropic compounds demonstrating that the action of the compounds is exerted at a stage of viral replication before viral RNA translation. Discussion The pathway of entry and penetration of toga(alpha)- viruses has been studied in considerable detail. After endocytosis, internalized virions are rapidly delivered to early endosomes where the mildly acidic conditions trigger an irreversible conformational change in the viral fusion proteins, thus inducing membrane fusion and penetration. This process occurs within the first hour of virus-cell contact (Marsh & Helenius, 1989). In this study we analysed the kinetics of the entry of RV, a toga(rubi)virus, into host cells. Previous studies have shown that binding of RV to cells occurs rapidly (Vaheri et al., 1967; Bardeletti et al., 1972). Based on the finding in this study that, following adsorption, a fraction of adsorbed virus remains sensitive to pronase treatment or to inactivation by ph 3'0 treatment for 7 h, it appears that RV internalization into cells is only complete 7 h after virus binding to the cell surface. The reason for the slow rate of entry is unknown but could involve slow migration of the virus receptor complex to a clathrincoated pit. RV infection is characterized by a long latent period (at least 12 h) and by the inability to achieve infection of all the cells at the time of initial virus adsorption (Hemphill et al., 1988). The slow kinetics of RV entry could be a significant factor in determining these characteristics. The infection of Vero cells by RV was inhibited by four different lysosomotropic compounds. Although we do not know the physical states of the virus in different phases of entry, the inhibitory activity of these compounds suggests that endosomes serve as the site of RV penetration and that low endosomal ph is critical in the penetration reaction. It has been reported (Maassab & Cochran, 1964) that the lysosomotropic agent amantadine inhibits an early phase of RV replication. In our experiments a complete inhibition of RV antigen synthesis was achieved only when the drugs were present in the culture medium during the first 9 h of infection. The time course of the sensitivity of RV multiplication to

6 308 R. Petruzziello and others lysosomotropic agents correlated with that observed for RV entry, indicating that the step of the virus replication cycle affected by these drugs was penetration. This was further confirmed by the finding that when extracellular virions bound to the cell surface were eliminated by ph 3.0 treatment following drug treatment, a complete inhibition of viral replication was observed for any time period after adsorption. On the basis of our results and the known effects of lysosomotropic agents we conclude that the drugs affect the uncoating step of RV by increasing the endosomal ph. Additional support for uncoating as the major site of action of the agents was provided by results obtained using transfection with RV RNA. The presence of the compounds before, during or after transfection did not affect RV replication, indicating that the effect of these drugs was at a step prior to translation of the genome. In conclusion, our results indicate that the delivery of rubella virus genome into the cytoplasm is mediated by a low ph-induced membrane fusion event between the viral and endosomal membranes. We wish to thank Dr L. Marr, Department of Biology, Georgia State University of Atlanta (GA) USA, for helpful technical assistance in the transfection experiments. This research was supported by grants from MURST and CNR. References BARDELETTI, G., HENRY, M., SOHIER, R. & GAUTHERON, D. C. (1972). Primary effects of the rubella virus on the metabolism of BHK-21 cells grown in suspension cultures. Archiv fiir die gesamte Virus-,/orschung 39, GOLLINS, S.W. & PORTeRFIELD, J.S. (1986). The uncoating and infectivity of the flavivirus West Nile on interaction with cells: effects of ph and ammonium chloride. Journal of General Virology 67, HELENIUS, A., KARTENBECK, J., SIMON& K. & FRIES, E. (1980). On the entry of Semliki Forest virus into BHK-21 cells. Journal of Cell Biology 84, HELENIUS, A., MARSH, M. & WHITE, J. (1982). Inhibition of Semliki Forest virus penetration by lysosomotropic weak bases. Journal of General Virology 58, HEMPHILL, M. L., FRONG, R.Y., ABERNATHY, E. S. & FREY, T.K. (1988). Time course of virus-specific macromolecular synthesis during rubella virus infection in Vero cells. Virology 162, KATOW, S. & SUGIURA, A. (1988). Low ph-induced conformational change of rubella virus envelope proteins. Journal of General Virology 69, KOBAYASHI, N. (1978). Hemolytic activity of rubella virus. Virology 89, MARSH, M. (1984). The entry of enveloped viruses into cells by endocytosis. Biochemical Journal 218, MARSH, M. & HELENIUS, A. (1980). Adsorptive endocytosis of Semliki Forest virus. Journal of Molecular Biology 142, MARSH, M. & HELENIUS, A. (1989). Virus entry into animal cells. Advances in Virus Research 36, 10~151. MAASSAB, H. F. & COCHRAN, K. V. (1964). Rubella virus: inhibition in vitro by amantadine hydrochloride. Science 145, MASTROMARINO, P., CIOE', L., RIETI, S. & ORSI, N. (1990). Role of membrane phospholipids and glycolipids in the Vero cell surface receptor for rubella virus. Medical Microbiology and Immunology 179, MATLIN, K.S., REc,~;go, H, HELENIUS, A. & SIMONS, K. (1981). Infectious entry pathway of influenza virus in a canine kidney cell line. Journal of Cell Biology 91, MATLIN, K.S., REGGIO, H., HELENIUS, A. & SIMONS, K. (1982). Pathway of vesicular stomatitis virus entry leading to infection. Journal of Molecular Biology 156, MAURACHER, C. A., GILLAM, S., SHUKIN, R. & TINGLE, A. J. (1991). ph-dependent solubility shift of rubella virus capsid protein. Virology 181, MAXF~ELD, F.R. (1982). Weak bases and ionophores rapidly and reversibly raise the ph in endocytic vesicles in cultured mouse fibroblasts. Journal of Cell Biology 95, ROSENTHAL, K.S., LEUTHER, M.B. & BARISAS, B.G. (1984). HSV binding and entry modulates cell surface protein mobility. Journal of Virology 49, SAWICKI, S. G., SAWICKI, D. L., K.~.~RIAINEN, L. & KER)~NEN, S. (1981). A Sindbis virus mutant temperature sensitive in the regulation of minus strand RNA synthesis. Virology 115, TALBOT, P. J. VANCE, D. E. (1982). Biochemical studies on the entry of Sindbis virus into BHK-21 cells and the effect of NH4CI. Virology 118, V.~,N,~NEN, P. ~ K,g,,~RI,~INEN, L. (1980). Fusion and hemolysis of erythrocytes caused by three togaviruses: Semliki Forest, Sindbis and rubella. Journal of General Virology 46, VAHERI, A., SEWICK, W. D. & PLOTKIN, S. A. (1967). Growth of rubella virus in BHK-21 cells. I. Production, assay, and adaptation of virus. Proceedings of the Society.['or Experimental Biology and Medicine 125, (Received 19 June 1995; Accepted 6 October 1995)