Poliovirus can enter and infect mammalian cells by way of an

Transcription

1 Proc. Nadl. Acad. Sci. USA Vol. 88, pp , May 1991 Medical Sciences Poliovirus can enter and infect mammalian cells by way of an intercellular adhesion molecule 1 pathway (virus receptor/viral attachment site/immunoglobulin-like domains) HANS-CHRISTOPH SELINKA*, ANDREE ZIBERT, AND ECKARD WIMMER Department of Microbiology, State University of New York at Stony Brook, Stony Brook, NY Communicated by Howard L. Bachrach, January 14, 1991 (received for review November 28, 1990) ABSTRACT Mouse fibroblast cell lines were transfected with truncated forms of the human poliovirus receptor () cdna and tested for the expression of functional receptors for poliovirus. Several receptor constructs, all containing the coding region of the first 143 amino acids of, were able to render mouse cells susceptible to poliovirus infection. A deletion of 65 amino acids in the first extracellular domain of prevented virus attachment and infection. These data suggest that domain 1 is necessary and sufficient for the virus-receptor interaction. A /intercellular adhesion molecule 1 hybrid receptor, expressing the variable domain on a truncated receptor molecule for human rhinovirus 14, was shown to be a functional receptor for poliovirus. This observation indicates that, subsequent to attachment to the -binding domain, poliovirus can use the same pathway as the major receptor group rhinoviruses to enter cells. The cellular receptor for poliovirus () has recently been identified as a protein belonging to the immunoglobulin supergene family (1, 2). The sequence of cdna encoding revealed a polypeptide with a predicted molecular mass of 46 kda prior to glycosylation. This polypeptide consists of three extracellular domains characteristic for certain immunoglobulin-like proteins (3) in the order V-C2-C2 (where V = variable and C = constant) that are followed by a hydrophobic membrane-spanning segment and a C-terminal tail (Fig. 1; dl ). The extracellular portion of the carries eight potential N-glycosylation sites (see Fig. 2), and, indeed, the protein migrates as a 67-kDa polypeptide when HeLa cell membranes are subjected to Western blot analyses (4) or when -specific cdna is expressed in vivo by means of a vaccinia virus vector (4) or in a baculovirus expression system (5) and in vitro by translation in a rabbit reticulocyte extract in the presence of dog pancreas microsomes (4). The molecular parameters of the interaction between poliovirus and its receptor, however, have not been elucidated. Poliovirus is a small RNA virus belonging to the genus Enterovirus of the family Picornaviridae. Another genus of this family is Rhinovirus, which encompasses two groups: the minor receptor group and the major receptor group rhinoviruses, of which the latter represents >80% of all known rhinoviruses (6, 7). Although entero- and rhinoviruses are very closely related with respect to their structure (8) and molecular biology (9), they use a variety of receptors most of which remain unknown (reviewed in ref. 10). However, the receptor for the major group rhinoviruses has recently been identified to be intercellular adhesion molecule 1 (ICAM-1) (11-13). Like, it is a protein of the immunoglobulin supergene family, but it differs from in that ICAM-1 consists of five C2-like folds (Fig. 1). The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C solely to indicate this fact. The pathway of penetration and uncoating of all picornaviruses is poorly understood. Evidence has been presented suggesting that after poliovirus and rhinovirus attach to their respective receptors, the particles enter host cells by means of endocytosis (reviewed in ref. 14). However, the requirement of low ph for uncoating of poliovirus has recently been called into question (15). Moreover, it is not known whether the pathway of uptake for poliovirus and for the major group rhinoviruses is the same. Mouse cells can be rendered sensitive to poliovirus infection by transfection of human DNA into mouse cell lines (16) or into the mouse genome (17, 18). Single domains ofthe molecule essential for its function as receptor, however, have not been characterized yet. We now report results that allow us to conclude that the V domain (domain 1) is the only extracellular portion of that is necessary for it to function as receptor, since expression of with a deletion of domains 2 and 3 allowed infection of mouse L cells. We also show that poliovirus can enter and infect cells by way of the ICAM-1 pathway if the V domain of is engineered onto a shortened ICAM-1 molecule. MATERIALS AND METHODS Construction of Expression Vectors. All plasmids were constructed and purified using standard procedures. Restriction sites used for deletions are given as cdna nucleotide numbers of puch20a (1). The pmt2 eukaryotic expression vector (19) was modified for use in mouse MOP8 cells by inserting the polyoma virus origin of replication. A 187-basepair (bp) Sma 1-HindI11 fragment from plasmid pucpy192 was blunt-ended and cloned into the Ssp I site of pmt2 to create pmt2py. The Sma I-Sca I fragment of puch20a, containing the complete coding sequence, was inserted into the blunt-ended EcoRI site, resulting in pmt, which was used to create the following deletion mutants. Plasmid pmtdl was constructed by deletion of the 571-bp Nco I [nucleotide (nt) 7251-Sac I (nt 1196) fragment. The 187-bp Kpn I (nt 1004)-Avr II (nt 1191) fragment was deleted in pmtdl + 2. In pmtdl + 3 the 279-bp Nco I (nt 725)-Kpn I (nt 1004) fragment was deleted. All three plasmid DNAs were incubated with Escherichia coli DNA polymerase I (large fragment) prior to religation. Plasmids ptmdl, ptmdl + 2, and ptmdl + 3 were derived from the deletion mutants described above by inserting the corresponding EcoRI (nt 517)-BamHI (nt 1467) into ptm that was cut with the identical restriction enzymes. The construction of ptm and ptmae was described recently (4). Abbreviations:, poliovirus receptor; ICAM-1, intercellular adhesion molecule 1; EMCV, encephalomyocarditis virus; V, variable; C, constant; nt, nucleotide; 5' NTR, 5' nontranslated region; HRV14, human rhinovirus 14. *To whom reprint requests should be addressed. 3598

2 Medical Sciences: Selinka et al. Proc. Natl. Acad. Sci. USA 88 (1991) 3599 NH2 2 4 COOH COOH COOH COOH d 1 /ICAM-1 dl+2 dl+3 dl+2+3 d1/ic ICAM-1 FIG. 1. Predicted structures of receptors for poliovirus () and the major group of rhinovirus (ICAM-1) spanning the plasma membrane. Characteristic intramolecular disulfide bridges (SS) form immunoglobulin-like extracellular domains of the V and C2 type. Types of immunoglobulin-like domains (V, C1, C2) are classified based on the number of 8 sheets and characteristic amino acid residues common to each domain class (3). Generally, V domains are larger (by two,8 strands) than C2 domains. consists of the three domains V-C2-C2. (A domain is depicted as "d." Numbers next to transmembrane and intracytoplasmic domains refer to the number of amino acids.) To generate pdl/ic, we amplified a DNA fragment of plasmid phrr19 (an ICAM-1 cdna clone given to us by A. McClelland) by using the polymerase chain reaction with the oligodeoxynucleotide primers 5'-CCGGCTAGCTGTCCT- GCCAGCGACTCCCCC-3' and 5'-GGCGCGGATC- CCCCGGGATAGGTTCAGGGAGG-3'. The 30-mer generates an Nhe I restriction site (underlined) just 5' to sequences encoding the ICAM-1 extracellular domains 3-5; the other primer generates a BamHI site after the ICAM-1 coding sequence. Polymerase chain reaction was performed for 25 cycles with a DNA amplification kit (Perkin-Elmer/Cetus). The polymerase chain product was treated with the Klenow fragment of E. coli DNA polymerase I, incubated with the restriction enzymes Nhe I and BamHI, and purified by gel electrophoresis. The 960-bp Nhe I-BamHI fragment was ligated into plasmid ptm, which contains a silent mutation harboring a restriction site for Nhe I (nt 663), replacing the 804-bp Nhe I-BamHI fragment of the sequence to create pdl/ic. Cell Lines and Transfections. Mouse MOP8 and OST7-1 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 8% calf serum. G418 (GIBCO; 400,ug/ml) was added in cultures of OST7-1 cells. DNA transfections into MOP8 and OST7-1 cell lines were performed by electroporation in a BRL Cell-Porator (Bethesda Research Laboratories). Cells in the exponential phase of growth were trypsinized, washed, and resuspended in serum-free DMEM at a concentration of about 1 x 107 cells per ml; 3 1Lg of expression plasmids was included in this medium. Following electroporation ( V; 800 tkf) and a 10-min recovery period, cells from a single cuvette were split into two or three equal portions and plated in DMEM containing 8% calf serum. After hr in culture the cells were used in binding and infectivity assays. Labeling of Poliovirus and Antibodies for Binding Assays. Poliovirus type 1 (Mahoney) [PV1(M)] was labeled with [35S]methionine (ICN; 1100,uCi/mmol; 1 Ci = 37 GBq) as described (20) except that incorporation of radioactivity was continued overnight. Monoclonal antibody D171 (21) was iodinated to a specific activity of 4,Ci/,g by using chloramine T. For virus-binding assays, semiconfluent monolayer cells (1 x 106) were washed in DMEM and 35S-labeled poliovirus (5 x 104 cpm per plate) was added. After 30 min of incubation at room temperature (or 2 hr at 4 C) cells were gently washed five times with DMEM to remove unattached virus, scraped off the plate, and lysed in 0.1 M NaOH/1% SDS, and radioactivity was measured in Ecolume scintillation fluid (ICN). Binding of 1251I-labeled antibodies (5 x 104 cpm per plate) to monolayer cells was performed at 4 C for 2 hr. Cells were washed and lysed as described above. Infectivity Assay. Mouse cells transfected with different constructs were infected with PV1(M) hr after transfection. Twenty plaque-forming units per cell were adsorbed at room temperature for 30 min, and cells were washed and incubated in DMEM/8% calf serum at 37 C. At the appropriate time points, medium was aspirated and cells were scraped off the plates and resuspended in 2 ml of phosphate-buffered saline. Virus was recovered from the cell suspension by freeze-thawing and titered by plaque assays on HeLa monolayer cells as described (22). RESULTS Binding Properties of Deletion Mutants. Modified molecules were constructed as indicated in Fig. 2 and inserted into mammalian expression vectors in which the gene is transcribed under the control of the AdMLP or T7 promoters and translated under the control of the 5' nontranslated region (5' NTR) of encephalomyocarditis virus (EMCV) RNA (23). In addition, we also cloned the origin of replication of polyoma virus into pmt2-derived plasmids to facilitate replication in MOP8 cells, a mouse fibroblast cell line stably expressing polyoma large tumor antigen (24). In each of the vectors, the -specific extracellular domains were joined

3 3600 Medical Sciences: Selinka et al.,41 UI,1 dl +2 S domain t domain 2 domain 3 TM Cyt C dl +3 L _ AE dl +2+3 _ FIG. 2. Schematic overview of the constructs coding for deletion mutants. The structure of the native with a signal peptide (S), three extracellular domains, the transmembrane domain (TM), and cytoplasmic domain (Cyt) is illustrated at the top. Dots show potential N-glycosylation sites. Numbers inside the scheme refer to the first and last amino acid deleted in each construct. These constructs, encoded by expression vectors pmt2 and ptm1, were expressed in mouse MOP8 and OST7-1 cells. to transmembrane and cytoplasmic domains in order to allow synthesis of potential cell surface receptors. A schematic diagram of different mutant polypeptides expected to be generated in transfected mouse cells is presented in Fig. 1. Mouse cells are not susceptible to poliovirus infection since expression is restricted to primate and human cells. Therefore, MOP8 cells transfected with expression vectors containing -specific sequences allowed us to test the ability of the recombinant molecules to bind poliovirus as well as a receptor-specific monoclonal antibody (D171) that efficiently blocks infection of HeLa cells with all three types of poliovirus (21). As a control, the wild-type receptor (dl ) was expressed. Cells transfected with the construct encoding only domain 1 (dl) showed a low but significant binding of 35S-labeled poliovirus (Fig. 3A). A substantial increase in virus binding was observed with cells expressing a receptor encoding the first two domains of (dl + 2). Another mutant receptor, consisting of the first and E U 3000 A B.:..: **V Proc. Natl. Acad. Sci. USA 88 (1991) third domain of (dl + 3), was very weak but also positive in virus binding. Data obtained in parallel binding assays with the 125I-labeled monoclonal antibody D171 (Fig. 3B) correlated with those observed for virus binding, an observation confirming the importance of the N-terminal domain of for virus attachment. Identical results were obtained using T7 RNA polymerase-mediated expression of or segments thereof in OST7-1 cells (data not shown), an expression system that we will describe below for the analysis of a /ICAM-1 hybrid receptor. Pollovirus Replication in Cells Expressing Truncated Receptors. Although mouse L cells fail to express receptors for poliovirus, they do support a single round of poliovirus replication after transfection with viral RNA (25). We took advantage of the ability of poliovirus to grow in mouse cells to test individual domains of transfected molecules for uptake and replication of the virus. All three truncated forms of the molecule carrying the N-terminal V domain were biologically active as specific receptors for poliovirus (Fig. 4). In a single-step growth curve, the yield ofinfectious virus from mouse cells expressing the complete (dl ) or the receptor with the first two domains of (dl + 2) was almost identical. Therefore, deletion of domain 3 does not severely affect receptor function. Virus titers measured just after infection (0 hr) reflected the different binding properties of the recombinant receptors (see Fig. 3). In agreement with the virus-binding data, less virus was recovered from cells expressing the V domain alone (dl). Production of infective particles in these cells, however, demonstrates that the presence of domain 1 is sufficient for specific uptake and replication of poliovirus. The receptor consisting of the domains 1 and 3 (dl + 3) resulted in lower virus titers but also had the capacity to confer susceptibility to poliovirus infection on transfected mouse cells. Nonspecific binding of poliovirus to receptor-negative MOP8 cells transfected with plasmid lacking (pmt2) did not lead to viral replication in these cells. It should be pointed out that with this expression system, we were unable to detect by immunoprecipitation regardless of whether we used constructs encoding the complete receptor or deletion mutants. All constructs yielded similar levels of receptor-specific transcription, as tested by dot-blot analyses (data not shown). This observation conforms to our previous studies of expression (4), although the reason for the low expression remains unexplained. E ***_ 6 * S..:.-.>.:.:::S FIG. 3. Binding properties of truncated s. Mouse MOP8 cells were transfected with the putative receptors (dl, dl + 2, dl + 3, dl ) or the vector alone (pmt2) and incubated with 35S-labeled poliovirus (A) or 125I-labeled monoclonal antibody D171 (B). Data shown are mean + SD of three independent experiments time (hours) FIG. 4. One-step growth curve of poliovirus PV1 (Mahoney) in mouse MOP8 cells transfected with wild-type and recombinant poliovirus receptors: dl (o), dl + 2 (-), dl + 3 (A), dl (o), and vector pmt2 (A). At the indicated times, virus was recovered from these cells and titered by plaque assay on HeLa cells. Each datum point represents the mean value of four experiments. PFU, plaque-forming units.

4 A /ICAM-1 Hybrid Molecule Confers Poliovirus Susceptibility to Mouse Cells. The low efficiency with which the V domain alone (Fig. 1; dl) functions as a receptor may have a variety of reasons, of which one could simply be spacing of the virus-binding site from the cell surface. Thus, removal of the two downstream C2 regions in may prevent accessibility and/or exposure of the virus-binding V domain and could also interfere with proper folding of the V domain. Moreover, the dl variant contained amino acid sequences (residues ) that we have preliminarily assigned as belonging to the /3 strand A of domain 2 (Fig. 2). To test whether these domain 2-specific amino acids and/or spacing play a role in virus binding, we fused the 143 amino acids of the V domain onto a truncated molecule of ICAM-1 giving rise to a polypeptide denoted dl/ic. Specifically, this construct encodes, in addition to the V domain, three C2-like domains, the transmembrane domain, and the cytoplasmic region of ICAM-1. For expression of this construct we used the vector ptm1, in which transcription is controlled by the phage T7 promoter, and translation is controlled by the EMCV 5' NTR (Fig. SA). When such a vector is transfected into mouse cells, which constitutively express cytoplasmic A B a a 4 / 2 I Q!QQ _ELr p 5 N-R -7 Medical Sciences: Selinka et al. r.a-, - i ^ 7 rea ptm1 dl dl/ic dl+2+3 AE Pv I (M) D r----i r--,-l - 4 mae - :::. ICAMpTM1 dl dl/ic dl+2+3 AE ptm1 AE dl dli/c d1+2+3 FIG. 5. (A) Schematic diagram of the chimeric /ICAM-1 receptor (dl/ic). Amino acids of were joined to amino acid 213 of ICAM-1 to form a /ICAM-1 hybrid receptor with four extracellular domains and the ICAM-1 transmembrane and cytoplasmic domains. (B) Binding of poliovirus and receptor-specific antibody D171 to chimeric and truncated molecules expressed in OST7-1 cells. After transfection cells were split and tested for binding of 35S-labeled poliovirus, PV1(M), and 1251-labeled D171 antibody. Specific binding to cells transfected with the wild-type was by definition 100%. (C) Infectivity assay of receptortransfected OST7-1 cells showing the increase in virus titers 0 and 12 hr after infection with PV1(M). Expression of the V domain alone (dl), on a truncated ICAM-1 molecule (dl/ic), or the native receptor (dl ) confers susceptibility to poliovirus infection. A deletion of 65 amino acids in the N-terminal V domain of (AE) prevents poliovirus infection. PFU, plaque-forming units. Proc. Natl. Acad. Sci. USA 88 (1991) 3601 T7 RNA polymerase (OST7-1 cells; ref. 38), synthesis of the gene product can be detected within 8 hr after transfection (4). As can be seen in Fig. SB, the expression of dl/ic leads to increased binding of virus and monoclonal antibody D171 when compared to the expression of the V domain alone (dl). On the other hand, binding of both probes to the hybrid receptor was lower than binding to the complete expressed (Fig. SB; dl ). It is interesting to note that the increase in binding from a single V domain receptor to a V-C2 (double domain) receptor may not be strongly dependent upon the nature of the C2 domain (see Figs. 3A and B and Fig. SB), although not every C2 domain can significantly augment binding (e.g., the small C-terminal C2 domain of ). It appears, therefore, that the spacing of the V domain from the cell surface is important for binding of virus or monoclonal antibody. We next tested whether the expression of the dl/ic construct allows infection of the OST7-1 mouse cells with poliovirus. As can be seen in Fig. 5C, cells expressing the hybrid receptor allowed poliovirus uptake and replication, roughly to the same extent as cells expressing the V domain alone. A Deletion in the V Domain Abolishes Poliovirus Binding. A variant with a partially deleted V domain was constructed by removing an Aat II-EcoRI fragment from the coding region of the cdna (Fig. 2). The resulting mutant, AE, which had a deletion of 65 amino acids in the N-terminal portion of the V domain, was unable to bind significantly poliovirus or monoclonal antibody D171 when expressed in OST7-1 cells (Fig. 5B). Accordingly, it also failed to render the OST7-1 cells sensitive to poliovirus infection (Fig. 5C). DISCUSSION Different viruses have adopted various cell surface molecules as the receptor by which their entry into host cells is mediated. This process of adoption seems to have occurred randomly, as virus receptors differ widely in their chemical properties. For example, the receptor for influenza virus is sialic acid, a simple carbohydrate, whereas the receptor for human immunodeficiency virus is CD4, a complex polypeptide (see ref. 14). Members of the same virus family can use very different receptors, and picornaviruses exemplify this phenomenon (10). Similarities in virion structure may have directed some picornaviruses to cell surface entities that appear to be related; however, subtle structural differences may result in other picornaviruses interacting with quite unrelated receptors. The particles of poliovirus, an enterovirus, and human rhinovirus 14 (HRV14), a rhinovirus, form pronounced "valleys" and "canyons" (26, 27) that surround the apex at the five-fold axis. For rhinovirus 14 it has been fairly well established that these canyon structures harbor the sites of receptor attachment (28, 29). Based on the similarity of the virion structures and the host cell receptors it is highly likely that poliovirus also uses these canyons as receptorbinding sites. As proposed by the "canyon hypothesis" (30), these deep canyon structures are inaccessible to antibodies, whereas narrower receptor domains can enter and interact with specific binding sites. It may not only be coincidence, then, that these viruses adopted monomeric immunoglobulinlike structures ( and ICAM-1) as receptors, and one may speculate that the receptor for enteroviruses of the coxsackie B type (31), has immunoglobulin-like structure also. In contrast, foot-and-mouth disease virus (a member of the genus Aphthovirus of the family Picornaviridae) lacks a canyon (32). Instead, the viral attachment site is probably within a pronounced surface loop, and the Aphthovirus receptor is likely to belong to the integrin supergene family (33). Based on these considerations, it may not be surprising that the data presented here reveal similarities between the

5 3602 Medical Sciences: Selinka et al. uptake of poliovirus and HRV14. We have shown that the V domain serves as the attachment site for poliovirus, just as the N-terminal C2 domain of ICAM-1 is the major point of contact in the HRV14-receptor interaction (28, 29, 34, 35). In an analogous manner to these picornavirus receptors, the N-terminal V domain of the immunoglobulin-like CD4 receptor interacts with the human immunodeficiency virus envelope protein gp120 (36, 37). Thus, in addition to virion structure, parameters such as accessibility may have influenced the choice of domains in immunoglobulin-like molecules for the binding of the virion. It is possible that docks into poliovirus in a similar way as ICAM-1 does into rhinovirus (35). However, the V domain with its two additional f3 strands may require more space than C-like domains and its orientation within the canyon may differ from the ICAM-1-rhinovirus interaction. Molecular modeling of the V domain of has allowed us to "dock" this domain into the canyon of poliovirus (M. decrombrugghe, D. Oren, J. Harber, E. Arnold, and E.W., unpublished data), but the precise nature and position of amino acids that form the points of contact between these structures remain to be determined. Our data clearly indicate that the V domain of alone is competent to serve as receptor. Binding of virus particles by the V domain, however, is not efficient, and a second domain augments the interaction. Expression of the N-terminal C2 domain of ICAM-1 has not been reported so far. It thus remains unknown whether a single domain of ICAM-1 alone is competent as a receptor for rhinovirus. The pathway by which entero- and rhinoviruses enter the cell and uncoat is not well understood but most likely involves endocytosis (for further references, see ref. 14). Clearly, and ICAM-1 must ultimately facilitate passage of the viral RNA through the plasma cell membrane. Interestingly, ICAM-1 is involved in intercellular communication and adhesion, and receptor activity for transport of ligands into the cell is not necessarily a property of intercellular attachment proteins. Nevertheless, ICAM-1 mediates transport of HRV14 into cells, and we have shown here that the uptake of poliovirus into mammalian cells can be facilitated even by a truncated ICAM-1 molecule. This chimeric molecule was functional, although the sequence of ICAM-1 differs significantly from that of, including the C-terminal, cytoplasmic tail. Thus, the V domain of, either alone or in cooperation with several V domains bound to the virus, can confer destabilization (and uncoating) of poliovirions even when presented on a heterologous backbone. We consider it possible that the V domain of can function as when attached to immunoglobulin-like molecules other than ICAM-1, as, for example, CD4. Note Added in Proof. Similar results confirming that the poliovirusbinding site resides in domain 1 in were obtained by Koike et al. (39). We thank Allen McClelland and Patrick Hearing for providing plasmids and Christopher U. T. Hellen for critical reading of the manuscript. H.-C.S. and A.Z. were supported by fellowships of the Deutsche Forschungsgemeinschaft and Deutscher Akademischer Austauschdienst, respectively. This research was supported in part by Public Health Service Grants AI and CA from the National Institutes of Health to E.W. 1. Mendelsohn, C. L., Wimmer, E. & Racaniello, V. R. (1989) Cell 56, Koike, S., Horie, H., Ise, I., Okitsu, A., Yoshida, M., Iizuka, N., Takeuchi, K., Takegami, T. & Nomoto, A. (1990) EMBO J. 9, Williams, A. F. & Barclay, A. N. (1988) Annu. Rev. Immunol. 6, Proc. Nati. Acad. Sci. USA 88 (1991) 4. Zibert, A., Selinka, H.-C., Elroy-Stein, O., Moss, B. & Wimmer, E. (1990) Virology 182, in press. 5. Kaplan, G., Freistadt, M. S. & Racaniello, V. R. (1990) J. Virol. 64, Lonberg-Holm, K., Crowell, R. L. & Philipson, L. (1976) Nature (London) 259, Colonno, R. J. (1986) BioEssays 5, Rossmann, M. G. & Johnson, J. E. (1989) Annu. Rev. Biochem. 58, Rueckert, R. R. (1985) in Virology, eds. Fields, B. N., Knipe, D. M., Chanock, R. M., Melnick, J. L. & Shope, R. E., (Raven, New York), pp Crowell, R. L. & Landau, B. J. (1983) in Comprehensive Virology, eds. Fraenkel-Conrat, H. & Wagner, R. R. (Plenum, New York), Vol. 18, pp Greve, J. M., Davis, G., Meyer, A. M., Forte, C. P., Yost, S. C., Marlor, C. W., Kamarck, M. E. & McClelland, A. (1989) Cell 56, Staunton, D. E., Merluzzi, V. J., Rothlein, R., Barton, R., Marlin, S. D. & Springer, T. A. (1989) Cell 56, Tomassini, J. E., Graham, D., DeWitt, C. M., Lineberger, D. W., Rodkey, J. A. & Colonno, R. J. (1989) Proc. Nail. Acad. Sci. USA 86, Marsh, M. & Helenius, A. (1989) Adv. Virus Res. 36, Grohmeyer, M. & Wetz, K. (1990) J. Virol. 64, Mendelsohn, C., Johnson, B., Lionetti, K. A., Nobis, P., Wimmer, E. & Racaniello, V. K. (1986) Proc. Nail. Acad. Sci. USA 83, Ren, R., Constantini, F., Gorgacz, E. G., Lee, J. J. & Racaniello, V. R. (1990) Cell 63, Koike, S., Taya, C., Kurata, T., Abe, S., Ise, I., Yonekawa, H. & Nomoto, A. (1991) Proc. Nail. Acad. Sci. USA 88, Kaufman, R. J., Murtha, P. & Davies, M. V. (1987) EMBO J. 6, Paul, A. V., Schultz, A., Pincus, S. E., Oroszlan, S. & Wimmer, E. (1987) Proc. Nail. Acad. Sci. USA 84, Nobis, P., Zibirre, R., Meyer, G., Kuehne, J., Warnecke, G. & Koch, G. (1985) J. Gen. Virol. 66, Emini, E. A., Jameson, B. A. & Wimmer, E. (1983) Nature (London) 304, Jang, S. K., Davies, M., Kaufman, R. J. & Wimmer, E. (1989) J. Virol. 63, Muller, W. J., Mueller, C. R., Mes, A.-M. & Hassel, J. A. (1983) J. Virol. 47, Holland, J. J., McLaren, J. C. & Syverton, J. T. (1959) J. Exp. Med. 110, Hogle, J. M., Chow, M. & Filman, D. J. (1985) Science 229, Rossmann, M. G., Arnold, E., Erickson, J. W., Frankenberger, E. A., Griffith, J. P., Hecht, H.-J., Johnson, J. E., Kamer, G., Luo, M., Mosser, A. G., Rueckert, R. R., Sherry, B. & Vriend, G. (1985) Nature (London) 317, Colonno, R. J., Condra, J. H., Mizutani, S., Callahan, P. L., Davies, M.-E., & Murcko, M. A. (1988) Proc. Natl. Acad. Sci. USA 85, Peaver, D. C., Fancher, M. J., Felock, P. J., Rossmann, M. G., Miller, M. S., Diana, G., Treasurywala, A. M., Mc- Kinlay, M. A. & Dutko, F. J. (1989) J. Virol. 63, Rossmann, M. G. (1989) J. Biol. Chem. 264, Mapoles, J. E., Krah, D. L. & Crowell, R. L. (1985) J. Virol. 55, Acharya, R., Fry, E., Stuart, D., Fox, G., Rowlands, D. & Brown, F. (1989) Nature (London) 337, Fox, G., Parry, N. R., Barnett, P., McGinn, B., Rowlands, D. J. & Brown, F. (1989) J. Gen. Virol. 70, Staunton, D. E., Dustin, M. L., Erickson, H. P. & Springer, T. A. (1990) Cell 61, Giranda, V. L., Chapman, M. S. & Rossmann, M. G. (1990) Proteins 7, Peterson, A. & Seed, B. (1988) Cell 54, Mizukami, T., Fuerst, T. R., Berger, E. A. & Moss, B. (1988) Proc. Natl. Acad. Sci. USA 85, Elroy-Stein, 0. & Moss, B. (1990) Proc. Nail. Acad. Sci. USA 87, Koike, S., Ise, I. & Nomoto, A. (1991) Proc. Natl. Acad. Sci. USA 88, in press.