Microbial Pathogenesis 1990 ; 9: 375-386
Host cell-dependent lateral mobility of viral glycoproteins Shari L . Lydy, Sukla Basak and Richard W . Compans* Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294, U.S.A . (Received 24 April 1990 ; accepted in revised form 24 August, 1990)
Lydy, S . L . (Dept of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294, U .S .A .), S . Basak and R . W . Compans . Host cell-dependent lateral mobility of viral glycoproteins . Microbial Pathogenesis 1990 ; 9 : 375-386 . The lateral mobility of viral envelope proteins on the plasma membranes of infected cells is an important factor in both virus assembly and pathogenesis. The envelope glycoproteins of measles and human parainfluenza virus are mobile on the surfaces of infected HeLa cells and undergo lateral redistribution in the presence of specific antibody, forming unipolar caps . In contrast, no such redistribution was observed with influenza virus hemagglutinin (HA) or vesicular stomatitis virus (VSV) G glycoproteins on infected HeLa cell surfaces . However, the HA and G glycoproteins were both found to be mobile in the plasma membrane of CV-1 cells, or human or murine peritoneal macrophages . These results indicate that host cell-dependent as well as virus-specific factors are involved in determining viral glycoprotein mobility . No significant differences in the patterns of synthesis of influenza or VSV viral proteins were found in the various cell types examined . The HA and G proteins, when expressed from vaccinia virus recombinants, were each found to be immobile in HeLa cells and mobile in CV-1 cells, thus indicating that the host cell-dependent differences in mobility are an intrinsic property of each viral glycoprotein molecule and not the result of interaction with other viral components . It is suggested that the association of viral glycoproteins with either the cytoskeleton or membraneassociated cellular proteins may be related to the observed differences in lateral mobility . Key words : viral glycoproteins ; lateral mobility ; plasma membrane .
Introduction Many viral antigens are mobile on the plasma membrane, and this mobility may play an important role during viral infection . During budding of enveloped viruses from the plasma membrane, viral antigens are assembled into specific domains on the cell surface from which host proteins are excluded, and lateral mobility of glycoproteins may be involved in forming such domains . In addition, viral antigens incorporated into cell membranes prior to virus budding may react with virus-specific divalent antibody, and undergo lateral redistribution to form a discrete cap on the cell surface ."' Such local concentration of viral antigen expression on infected cell surfaces may facilitate recognition by the host immune system, leading to immune lysis of infected cells in the presence of complement or immune lymphocytes, or to virally induced autoimmune disease ." It has been suggested that viral antigen-antibody complexes shed from infected cell surfaces after capping may cause immune complex disease and may ultimately be a survival mechanism for many viruses since infected cells escape surveillance from host defenses, leading to the establishment of persistent viral infection . 3' 5 Finally, many viral glycoproteins exhibit the capacity to induce membrane * Author to whom correspondence should be addressed . 0882-4010/90/120375+12 803 .00/0
© 1990 Academic Press Limited
37 6
S . L . Lydy et al .
fusion leading to the formation of syncytia, and it is likely that high local concentrations of glycoproteins are needed for this biological activity . It was previously reported that the glycoproteins of measles virus are mobile on the surfaces of infected HeLa cells` whereas, under similar experimental conditions, the mobility of influenza HA was restricted on HeLa cell plasma membranes .' In HeLa cells doubly infected with measles and influenza viruses, measles virus HA was found to be mobile whereas no capping of influenza HA was observed .' It has been proposed that cytoskeletal components of virus-infected cells may control the movement of viral structures at the plasma membrane 8.9 and be involved in the viral budding process .", " In addition, many viruses cause changes in the organization of host cell microfilament and microtubular networks during infection .' 1,13 Based on these observations, it is possible that viral glycoproteins differ in their association with cytoskeletal components in different cells during the infection process, and that this interaction affects the lateral mobility of the viral glycoproteins . The present study was undertaken to investigate further the basis for the observed differences in lateral mobility of viral glycoproteins in various cell lines . We have studied the antibody-induced redistribution of influenza HA, the G glycoprotein of vesicular stomatitis virus (VSV), hemagglutinin-neuraminidase (HN) and fusion (F) glycoproteins of human parainfluenza virus type 3 (P13), and F protein of respiratory syncytial virus (RSV) on HeLa and CV-1 cell plasma membranes . Vaccinia virus recombinants were used to express several of the viral glycoproteins in the absence of other viral components, to determine whether the restricted mobility of certain glycoproteins is an intrinsic property of these glycoprotein molecules .
Results Mobility of viral glycoproteins on HeLa plasma membranes We analysed the lateral mobility of viral glycoproteins by determining their ability to undergo lateral redistribution in response to specific antibody . In control experiments, cells immunolabeled with antibody on ice showed uniform surface distribution of viral antigens . Antibody-induced redistribution of the viral glycoprotein was shown by formation of irregular patches of viral antigen-antibody complexes or a discrete unipolar cap on cells immunolabeled at 37°C, whereas uniform surface fluorescence on cells labeled under such conditions indicated restricted mobility of the surface antigen . Antibody-induced capping of virus antigen was previously reported to occur optimally at 37°C . 1 We incubated virus-infected cells with antibody conjugates at 37°C for intervals from 30 to 90 min to determine the optimal incubation time for maximal cell capping, and found that capping of viral glycoproteins occurred on cell surfaces within 30 min . In addition, the time course of expression of influenza and VSV viral polypeptides in HeLa and CV-1 cells was determined to ensure that capping experiments were done at times when the infected cells were expressing large amounts of viral glycoproteins (Fig . 1) . At 8 h post-infection, a diffuse and random pattern of surface fluorescence on influenza virus-infected HeLa cells was observed after immunolabeling at 37°C with specific antibody to the HA glycoprotein [Fig . 2(A)], as previously reported .' Under similar experimental conditions, the VSV G glycoprotein also was found to be uniformly distributed on HeLa cell surfaces [Fig . 2(B)] . In both cases the patterns of fluorescence closely resembled those seen on cells immunolabeled on ice, indicating the lack of antigen redistribution . In contrast, lateral redistribution was observed when measles virus-infected [Fig . 2(C)] or P13-infected HeLa cells [Fig . 2(D)] were immunolabeled with antibodies specific for the HA or HN protein, respectively . The fusion (F) glycoprotein of P13 virus also exhibited antibody-induced
Host cell-dependent lateral mobility of viral glycoproteins
U
2
4
6 8
10
U 2 4 6
377
8 10
2
4 6
8
10
2 4 6 8
10
P,- 3 HA G NP-
N, NS -
M NS, -
HeLa
CV_ I
HeLa
CV_ I
Fig . 1 . Time course of VSV and influenza protein synthesis in HeLa and CV-1 cells . At 2-10 h postinfection, the infected HeLa and CV-1 cells were radiolabeled with 35S-methionine and the radioactive VSV (A) and influenza (B) polypeptides analysed on 10% and 15% reducing polyacrylamide gels, respectively . U designates uninfected cells .
redistribution on the surfaces of infected HeLa cells (data not shown) . Joseph and Oldstone' previously reported the mobility of measles HA on the majority of infected HeLa cells, and the present results indicate that the HN and F glycoproteins of P13 virus are similar in this regard . Thus, glycoproteins of various enveloped viruses show significant differences in their lateral mobility on the surfaces of infected HeLa cells . Host cell-dependent differences in the lateral mobility of viral glycoproteins The finding of significant differences in the mobility of various viral glycoproteins on surfaces of infected HeLa cells led us to investigate viral glycoprotein mobility further in other cell types . CV-1 cells were infected with either influenza virus or VSV, and the pattern of surface fluorescence of the HA and G proteins was examined after immunolabeling with specific antibody at 37°C . Significant patching or capping of the influenza HA [Fig . 3(A)] and VSV G [Fig . 3(B)] glycoproteins occurred on 90% and 62% of the infected CV-1 cell surfaces, respectively, whereas a uniform distribution of antigens was observed in cells immunolabeled on ice (not shown) . We also examined the pattern of surface immunofluorescence on virus-infected human and mouse peritoneal macrophages . In control cells immunolabeled on ice, the viral antigen remained uniformly distributed over the macrophage surfaces [Fig . 4(A)] . In contrast, nearly all of the influenza virus-infected murine macrophages [Fig . 3(B)] and human macrophages (not shown) were found to show a pattern of influenza HA in unipolar caps after immunolabeling at 37°C . Under similar experimental conditions, the G protein of VSV was also found to redistribute in both human and murine peritoneal macrophages (not shown) . These results indicate that certain viral glycoproteins
378
S . L . Lydy et at.
Fig . 2 . Distribution of viral antigens on the surfaces of infected HeLa cells after immunolabeling at 37°C . Cells were infected and immunolabeled as described in Materials and methods . (A) Influenza HA (8 h postinfection) and (B) VSV-G (6 h post-infection) showing diffuse fluorescence . (C) Measles HA (24 h postinfection) and (D) P13-HN (24 h post-infection) showing redistribution into caps in many cells .
exhibit a high degree of lateral mobility in some cell types, and restricted mobility in another cell type . Both the influenza HA and VSV G glycoproteins are immobile on surfaces of infected HeLa cells, but exhibit patch and cap formation on CV-1 cells or macrophages .
Protein synthesis and glycoprotein mobility To investigate whether the observed differences in the lateral mobility of viral glycoproteins might involve a difference in the level of synthesis of some other viral protein which might be involved in the assembly of virus into budding structures, we analysed the patterns of viral polypeptide synthesis in the various cell types (Fig . 1)
Host cell-dependent lateral mobility of viral glycoproteins
379
Fig . 3 . Surface immunofluorescence of influenza HA and VSV-G on CV-1 cells . Cells were infected with influenza virus or VSV and immunolabeled at 37°C . Lateral mobility of (A) influenza HA and (B) VSV-G is shown by marked patching and formation of unipolar caps .
Fig . 4 . Surface distribution of influenza HA on infected mouse peritoneal macrophages . Cells were infected and immunolabeled as described in Materials and methods . (A) Infected macrophages (6 h postinfection) immunolabeled on ice, showing uniform surface distribution of HA . (B) Capping of influenza HA after immunolabeling at 37'C .
Densitometer analysis of the viral protein bands showed no significant differences in the ratios of the major structural proteins of VSV at 6 post-infection in HeLa or CV-1 cells [Fig . 1 (A)] . Similarly, no significant difference in the pattern of polypeptide synthesis was observed between influenza-infected HeLa or CV-1 cells [Fig . 1 (B)] . These results indicate that the observed differences in glycoprotein mobility are not
380
S . L . Lydy et al.
Fig . 5 . Lateral mobility of viral glycoproteins after vaccinia superinfection of P13 virus-infected HeLa cells . Cells were infected and immunolabeled as described in Materials and methods . A double exposure of the same cells is shown, with the top exposure showing uncapped rhodamine-tagged vaccinia surface glycoprotein and the bottom exposure showing capping of fluorescein-tagged P13 HN (denoted by arrows) .
related to a difference in the level of synthesis of a specific viral structural protein in the cell types examined . Mobility of viral glycoproteins expressed using vaccinia virus recombinants To determine whether host cell-dependent differences in mobility of glycoproteins are an intrinsic property of the protein molecules themselves or depend on interactions with other viral components, HeLa and CV-1 cells were infected with vaccinia virus recombinants expressing influenza HA or VSV G proteins . If viral glycoproteins expressed from cloned genes were found to be immobile on the surfaces of infected HeLa cells and mobile on other cell types, this would indicate that host cell-dependent differences in mobility are an intrinsic property of the glycoprotein molecules . Control experiments were first carried out to determine whether vaccinia virus infection would affect the lateral redistribution of proteins . In HeLa cells which were infected with P13 virus and superinfected with the IH D-J strain of vaccinia virus, the P13-HN glycoprotein exhibited lateral mobility at 37°C (Fig . 5) . Under similar experimental conditions, the fusion protein of P13 was also mobile at 37°C (data not shown) . In HeLa or CV-1 cells infected with influenza virus and superinfected with vaccinia virus, the restricted mobility of influenza HA in HeLa cells and mobility in CV-1 cells were not found to be altered (data not shown) . Therefore, the vaccinia virus infection did not affect lateral mobility of the glycoproteins tested . The vaccinia-expressed recombinant HA and G glycoproteins showed restricted mobility in HeLa cells in the presence of specific antibody at 37°C [Fig . 6(B, D)] In comparison, the F glycoprotein of RSV expressed from a vaccinia recombinant exhibited patching or capping [Fig . 6(F)], demonstrating that the three vaccinia recombinant-expressed viral glycoproteins exhibit marked differences in lateral mobility on HeLa cell surfaces, as was observed following virus infection . A more punctuate surface fluorescence staining pattern was evident in some recombinant HA and G-
Host cell-dependent lateral mobility of viral glycoproteins
38 1
Fig . 6 . Surface immunofluorescence of vaccinia recombinant-expressed viral glycoproteins on HeLa cells . Diffuse surface fluorescence of (A) VVHA ; (C) VVG and (E) VVF after immunolabeling on ice . At 37°C, diffuse surface fluorescence indicates the restricted mobility of (B) VVHA and (D) VVG . (F) VVF exhibiting patching and capping at 37°C .
expressing HeLa cells at both 0 ° and 37 ° C, in contrast to the more diffuse surface fluorescence seen with glycoproteins produced during influenza or VSV virus infection when cells are immunolabeled on ice . Under similar experimental conditions, the recombinant HA and G glycoproteins were found to be mobile on the surfaces of infected CV-1 cells (Fig . 7) . The finding that the HA and G glycoproteins expressed
3 82
S . L. Lydy et al.
Fig . 7 . Surface immunofluorescence of recombinant HA and G glycoproteins on CV-1 cells . Patching and capping of (A) VVHA and (B) VVG is evident at 37°C .
from vaccinia recombinants are immobile on the surfaces of HeLa cells, but mobile on CV-1 cells, indicates that the host cell-dependent differences in mobility are determined by the structure of these glycoprotein molecules, and do not involve other viral gene products .
Discussion The present results demonstrate that the structural features of viral glycoproteins determine their mobility on the host cell plasma membrane . The lateral mobility of viral glycoproteins on the surfaces of infected cells may be an important factor in viral pathogenesis . In cell types which exhibit restricted lateral mobility of viral glycoproteins on the cell surface, viral maturation may be impaired, preventing the release and spread of infectious virions from the infected cells . Thus the virus infection becomes selflimiting . Caliguiri and Holmes 14 have reported that the budding of influenza virus is restricted in HeLa cells at the stage of virion release . It will be of interest to determine whether this is related to the observed restriction in mobility of influenza HA . Incubation of measles virus-infected cells in the presence of specific antibody has been demonstrated to not only stimulate redistribution of measles antigens on the infected cell surfaces, but to also prevent complement or immune lymphocyte-mediated killing of infected cells, suggesting that shedding of capped antigen-antibody complexes from infected cell surfaces enables the cells to avoid immune lysis and the virus to persist ." In this instance, lateral mobility of viral antigen-antibody complexes is a modulation mechanism which allows the virus infection to continue in a chronic persistent state without being cleared from the host . In addition, lateral mobility of viral glycoproteins may also play a role in cytopathology . For example, lateral mobility and clustering of viral antigens on the infected cell surface may be necessary for virusinduced cell fusion, leading to syncytium formation . Such cell fusion is a means by which virus can spread from cell to cell without the release of free infectious virions,
Host cell-dependent lateral mobility of viral glycoproteins
383
thus allowing the virus infection to continue spreading while evading host humoral immune responses . The present results indicate that antibody-induced mobility of viral glycoproteins on surfaces of infected cells is dependent both on the structure of the viral glycoprotein and some undetermined property of the host cell . Both the influenza HA and VSV G protein were found to undergo antibody-induced redistribution in several cell types, but not in HeLa cells . In contrast, the glycoproteins of measles, P13 and RSV exhibited lateral mobility in HeLa cells . These observations led us to investigate the basis for these differences in lateral mobility . In particular, we have investigated the possibility that restriction in lateral mobility might result from interaction of viral glycoproteins with other viral internal proteins . Specifically, the matrix proteins of influenza virus or VSV were considered to be likely candidates for possible involvement in determining lateral mobility of surface glycoproteins because they are located beneath the viral membrane and may interact with the cytoplasmic domains of the glycoproteins . Analysis of the patterns of viral polypeptide synthesis indicated that host celldependent differences in the lateral mobility of the glycoproteins showed no obvious correlation with the levels of synthesis of other viral proteins . Further, using vaccinia virus vectors to express viral glycoproteins, we demonstrated that the HA and G glycoproteins exhibited the same patterns of host cell-dependent mobility as were found on cells infected with the respective wild type viruses . Thus, we conclude that the observed differences in mobility are an intrinsic property of the glycoprotein molecules, and not due to an interaction with other viral components . Griffin and Mullinax 16 had previously reported that a murine lymphokine could alter the restricted mobility of complement C3b receptors on mouse peritoneal macrophage plasma membranes . To determine whether this lymphokine would have a similar effect on the restricted mobility of viral glycoproteins on HeLa cell surfaces, we treated influenza or VSV-infected HeLa cells with the lymphokine, but did not observe a change in the restricted mobility of the influenza HA or VSV-G glycoproteins (results not shown) . It is possible, however, that the virus-infected HeLa cells could not respond to the lymphokine used because of their epithelial nature . The mechanism by which this lymphokine enables C3b receptors to become mobile within the plasma membrane of macrophages remains to be determined . The present results are consistent with the view that viral glycoproteins differ in their association with either the cytoskeleton, or other membrane-associated components of various host cells, and that this interaction affects the lateral mobility of the glycoproteins . Wang and co-workers 17 reported that actin is found associated with enveloped viruses including retroviruses, paramyxoviruses, and influenza virus . Evidence has been obtained that actin does not play a role in the budding of influenza virus 13 or VSV 19 although it may be involved in the assembly of paramyxoviruses .20 22 A proposed molecular mechanism for redistribution of cell surface antigens involves formation of divalent antibody-induced clusters of cell surface antigens which become associated with concentrations of actin and myosin under the plasma membrane either directly or indirectly via a host protein . 23-25 It was previously reported that when influenza virus-infected HeLa cells were treated with cytochalasin D or colchicine to disrupt either the microfilament or microtubule network, the mobility of HA was still completely restricted .' These same agents were found to inhibit the capping of measles HA in infected HeLa cells, 2 suggesting that microtubules and microfilaments are involved in the lateral mobility of some glycoproteins . The precise structural features of glycoprotein molecules which determine their lateral mobility are not obvious from the known properties of the molecules . Lateral mobility may be affected by the conformation or degree of glycosylation of the
384
S . L . Lydy et al .
glycoprotein ectodomain, 26 interaction with the host cell cytoskeleton or membraneassociated cellular proteins, or self-association of glycoproteins on cell surfaces . Except for the H N of P13 virus, all the glycoproteins studied are bitopic transmembrane proteins with an external hydrophilic N-terminal domain, a hydrophobic transmembrane domain near the carboxy terminus and a hydrophilic cytoplasmic tai 1 . 17 30 In contrast, HN is anchored in the membrane by an uncleaved hydrophobic domain near the amino terminus ." The degree of glycosylation does not appear to be correlated with glycoprotein mobility, since influenza HA exhibits restricted mobility whereas RSV-F is mobile in HeLa cells and both possess five N-linked glycosylation sites . The length of the cytoplasmic domains of influenza HA, VSV-G, and paramyxovirus glycoproteins also shows no obvious correlation with mobility . All of the glycoproteins studied appear to be oligomeric structures : influenza HA is a trimer32 and the G protein of VSV is most likely also a trimer ." Walsh and co-workers" found the F glycoprotein of RSV to purify as a dimer although in other paramyxoviruses, the F glycoprotein may be tetrameric ." We are presently investigating the essential molecular features involved in glycoprotein mobility by construction of chimeric protein molecules .
Materials and methods Virus . Stocks of influenza virus (A/WSN/33) were prepared on Madin-Darby bovine kidney (MDBK) cells36 and titers determined by plaque assay in Madin-Darby canine kidney (MDCK) cells . 37 VSV (Indiana strain) was grown in BHK21 cells and titered by procedures previously described . 37 Parainfluenza virus type 3 was grown in LLC-MK 2 cells and titered as described previously . 38 A recombinant vaccinia virus expressing the VSV-G glycoprotein (Indiana strain) was provided by Dr Bernard Moss . The recombinant expressing the hemagglutinin of influenza (A/WSN/33) was described previously . 39 Recombinant vaccinia virus expressing the fusion (F) glycoprotein of RSV 40 was kindly provided by Dr Gail Wertz . Vaccinia virus wild-type and recombinant stocks (VVHA, VVG, and VVF) were grown in HeLa cells and titered in cell lines used for expression studies .
Cells and virus infection . HeLa or CV-1 cells were grown in Dulbecco's modified minimum essential medium (DMEM) with 10% heat inactivated fetal calf serum (FCS), and infected at a multiplicity of infection (moi) of 10 plaque-forming units/cell (pfu/cell) . Two hours post-infection, cells were washed with Ca t +, Mg"-deficient phosphate buffered saline (PBS deficient) and then incubated with Eagle's minimal essential medium (EMEM) minus CaCl 2 with 30 mm EGTA (pH 8) for 30 min at 37°C+5% CO 2 . Infected cells were suspended with a pipette washed two times in Joklik's modified MEM (SMEM)+2% FCS, resuspended in SMEM and maintained at 37°C on a rocker platform . The viability of the infected cells exceeded 95% as determined by trypan blue dye exclusion, and most cells expressed viral antigens randomly distributed on their surface when assayed by indirect immunofluorescence microscopy before experiments were initiated . Human peritoneal cells were obtained by peritoneal lavage from women undergoing medically indicated laparoscopy . Murine peritoneal macrophages were obtained as described by Griffin and Mullinax 41 from 20-30-g female CD-1 Swiss mice . Peritoneal exudate cells (PEC) were isolated by density gradient centrifugation through Ficoll-Hypaque . 42 The interface layer containing lymphocytes and macrophages was carefully harvested, the cells rinsed and pelleted twice in PBS deficient, and resuspended in DMEM supplemented with 10% FCS and 0 .1 mg/ml gentamycin sulfate . Macrophages were separated from lymphocytes by a modified adherence procedure .43 Macrophage cultures were then maintained in SMEM and 10% FCS for 72 h at 37°C with gentle agitation before infection with virus at an moi of 20 pfu/cell . Antibody. Mouse monoclonal antibodies to influenza HA and VSV-G were previously described ." Rabbit anti-VSV and anti-influenza virus sera were prepared as described by Roth and Compans . 45 Rabbit polyvalent antibody to solubilized RSV was a gift from Dr Gail Wertz . Mouse monoclonal antibodies to P13 HN and F glycoproteins 38 were a gift from Dr Ranjit Ray .
Host cell-dependent lateral mobility of viral glycoproteins
385
Rhodamine or fluorescein-conjugated goat anti-rabbit or goat anti-mouse immunoglobulin were purchased from Southern Biotechnology Associates (Birmingham, Alabama) . Immunofluorescence . Aliquots of 2x105 virus-infected cells were washed with PBS and incubated in glass tubes on ice for 30 min with an appropriate dilution of either monoclonal antibodies or polyclonal antisera . Cell clumping was prevented by resuspending the cells every 5 min . After the incubation period, the cells were washed with cold EMEM, pelleted, and incubated with appropriate fluorochrome-conjugated anti- immunoglobulin antisera at 0°C or 37°C for 30 min . After incubation, cells were again washed in cold EMEM, fixed in 1% formaldehyde in PBS, mounted in 1 : 1 PBS/glycerol containing 3% propyl gallate, and examined with a Nikon Optiphot fluorescent microscope .
Radiolabeling and electrophoretic analysis of viral polypeptides . Cells were infected at an moi of 10 pfu/cell . At intervals 2-10 h post-infection, cells were incubated with serum-free EMEM deficient in methionine for 15 min at 37°C, and then pulse-labeled with 25 µCi/ml of 32 5methionine in methionine-deficient EMEM for 30 min at 37°C . The cells were washed five times with ice-cold PBS and solubilized in sample reducing buffer (0 .625 M Tris-HCI, pH 6 .8, containing 0 .5% 2-mercaptoethanol, 10% glycerol, and 2 .3% SDS) . Samples were analysed by gel electrophoresis under reducing conditions on either 15% gels with a 130 : 1 acrylamide/ bisacrylamide ratio 46 or 10% gels with a ratio of 30 : 0 .8 . 4' We thank Betty Jeffrey for preparing the manuscript and Eugene Arms assistance with photography . This research was supported by grant Al 12680 from the National Institutes of Health .
References 1 . Joseph BS, Oldstone MBA . Antibody-induced redistribution of measles virus antigens on the cell surface . J Immunol 1974; 113 : 1205-9 . 2 . Oldstone MBA, Fujinami RS, Lampert PW . Membrane and cytoplasmic changes in virus-infected cells induced by interactions of antiviral antibody with surface viral antigen . In Melnick JL, ed . Progress in medical virology . Vol . 26 . Basel : S . Karger, 1980 ; 45-93 . 3 . Mims CA, White, DO . Viral pathogenesis and immunology . Oxford : Blackwell Scientific Publications, 1984 . 4 . Whitton JL, Oldstone MBA . Virus-induced immune response interactions. In : Fields BN, Knipe DM, Chanock RM, Hirsch MS, Melnick JL, Monath TP, Roizman B, eds . Virology . Vol . 1 . New York: Raven Press, 1990 ; 369-81 . 5 . Ahmed R, Stevens JG . Viral persistence . In Fields BN et al ., eds . Virology . Vol . 1 . New York : Raven Press, 1990 . 6 . Lampert PW, Joseph BS, Oldstone MBA . Antibody-induced capping of measles virus antigens on plasma membrane studied by electron microscopy . J Virol 1975; 15 : 1248-55 . 7 . Basak S, Compans RW, Oldstone MBA . Restricted mobility of influenza hemagglutinin on HeLa cell plasma membranes. In : Compans RW, Bishop DHL, eds . Segmented negative strand virusesarenaviruses, bunyaviruses and orthomyxoviruses . New York: Academic Press, 1984 ; 361-4 . 8 . Ehrnst A, Sundqvist KG . Mechanisms of appearance of viral glycoproteins at the cell surface membrane . Exp Cell Biol 1976 ; 44 :198-225 . 9 . Geiger B, Singer SJ . The participation of a-actinin in the capping of cell membrane components . Cell 1979 ; 16 : 213-22 . 10 . Hiller G, Weber K, Schneider L, Parajsz C, Jungwirth C . Interaction of assembled progeny pox viruses with the cellular cytoskeleton . Virology 1979 ; 98 : 142-53 . 11 . Bohn W, Rutter G, Hohenberg H, Mannweiler K, Nobris P . Involvement of actin filaments in budding of measles virus : studies on cytoskeletons of infected cells . Virology 1986 ; 149 : 91-106 . 12, Rutter G, Mannweiler K . Alterations of actin-containing structures in BHK21 cells infected with Newcastle disease virus and vesicular stomatitis virus. J Gen Virol 1977 ; 37 : 233-42 . 13, Bedows E, Rao KMK, Walse MJ . Fate of microfilaments in Vero cells infected with measles virus and herpes simplex virus 1 . Mol Cell Biol 1983 ; 3 : 712-19 . 14 . Caliguiri L, Holmes KV . Host-dependent restrictions of influenza virus maturation . Virology 1979 ; 92 : 15-30 . 15 . Fujinami RS, Oldstone MBA . Antibody initiates virus persistence : immune modulation and measles virus infection . In : Notkins AL, Oldstone MBA, eds . Concepts in viral pathogenesis . New York : SpringerVerlag, 1984 ; 187-93 . 16 . Griffin FM Jr, Mullinax PJ . Effects of differentiation in vivo and of lymphokine treatment in vitro on the mobility of C3 receptors of human and mouse mononuclear phagocytes . J Immunol 1985 ; 135 : 3394-7 . 17 . Wang E, Wolf A, Lamb RA, Choppin PW, Goldberg AR . The presence of actin in enveloped viruses .
386
S . L . Lydy et al.
In : Goldman R, Pollard T, Rosenbaum J, eds . Cell motility . Cold Spring Harbor: Cold Spring Laboratory Harbor Press, 1976; 589-99 . 18 . Griffin JA, Compans RW . Effect of cytochalasin B on the maturation of enveloped viruses . J Exp Med 1979; 150 : 379-91 . 19 . Gentry N, Bussereau F . Is the cytoskeleton involved in VSV reproduction? J Virol 1986 ; 34 : 777-81 . 20 . Tyrrell DLJ, Ehrnst E . Transmembrane communication in cells chronically infected with measles virus . J Cell Biol 1979 ; 81 : 396-402 . 21 . Stallcup KC, Raine ES, Fields BN . Cytochalasin B inhibits the maturation of measles virus . Virology 1983;124 :59-74 . 22 . Guiffre RM, Tovell DR, Kay CM, Tyrrell DLJ . Evidence for an interaction between the membrane protein of a paramyxovirus virus and actin . J Virol 1982; 42 : 963-8 . 23 . Singer SJ, Ash JF, Bourguignon LYW, Heggeness MH, Louvard D . Transmembrane interactions and the mechanisms of transport of proteins across membranes . J Supramol Struct 1978 ; 373-89 . 24 . Pasternak C, Spudlich JA, Elson EL . Capping of surface receptors and concomitant cortical tension are generated by conventional myosin . Nature 1989 ; 341 : 549-51 . 25 . Sheetz MP, Turney S, Qian H, Elson EL . Nanometre-level analysis demonstrates that lipid flow does not drive membrane glycoprotein movements . Nature 1989 ; 340 : 284-8 . 26 . Wier M, Edidin M . Constraint of the translational diffusion of a membrane glycoprotein by its external domains . Science 1988; 244 : 412-14 . 27 . Hiti AL, Davis AR, Nayak DP . Complete sequence analysis shows that the hemagglutinins of the HO and H2 subtypes of human influenza viruses are closely related . Virology 1981 ; 111 : 113-24 . 28 . Rose JK, Gallione CJ . Nucleotide sequences of the mRNAs encoding the vesicular stomatitis G and M proteins determined from cDNA clones containing the complete coding regions . J Virol 1981 ; 39 : 51928 . 29 . Collins PL, Huang YT, Wertz GW . Nucleoticle sequence of the geneencoding the fusion (F) glycoprotein of human respiratory syncytial virus. Proc Natl Acad Sci USA 1984 ; 81 : 7683-7 . 30 . Spriggs M, Olmsted Venkatesan RS, Coligan J, Collins P . Fusion glycoprotein of human parainfluenza type 3 virus: nucleotide sequences of the gene, direct identification of the cleavage-activation site, and comparison with other para myxovi ruses. Virology 1986 ; 152 : 241-51 . 31 . Elango N, Coligan J, Jambou R, Venkatasan S . Human parainfluenza type 3 virus hemagglutininneuraminidase glycoprotein : nucleotide sequence of the mRNA and limited amino acid sequence of the purified protein . J Virol 1986 ; 57 : 481-9 . 32 . Wiley DC, Skehel JJ, Waterfield M . Evidence from studies with a cross-linking reagent that the hemagglutinin of influenza virus is a trimer . Virology 1977 ; 79 : 446-9 . 33 . Kreis TE, Lodish HF . Oligomerization is essential for transport of vesicular stomatitis viral glycoprotein to the cell surface . Cell 1986 ; 46 : 929-37 . 34 . Walsh EE, Brandriss MW, Schlesinger JJ . Purification and characterization of respiratory syncytial virus fusion protein . J Gen Virol 1985; 66 : 409-15 . 35 . Sechoy 0, Philippot JR, Bienvenue A . F protein-F protein interaction with the Sendai virus identified by native-bonding or chemical cross-linking . J Biol Chem 1987 ; 262 : 11519-23 . 36 . Choppin PW . Replication of influenza virus in a continuous cell line : high yield of infective virus from cells infected at high multiplicity . Virology 1969 ; 39 : 130-4 . 37 . Roth MG, Fitzpatrick JP, Compans RW . Polarity of influenza and vesicular stomatitis virus maturation in MDCK cells : lack of a requirement for glycosylation of viral glycoproteins . Proc Natl Acad Sci USA 1979; 76 : 6430-4 . 38 . Ray R, Brown VE, Compans RW . Glycoproteins of human parainfluenza virus type 3 : characterization and evaluation as a subunit vaccine. J Infect Dis 1985 ; 152 : 1219-29 . 39 . Stephens EB, Compans RW, Earl P, Moss B . Surface expression of viral glycoproteins is polarized in epithelial cells infected with recombinant vaccinia viral vectors . EMBO J 1986 ; 5 : 237-45 . 40. Wertz GW, Stott EJ, Young KKY, Anderson K, Ball LA . Expression of the fusion protein of human respiratory syncytial virus from recombinant vaccinia virus vectors and protection of vaccinated mice . J Virol 1987 ; 61 : 293-301 . 41 . Griffin FM Jr, Mullinax PJ . Augmentation of macrophage complement receptor function in vitro . J Exp Med 1981 ; 154 : 291-305 . 42 . Colotta F, Peri G, Villa A, Mantovani A . Rapid killing of actinomycin-D treated tumor cells by human mononuclear cells . J Immunol 1984 ; 132: 936-44 . 43 . Kumagai K, Itoh K, Hinuma S, Tada M . Pretreatment of plastic petri plates with fetal calf serum . A simple method for macrophage isolation . J Immunol Methods 1979 ; 29 : 17-25 . 44 . Alonso-Caplen FV, Compans RW . Modulation of glycosylation and transport of viral membrane glycoproteins by a sodium ionophore . J Cell Biol 1983 ; 97 : 659-68 . 45 . Roth MG, Compans RW. Antibody-resistant spread of vesicular stomatitis virus infection in cell lines of epithelial origin . J Virol 1980; 35 : 547-50 . 46 . Lamb RA, Choppin PW . Synthesis of influenza virus proteins in infected cells : translation of viral polypeptides, including three P polypeptides, from RNA produced by primary transcription . Virology 1976 ;74 :504-19 . 47 . Laemmli UL . Cleavage of structural proteins during the assembly of the head bacteriophage T4 . Nature 1970;227 :650-62 .
Host cell-dependent lateral mobility of viral glycoproteins Shari L . Lydy, Sukla Basak and Richard W . Compans* Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294, U.S.A . (Received 24 April 1990 ; accepted in revised form 24 August, 1990)
Lydy, S . L . (Dept of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294, U .S .A .), S . Basak and R . W . Compans . Host cell-dependent lateral mobility of viral glycoproteins . Microbial Pathogenesis 1990 ; 9 : 375-386 . The lateral mobility of viral envelope proteins on the plasma membranes of infected cells is an important factor in both virus assembly and pathogenesis. The envelope glycoproteins of measles and human parainfluenza virus are mobile on the surfaces of infected HeLa cells and undergo lateral redistribution in the presence of specific antibody, forming unipolar caps . In contrast, no such redistribution was observed with influenza virus hemagglutinin (HA) or vesicular stomatitis virus (VSV) G glycoproteins on infected HeLa cell surfaces . However, the HA and G glycoproteins were both found to be mobile in the plasma membrane of CV-1 cells, or human or murine peritoneal macrophages . These results indicate that host cell-dependent as well as virus-specific factors are involved in determining viral glycoprotein mobility . No significant differences in the patterns of synthesis of influenza or VSV viral proteins were found in the various cell types examined . The HA and G proteins, when expressed from vaccinia virus recombinants, were each found to be immobile in HeLa cells and mobile in CV-1 cells, thus indicating that the host cell-dependent differences in mobility are an intrinsic property of each viral glycoprotein molecule and not the result of interaction with other viral components . It is suggested that the association of viral glycoproteins with either the cytoskeleton or membraneassociated cellular proteins may be related to the observed differences in lateral mobility . Key words : viral glycoproteins ; lateral mobility ; plasma membrane .
Introduction Many viral antigens are mobile on the plasma membrane, and this mobility may play an important role during viral infection . During budding of enveloped viruses from the plasma membrane, viral antigens are assembled into specific domains on the cell surface from which host proteins are excluded, and lateral mobility of glycoproteins may be involved in forming such domains . In addition, viral antigens incorporated into cell membranes prior to virus budding may react with virus-specific divalent antibody, and undergo lateral redistribution to form a discrete cap on the cell surface ."' Such local concentration of viral antigen expression on infected cell surfaces may facilitate recognition by the host immune system, leading to immune lysis of infected cells in the presence of complement or immune lymphocytes, or to virally induced autoimmune disease ." It has been suggested that viral antigen-antibody complexes shed from infected cell surfaces after capping may cause immune complex disease and may ultimately be a survival mechanism for many viruses since infected cells escape surveillance from host defenses, leading to the establishment of persistent viral infection . 3' 5 Finally, many viral glycoproteins exhibit the capacity to induce membrane * Author to whom correspondence should be addressed . 0882-4010/90/120375+12 803 .00/0
© 1990 Academic Press Limited
37 6
S . L . Lydy et al .
fusion leading to the formation of syncytia, and it is likely that high local concentrations of glycoproteins are needed for this biological activity . It was previously reported that the glycoproteins of measles virus are mobile on the surfaces of infected HeLa cells` whereas, under similar experimental conditions, the mobility of influenza HA was restricted on HeLa cell plasma membranes .' In HeLa cells doubly infected with measles and influenza viruses, measles virus HA was found to be mobile whereas no capping of influenza HA was observed .' It has been proposed that cytoskeletal components of virus-infected cells may control the movement of viral structures at the plasma membrane 8.9 and be involved in the viral budding process .", " In addition, many viruses cause changes in the organization of host cell microfilament and microtubular networks during infection .' 1,13 Based on these observations, it is possible that viral glycoproteins differ in their association with cytoskeletal components in different cells during the infection process, and that this interaction affects the lateral mobility of the viral glycoproteins . The present study was undertaken to investigate further the basis for the observed differences in lateral mobility of viral glycoproteins in various cell lines . We have studied the antibody-induced redistribution of influenza HA, the G glycoprotein of vesicular stomatitis virus (VSV), hemagglutinin-neuraminidase (HN) and fusion (F) glycoproteins of human parainfluenza virus type 3 (P13), and F protein of respiratory syncytial virus (RSV) on HeLa and CV-1 cell plasma membranes . Vaccinia virus recombinants were used to express several of the viral glycoproteins in the absence of other viral components, to determine whether the restricted mobility of certain glycoproteins is an intrinsic property of these glycoprotein molecules .
Results Mobility of viral glycoproteins on HeLa plasma membranes We analysed the lateral mobility of viral glycoproteins by determining their ability to undergo lateral redistribution in response to specific antibody . In control experiments, cells immunolabeled with antibody on ice showed uniform surface distribution of viral antigens . Antibody-induced redistribution of the viral glycoprotein was shown by formation of irregular patches of viral antigen-antibody complexes or a discrete unipolar cap on cells immunolabeled at 37°C, whereas uniform surface fluorescence on cells labeled under such conditions indicated restricted mobility of the surface antigen . Antibody-induced capping of virus antigen was previously reported to occur optimally at 37°C . 1 We incubated virus-infected cells with antibody conjugates at 37°C for intervals from 30 to 90 min to determine the optimal incubation time for maximal cell capping, and found that capping of viral glycoproteins occurred on cell surfaces within 30 min . In addition, the time course of expression of influenza and VSV viral polypeptides in HeLa and CV-1 cells was determined to ensure that capping experiments were done at times when the infected cells were expressing large amounts of viral glycoproteins (Fig . 1) . At 8 h post-infection, a diffuse and random pattern of surface fluorescence on influenza virus-infected HeLa cells was observed after immunolabeling at 37°C with specific antibody to the HA glycoprotein [Fig . 2(A)], as previously reported .' Under similar experimental conditions, the VSV G glycoprotein also was found to be uniformly distributed on HeLa cell surfaces [Fig . 2(B)] . In both cases the patterns of fluorescence closely resembled those seen on cells immunolabeled on ice, indicating the lack of antigen redistribution . In contrast, lateral redistribution was observed when measles virus-infected [Fig . 2(C)] or P13-infected HeLa cells [Fig . 2(D)] were immunolabeled with antibodies specific for the HA or HN protein, respectively . The fusion (F) glycoprotein of P13 virus also exhibited antibody-induced
Host cell-dependent lateral mobility of viral glycoproteins
U
2
4
6 8
10
U 2 4 6
377
8 10
2
4 6
8
10
2 4 6 8
10
P,- 3 HA G NP-
N, NS -
M NS, -
HeLa
CV_ I
HeLa
CV_ I
Fig . 1 . Time course of VSV and influenza protein synthesis in HeLa and CV-1 cells . At 2-10 h postinfection, the infected HeLa and CV-1 cells were radiolabeled with 35S-methionine and the radioactive VSV (A) and influenza (B) polypeptides analysed on 10% and 15% reducing polyacrylamide gels, respectively . U designates uninfected cells .
redistribution on the surfaces of infected HeLa cells (data not shown) . Joseph and Oldstone' previously reported the mobility of measles HA on the majority of infected HeLa cells, and the present results indicate that the HN and F glycoproteins of P13 virus are similar in this regard . Thus, glycoproteins of various enveloped viruses show significant differences in their lateral mobility on the surfaces of infected HeLa cells . Host cell-dependent differences in the lateral mobility of viral glycoproteins The finding of significant differences in the mobility of various viral glycoproteins on surfaces of infected HeLa cells led us to investigate viral glycoprotein mobility further in other cell types . CV-1 cells were infected with either influenza virus or VSV, and the pattern of surface fluorescence of the HA and G proteins was examined after immunolabeling with specific antibody at 37°C . Significant patching or capping of the influenza HA [Fig . 3(A)] and VSV G [Fig . 3(B)] glycoproteins occurred on 90% and 62% of the infected CV-1 cell surfaces, respectively, whereas a uniform distribution of antigens was observed in cells immunolabeled on ice (not shown) . We also examined the pattern of surface immunofluorescence on virus-infected human and mouse peritoneal macrophages . In control cells immunolabeled on ice, the viral antigen remained uniformly distributed over the macrophage surfaces [Fig . 4(A)] . In contrast, nearly all of the influenza virus-infected murine macrophages [Fig . 3(B)] and human macrophages (not shown) were found to show a pattern of influenza HA in unipolar caps after immunolabeling at 37°C . Under similar experimental conditions, the G protein of VSV was also found to redistribute in both human and murine peritoneal macrophages (not shown) . These results indicate that certain viral glycoproteins
378
S . L . Lydy et at.
Fig . 2 . Distribution of viral antigens on the surfaces of infected HeLa cells after immunolabeling at 37°C . Cells were infected and immunolabeled as described in Materials and methods . (A) Influenza HA (8 h postinfection) and (B) VSV-G (6 h post-infection) showing diffuse fluorescence . (C) Measles HA (24 h postinfection) and (D) P13-HN (24 h post-infection) showing redistribution into caps in many cells .
exhibit a high degree of lateral mobility in some cell types, and restricted mobility in another cell type . Both the influenza HA and VSV G glycoproteins are immobile on surfaces of infected HeLa cells, but exhibit patch and cap formation on CV-1 cells or macrophages .
Protein synthesis and glycoprotein mobility To investigate whether the observed differences in the lateral mobility of viral glycoproteins might involve a difference in the level of synthesis of some other viral protein which might be involved in the assembly of virus into budding structures, we analysed the patterns of viral polypeptide synthesis in the various cell types (Fig . 1)
Host cell-dependent lateral mobility of viral glycoproteins
379
Fig . 3 . Surface immunofluorescence of influenza HA and VSV-G on CV-1 cells . Cells were infected with influenza virus or VSV and immunolabeled at 37°C . Lateral mobility of (A) influenza HA and (B) VSV-G is shown by marked patching and formation of unipolar caps .
Fig . 4 . Surface distribution of influenza HA on infected mouse peritoneal macrophages . Cells were infected and immunolabeled as described in Materials and methods . (A) Infected macrophages (6 h postinfection) immunolabeled on ice, showing uniform surface distribution of HA . (B) Capping of influenza HA after immunolabeling at 37'C .
Densitometer analysis of the viral protein bands showed no significant differences in the ratios of the major structural proteins of VSV at 6 post-infection in HeLa or CV-1 cells [Fig . 1 (A)] . Similarly, no significant difference in the pattern of polypeptide synthesis was observed between influenza-infected HeLa or CV-1 cells [Fig . 1 (B)] . These results indicate that the observed differences in glycoprotein mobility are not
380
S . L . Lydy et al.
Fig . 5 . Lateral mobility of viral glycoproteins after vaccinia superinfection of P13 virus-infected HeLa cells . Cells were infected and immunolabeled as described in Materials and methods . A double exposure of the same cells is shown, with the top exposure showing uncapped rhodamine-tagged vaccinia surface glycoprotein and the bottom exposure showing capping of fluorescein-tagged P13 HN (denoted by arrows) .
related to a difference in the level of synthesis of a specific viral structural protein in the cell types examined . Mobility of viral glycoproteins expressed using vaccinia virus recombinants To determine whether host cell-dependent differences in mobility of glycoproteins are an intrinsic property of the protein molecules themselves or depend on interactions with other viral components, HeLa and CV-1 cells were infected with vaccinia virus recombinants expressing influenza HA or VSV G proteins . If viral glycoproteins expressed from cloned genes were found to be immobile on the surfaces of infected HeLa cells and mobile on other cell types, this would indicate that host cell-dependent differences in mobility are an intrinsic property of the glycoprotein molecules . Control experiments were first carried out to determine whether vaccinia virus infection would affect the lateral redistribution of proteins . In HeLa cells which were infected with P13 virus and superinfected with the IH D-J strain of vaccinia virus, the P13-HN glycoprotein exhibited lateral mobility at 37°C (Fig . 5) . Under similar experimental conditions, the fusion protein of P13 was also mobile at 37°C (data not shown) . In HeLa or CV-1 cells infected with influenza virus and superinfected with vaccinia virus, the restricted mobility of influenza HA in HeLa cells and mobility in CV-1 cells were not found to be altered (data not shown) . Therefore, the vaccinia virus infection did not affect lateral mobility of the glycoproteins tested . The vaccinia-expressed recombinant HA and G glycoproteins showed restricted mobility in HeLa cells in the presence of specific antibody at 37°C [Fig . 6(B, D)] In comparison, the F glycoprotein of RSV expressed from a vaccinia recombinant exhibited patching or capping [Fig . 6(F)], demonstrating that the three vaccinia recombinant-expressed viral glycoproteins exhibit marked differences in lateral mobility on HeLa cell surfaces, as was observed following virus infection . A more punctuate surface fluorescence staining pattern was evident in some recombinant HA and G-
Host cell-dependent lateral mobility of viral glycoproteins
38 1
Fig . 6 . Surface immunofluorescence of vaccinia recombinant-expressed viral glycoproteins on HeLa cells . Diffuse surface fluorescence of (A) VVHA ; (C) VVG and (E) VVF after immunolabeling on ice . At 37°C, diffuse surface fluorescence indicates the restricted mobility of (B) VVHA and (D) VVG . (F) VVF exhibiting patching and capping at 37°C .
expressing HeLa cells at both 0 ° and 37 ° C, in contrast to the more diffuse surface fluorescence seen with glycoproteins produced during influenza or VSV virus infection when cells are immunolabeled on ice . Under similar experimental conditions, the recombinant HA and G glycoproteins were found to be mobile on the surfaces of infected CV-1 cells (Fig . 7) . The finding that the HA and G glycoproteins expressed
3 82
S . L. Lydy et al.
Fig . 7 . Surface immunofluorescence of recombinant HA and G glycoproteins on CV-1 cells . Patching and capping of (A) VVHA and (B) VVG is evident at 37°C .
from vaccinia recombinants are immobile on the surfaces of HeLa cells, but mobile on CV-1 cells, indicates that the host cell-dependent differences in mobility are determined by the structure of these glycoprotein molecules, and do not involve other viral gene products .
Discussion The present results demonstrate that the structural features of viral glycoproteins determine their mobility on the host cell plasma membrane . The lateral mobility of viral glycoproteins on the surfaces of infected cells may be an important factor in viral pathogenesis . In cell types which exhibit restricted lateral mobility of viral glycoproteins on the cell surface, viral maturation may be impaired, preventing the release and spread of infectious virions from the infected cells . Thus the virus infection becomes selflimiting . Caliguiri and Holmes 14 have reported that the budding of influenza virus is restricted in HeLa cells at the stage of virion release . It will be of interest to determine whether this is related to the observed restriction in mobility of influenza HA . Incubation of measles virus-infected cells in the presence of specific antibody has been demonstrated to not only stimulate redistribution of measles antigens on the infected cell surfaces, but to also prevent complement or immune lymphocyte-mediated killing of infected cells, suggesting that shedding of capped antigen-antibody complexes from infected cell surfaces enables the cells to avoid immune lysis and the virus to persist ." In this instance, lateral mobility of viral antigen-antibody complexes is a modulation mechanism which allows the virus infection to continue in a chronic persistent state without being cleared from the host . In addition, lateral mobility of viral glycoproteins may also play a role in cytopathology . For example, lateral mobility and clustering of viral antigens on the infected cell surface may be necessary for virusinduced cell fusion, leading to syncytium formation . Such cell fusion is a means by which virus can spread from cell to cell without the release of free infectious virions,
Host cell-dependent lateral mobility of viral glycoproteins
383
thus allowing the virus infection to continue spreading while evading host humoral immune responses . The present results indicate that antibody-induced mobility of viral glycoproteins on surfaces of infected cells is dependent both on the structure of the viral glycoprotein and some undetermined property of the host cell . Both the influenza HA and VSV G protein were found to undergo antibody-induced redistribution in several cell types, but not in HeLa cells . In contrast, the glycoproteins of measles, P13 and RSV exhibited lateral mobility in HeLa cells . These observations led us to investigate the basis for these differences in lateral mobility . In particular, we have investigated the possibility that restriction in lateral mobility might result from interaction of viral glycoproteins with other viral internal proteins . Specifically, the matrix proteins of influenza virus or VSV were considered to be likely candidates for possible involvement in determining lateral mobility of surface glycoproteins because they are located beneath the viral membrane and may interact with the cytoplasmic domains of the glycoproteins . Analysis of the patterns of viral polypeptide synthesis indicated that host celldependent differences in the lateral mobility of the glycoproteins showed no obvious correlation with the levels of synthesis of other viral proteins . Further, using vaccinia virus vectors to express viral glycoproteins, we demonstrated that the HA and G glycoproteins exhibited the same patterns of host cell-dependent mobility as were found on cells infected with the respective wild type viruses . Thus, we conclude that the observed differences in mobility are an intrinsic property of the glycoprotein molecules, and not due to an interaction with other viral components . Griffin and Mullinax 16 had previously reported that a murine lymphokine could alter the restricted mobility of complement C3b receptors on mouse peritoneal macrophage plasma membranes . To determine whether this lymphokine would have a similar effect on the restricted mobility of viral glycoproteins on HeLa cell surfaces, we treated influenza or VSV-infected HeLa cells with the lymphokine, but did not observe a change in the restricted mobility of the influenza HA or VSV-G glycoproteins (results not shown) . It is possible, however, that the virus-infected HeLa cells could not respond to the lymphokine used because of their epithelial nature . The mechanism by which this lymphokine enables C3b receptors to become mobile within the plasma membrane of macrophages remains to be determined . The present results are consistent with the view that viral glycoproteins differ in their association with either the cytoskeleton, or other membrane-associated components of various host cells, and that this interaction affects the lateral mobility of the glycoproteins . Wang and co-workers 17 reported that actin is found associated with enveloped viruses including retroviruses, paramyxoviruses, and influenza virus . Evidence has been obtained that actin does not play a role in the budding of influenza virus 13 or VSV 19 although it may be involved in the assembly of paramyxoviruses .20 22 A proposed molecular mechanism for redistribution of cell surface antigens involves formation of divalent antibody-induced clusters of cell surface antigens which become associated with concentrations of actin and myosin under the plasma membrane either directly or indirectly via a host protein . 23-25 It was previously reported that when influenza virus-infected HeLa cells were treated with cytochalasin D or colchicine to disrupt either the microfilament or microtubule network, the mobility of HA was still completely restricted .' These same agents were found to inhibit the capping of measles HA in infected HeLa cells, 2 suggesting that microtubules and microfilaments are involved in the lateral mobility of some glycoproteins . The precise structural features of glycoprotein molecules which determine their lateral mobility are not obvious from the known properties of the molecules . Lateral mobility may be affected by the conformation or degree of glycosylation of the
384
S . L . Lydy et al .
glycoprotein ectodomain, 26 interaction with the host cell cytoskeleton or membraneassociated cellular proteins, or self-association of glycoproteins on cell surfaces . Except for the H N of P13 virus, all the glycoproteins studied are bitopic transmembrane proteins with an external hydrophilic N-terminal domain, a hydrophobic transmembrane domain near the carboxy terminus and a hydrophilic cytoplasmic tai 1 . 17 30 In contrast, HN is anchored in the membrane by an uncleaved hydrophobic domain near the amino terminus ." The degree of glycosylation does not appear to be correlated with glycoprotein mobility, since influenza HA exhibits restricted mobility whereas RSV-F is mobile in HeLa cells and both possess five N-linked glycosylation sites . The length of the cytoplasmic domains of influenza HA, VSV-G, and paramyxovirus glycoproteins also shows no obvious correlation with mobility . All of the glycoproteins studied appear to be oligomeric structures : influenza HA is a trimer32 and the G protein of VSV is most likely also a trimer ." Walsh and co-workers" found the F glycoprotein of RSV to purify as a dimer although in other paramyxoviruses, the F glycoprotein may be tetrameric ." We are presently investigating the essential molecular features involved in glycoprotein mobility by construction of chimeric protein molecules .
Materials and methods Virus . Stocks of influenza virus (A/WSN/33) were prepared on Madin-Darby bovine kidney (MDBK) cells36 and titers determined by plaque assay in Madin-Darby canine kidney (MDCK) cells . 37 VSV (Indiana strain) was grown in BHK21 cells and titered by procedures previously described . 37 Parainfluenza virus type 3 was grown in LLC-MK 2 cells and titered as described previously . 38 A recombinant vaccinia virus expressing the VSV-G glycoprotein (Indiana strain) was provided by Dr Bernard Moss . The recombinant expressing the hemagglutinin of influenza (A/WSN/33) was described previously . 39 Recombinant vaccinia virus expressing the fusion (F) glycoprotein of RSV 40 was kindly provided by Dr Gail Wertz . Vaccinia virus wild-type and recombinant stocks (VVHA, VVG, and VVF) were grown in HeLa cells and titered in cell lines used for expression studies .
Cells and virus infection . HeLa or CV-1 cells were grown in Dulbecco's modified minimum essential medium (DMEM) with 10% heat inactivated fetal calf serum (FCS), and infected at a multiplicity of infection (moi) of 10 plaque-forming units/cell (pfu/cell) . Two hours post-infection, cells were washed with Ca t +, Mg"-deficient phosphate buffered saline (PBS deficient) and then incubated with Eagle's minimal essential medium (EMEM) minus CaCl 2 with 30 mm EGTA (pH 8) for 30 min at 37°C+5% CO 2 . Infected cells were suspended with a pipette washed two times in Joklik's modified MEM (SMEM)+2% FCS, resuspended in SMEM and maintained at 37°C on a rocker platform . The viability of the infected cells exceeded 95% as determined by trypan blue dye exclusion, and most cells expressed viral antigens randomly distributed on their surface when assayed by indirect immunofluorescence microscopy before experiments were initiated . Human peritoneal cells were obtained by peritoneal lavage from women undergoing medically indicated laparoscopy . Murine peritoneal macrophages were obtained as described by Griffin and Mullinax 41 from 20-30-g female CD-1 Swiss mice . Peritoneal exudate cells (PEC) were isolated by density gradient centrifugation through Ficoll-Hypaque . 42 The interface layer containing lymphocytes and macrophages was carefully harvested, the cells rinsed and pelleted twice in PBS deficient, and resuspended in DMEM supplemented with 10% FCS and 0 .1 mg/ml gentamycin sulfate . Macrophages were separated from lymphocytes by a modified adherence procedure .43 Macrophage cultures were then maintained in SMEM and 10% FCS for 72 h at 37°C with gentle agitation before infection with virus at an moi of 20 pfu/cell . Antibody. Mouse monoclonal antibodies to influenza HA and VSV-G were previously described ." Rabbit anti-VSV and anti-influenza virus sera were prepared as described by Roth and Compans . 45 Rabbit polyvalent antibody to solubilized RSV was a gift from Dr Gail Wertz . Mouse monoclonal antibodies to P13 HN and F glycoproteins 38 were a gift from Dr Ranjit Ray .
Host cell-dependent lateral mobility of viral glycoproteins
385
Rhodamine or fluorescein-conjugated goat anti-rabbit or goat anti-mouse immunoglobulin were purchased from Southern Biotechnology Associates (Birmingham, Alabama) . Immunofluorescence . Aliquots of 2x105 virus-infected cells were washed with PBS and incubated in glass tubes on ice for 30 min with an appropriate dilution of either monoclonal antibodies or polyclonal antisera . Cell clumping was prevented by resuspending the cells every 5 min . After the incubation period, the cells were washed with cold EMEM, pelleted, and incubated with appropriate fluorochrome-conjugated anti- immunoglobulin antisera at 0°C or 37°C for 30 min . After incubation, cells were again washed in cold EMEM, fixed in 1% formaldehyde in PBS, mounted in 1 : 1 PBS/glycerol containing 3% propyl gallate, and examined with a Nikon Optiphot fluorescent microscope .
Radiolabeling and electrophoretic analysis of viral polypeptides . Cells were infected at an moi of 10 pfu/cell . At intervals 2-10 h post-infection, cells were incubated with serum-free EMEM deficient in methionine for 15 min at 37°C, and then pulse-labeled with 25 µCi/ml of 32 5methionine in methionine-deficient EMEM for 30 min at 37°C . The cells were washed five times with ice-cold PBS and solubilized in sample reducing buffer (0 .625 M Tris-HCI, pH 6 .8, containing 0 .5% 2-mercaptoethanol, 10% glycerol, and 2 .3% SDS) . Samples were analysed by gel electrophoresis under reducing conditions on either 15% gels with a 130 : 1 acrylamide/ bisacrylamide ratio 46 or 10% gels with a ratio of 30 : 0 .8 . 4' We thank Betty Jeffrey for preparing the manuscript and Eugene Arms assistance with photography . This research was supported by grant Al 12680 from the National Institutes of Health .
References 1 . Joseph BS, Oldstone MBA . Antibody-induced redistribution of measles virus antigens on the cell surface . J Immunol 1974; 113 : 1205-9 . 2 . Oldstone MBA, Fujinami RS, Lampert PW . Membrane and cytoplasmic changes in virus-infected cells induced by interactions of antiviral antibody with surface viral antigen . In Melnick JL, ed . Progress in medical virology . Vol . 26 . Basel : S . Karger, 1980 ; 45-93 . 3 . Mims CA, White, DO . Viral pathogenesis and immunology . Oxford : Blackwell Scientific Publications, 1984 . 4 . Whitton JL, Oldstone MBA . Virus-induced immune response interactions. In : Fields BN, Knipe DM, Chanock RM, Hirsch MS, Melnick JL, Monath TP, Roizman B, eds . Virology . Vol . 1 . New York: Raven Press, 1990 ; 369-81 . 5 . Ahmed R, Stevens JG . Viral persistence . In Fields BN et al ., eds . Virology . Vol . 1 . New York : Raven Press, 1990 . 6 . Lampert PW, Joseph BS, Oldstone MBA . Antibody-induced capping of measles virus antigens on plasma membrane studied by electron microscopy . J Virol 1975; 15 : 1248-55 . 7 . Basak S, Compans RW, Oldstone MBA . Restricted mobility of influenza hemagglutinin on HeLa cell plasma membranes. In : Compans RW, Bishop DHL, eds . Segmented negative strand virusesarenaviruses, bunyaviruses and orthomyxoviruses . New York: Academic Press, 1984 ; 361-4 . 8 . Ehrnst A, Sundqvist KG . Mechanisms of appearance of viral glycoproteins at the cell surface membrane . Exp Cell Biol 1976 ; 44 :198-225 . 9 . Geiger B, Singer SJ . The participation of a-actinin in the capping of cell membrane components . Cell 1979 ; 16 : 213-22 . 10 . Hiller G, Weber K, Schneider L, Parajsz C, Jungwirth C . Interaction of assembled progeny pox viruses with the cellular cytoskeleton . Virology 1979 ; 98 : 142-53 . 11 . Bohn W, Rutter G, Hohenberg H, Mannweiler K, Nobris P . Involvement of actin filaments in budding of measles virus : studies on cytoskeletons of infected cells . Virology 1986 ; 149 : 91-106 . 12, Rutter G, Mannweiler K . Alterations of actin-containing structures in BHK21 cells infected with Newcastle disease virus and vesicular stomatitis virus. J Gen Virol 1977 ; 37 : 233-42 . 13, Bedows E, Rao KMK, Walse MJ . Fate of microfilaments in Vero cells infected with measles virus and herpes simplex virus 1 . Mol Cell Biol 1983 ; 3 : 712-19 . 14 . Caliguiri L, Holmes KV . Host-dependent restrictions of influenza virus maturation . Virology 1979 ; 92 : 15-30 . 15 . Fujinami RS, Oldstone MBA . Antibody initiates virus persistence : immune modulation and measles virus infection . In : Notkins AL, Oldstone MBA, eds . Concepts in viral pathogenesis . New York : SpringerVerlag, 1984 ; 187-93 . 16 . Griffin FM Jr, Mullinax PJ . Effects of differentiation in vivo and of lymphokine treatment in vitro on the mobility of C3 receptors of human and mouse mononuclear phagocytes . J Immunol 1985 ; 135 : 3394-7 . 17 . Wang E, Wolf A, Lamb RA, Choppin PW, Goldberg AR . The presence of actin in enveloped viruses .
386
S . L . Lydy et al.
In : Goldman R, Pollard T, Rosenbaum J, eds . Cell motility . Cold Spring Harbor: Cold Spring Laboratory Harbor Press, 1976; 589-99 . 18 . Griffin JA, Compans RW . Effect of cytochalasin B on the maturation of enveloped viruses . J Exp Med 1979; 150 : 379-91 . 19 . Gentry N, Bussereau F . Is the cytoskeleton involved in VSV reproduction? J Virol 1986 ; 34 : 777-81 . 20 . Tyrrell DLJ, Ehrnst E . Transmembrane communication in cells chronically infected with measles virus . J Cell Biol 1979 ; 81 : 396-402 . 21 . Stallcup KC, Raine ES, Fields BN . Cytochalasin B inhibits the maturation of measles virus . Virology 1983;124 :59-74 . 22 . Guiffre RM, Tovell DR, Kay CM, Tyrrell DLJ . Evidence for an interaction between the membrane protein of a paramyxovirus virus and actin . J Virol 1982; 42 : 963-8 . 23 . Singer SJ, Ash JF, Bourguignon LYW, Heggeness MH, Louvard D . Transmembrane interactions and the mechanisms of transport of proteins across membranes . J Supramol Struct 1978 ; 373-89 . 24 . Pasternak C, Spudlich JA, Elson EL . Capping of surface receptors and concomitant cortical tension are generated by conventional myosin . Nature 1989 ; 341 : 549-51 . 25 . Sheetz MP, Turney S, Qian H, Elson EL . Nanometre-level analysis demonstrates that lipid flow does not drive membrane glycoprotein movements . Nature 1989 ; 340 : 284-8 . 26 . Wier M, Edidin M . Constraint of the translational diffusion of a membrane glycoprotein by its external domains . Science 1988; 244 : 412-14 . 27 . Hiti AL, Davis AR, Nayak DP . Complete sequence analysis shows that the hemagglutinins of the HO and H2 subtypes of human influenza viruses are closely related . Virology 1981 ; 111 : 113-24 . 28 . Rose JK, Gallione CJ . Nucleotide sequences of the mRNAs encoding the vesicular stomatitis G and M proteins determined from cDNA clones containing the complete coding regions . J Virol 1981 ; 39 : 51928 . 29 . Collins PL, Huang YT, Wertz GW . Nucleoticle sequence of the geneencoding the fusion (F) glycoprotein of human respiratory syncytial virus. Proc Natl Acad Sci USA 1984 ; 81 : 7683-7 . 30 . Spriggs M, Olmsted Venkatesan RS, Coligan J, Collins P . Fusion glycoprotein of human parainfluenza type 3 virus: nucleotide sequences of the gene, direct identification of the cleavage-activation site, and comparison with other para myxovi ruses. Virology 1986 ; 152 : 241-51 . 31 . Elango N, Coligan J, Jambou R, Venkatasan S . Human parainfluenza type 3 virus hemagglutininneuraminidase glycoprotein : nucleotide sequence of the mRNA and limited amino acid sequence of the purified protein . J Virol 1986 ; 57 : 481-9 . 32 . Wiley DC, Skehel JJ, Waterfield M . Evidence from studies with a cross-linking reagent that the hemagglutinin of influenza virus is a trimer . Virology 1977 ; 79 : 446-9 . 33 . Kreis TE, Lodish HF . Oligomerization is essential for transport of vesicular stomatitis viral glycoprotein to the cell surface . Cell 1986 ; 46 : 929-37 . 34 . Walsh EE, Brandriss MW, Schlesinger JJ . Purification and characterization of respiratory syncytial virus fusion protein . J Gen Virol 1985; 66 : 409-15 . 35 . Sechoy 0, Philippot JR, Bienvenue A . F protein-F protein interaction with the Sendai virus identified by native-bonding or chemical cross-linking . J Biol Chem 1987 ; 262 : 11519-23 . 36 . Choppin PW . Replication of influenza virus in a continuous cell line : high yield of infective virus from cells infected at high multiplicity . Virology 1969 ; 39 : 130-4 . 37 . Roth MG, Fitzpatrick JP, Compans RW . Polarity of influenza and vesicular stomatitis virus maturation in MDCK cells : lack of a requirement for glycosylation of viral glycoproteins . Proc Natl Acad Sci USA 1979; 76 : 6430-4 . 38 . Ray R, Brown VE, Compans RW . Glycoproteins of human parainfluenza virus type 3 : characterization and evaluation as a subunit vaccine. J Infect Dis 1985 ; 152 : 1219-29 . 39 . Stephens EB, Compans RW, Earl P, Moss B . Surface expression of viral glycoproteins is polarized in epithelial cells infected with recombinant vaccinia viral vectors . EMBO J 1986 ; 5 : 237-45 . 40. Wertz GW, Stott EJ, Young KKY, Anderson K, Ball LA . Expression of the fusion protein of human respiratory syncytial virus from recombinant vaccinia virus vectors and protection of vaccinated mice . J Virol 1987 ; 61 : 293-301 . 41 . Griffin FM Jr, Mullinax PJ . Augmentation of macrophage complement receptor function in vitro . J Exp Med 1981 ; 154 : 291-305 . 42 . Colotta F, Peri G, Villa A, Mantovani A . Rapid killing of actinomycin-D treated tumor cells by human mononuclear cells . J Immunol 1984 ; 132: 936-44 . 43 . Kumagai K, Itoh K, Hinuma S, Tada M . Pretreatment of plastic petri plates with fetal calf serum . A simple method for macrophage isolation . J Immunol Methods 1979 ; 29 : 17-25 . 44 . Alonso-Caplen FV, Compans RW . Modulation of glycosylation and transport of viral membrane glycoproteins by a sodium ionophore . J Cell Biol 1983 ; 97 : 659-68 . 45 . Roth MG, Compans RW. Antibody-resistant spread of vesicular stomatitis virus infection in cell lines of epithelial origin . J Virol 1980; 35 : 547-50 . 46 . Lamb RA, Choppin PW . Synthesis of influenza virus proteins in infected cells : translation of viral polypeptides, including three P polypeptides, from RNA produced by primary transcription . Virology 1976 ;74 :504-19 . 47 . Laemmli UL . Cleavage of structural proteins during the assembly of the head bacteriophage T4 . Nature 1970;227 :650-62 .
