N glycosylation of the virus binding domain is not essential for function of the human poliovirus receptor.

The human poliovirus receptor (hPVR) is a glycoprotein with three immunoglobulin-like extracellular domains, of which the N-terminal domain (V-type domain) is necessary and sufficient for virus binding and uptake. The effect of N glycosylation of the V domain of hPVR on binding and entry of poliovirus was studied. Stable mouse L-cell lines were generated that express PVR-specific cDNA. One of the cell lines expressed a mutant of hPVR, in which both asparagine residues of the two N-glycosylation sites of the V domain were changed to aspartate (N105D) and serine (N120S), respectively. In the second mutant cell line, the portion of the cDNA encoding the V domain of hPVR was substituted by the homologous sequence of the recently isolated PVR cDNA from monkey cells. This V domain naturally lacks both N glycosylation sites and encodes D105 and S120 at the respective positions of the open reading frame. Absence of N glycosylation at these sites was demonstrated by in vitro translation of the two mutant coding sequences in the presence of microsomal membranes. Both PVR mutant cell lines were capable of poliovirus binding and replication. However, binding of anti-PVR monoclonal antibody D171 and protection from viral replication by this antibody were observed only with the glycosylation mutant carrying the human V domain. In contrast, infection of the cell line expressing the monkey-human hybrid receptor was not blocked even though monkey cells are fully protected by monoclonal antibody D171. The data suggest that N glycosylation of the V domain of hPVR is not essential for viral replication in human tissues and that differential glycosylation of hPVR at these sites is likely not a determinant of viral tissue tropism. Furthermore, the virus binding site and the epitope recognized by monoclonal antibody D171 do not appear to overlap.     

Mutational analysis of the cellular receptor for poliovirus.

To identify sequences of the cellular poliovirus receptor (PVR) involved in viral infection, mutant PVR cDNAs were constructed and assayed for biological activity in mouse L cells. To confirm that mutant PVRs reached the cell surface, an immunological tag, consisting of part of CH3 from human immunoglobulin G1, was engineered into the PVR. Deletion of PVR amino acids 256 to 320 or 385 to the carboxy terminus yielded receptors that were able to support poliovirus infection. PVRs lacking amino acids 40 to 136 or 137 to 256 were expressed at the cell surface but were not active as receptors for poliovirus. The results show that immunoglobulin-type domain 3 and the extreme carboxy terminus of the PVR are not required for viral receptor function, but sequences within the two amino-terminal domains contribute to the initiation of poliovirus infection.     

The poliovirus receptor protein is produced both as membrane-bound and secreted forms.

Both genomic and complementary DNA clones encoding poliovirus receptors were isolated from genomic and complementary DNA libraries prepared from HeLa S3 cells, respectively. Nucleotide sequence analysis of these cloned DNAs revealed that the poliovirus receptor gene is approximately 20 kb long and contains seven introns in the coding region, and that at least four mRNA isoforms referring to the coding sequence are generated by alternative splicing and appear to encode four different molecules, that is, PVR alpha, PVR beta, PVR gamma and PVR delta. The predicted amino acid sequences indicate that PVR alpha and PVR delta, corresponding to the previously described cDNA clones H20A and H20B, respectively, are integral membrane proteins while the other two molecules described here for the first time lack a putative transmembrane domain. Mouse cell transformants carrying PVR alpha were permissive for poliovirus infection, but those carrying PVR beta were hardly permissive. In contrast to PVR alpha, PVR beta was not detected on the surface of the mouse cell transformants but was detected in the culture fluid by an immunological method using a monoclonal antibody against poliovirus receptor. Three types of splicing products for PVR alpha, PVR beta and PVR gamma were detected by polymerase chain reactions using appropriate primers in poly(A)+ RNAs of the brain, leukocyte, liver, lung and placenta of humans; the choice of primers used did not permit detection of PVR delta. In situ hybridization using a cDNA fragment as a probe demonstrated that the PVR gene is located at the band q13.1----13.2 of human chromosome 19.     

Molecular cloning and expression of a murine homolog of the human poliovirus receptor gene.

The poliovirus receptor (Pvr) is a member of the immunoglobulin superfamily of proteins, but its function in the cell is not known. Southern blot hybridization analysis indicated that the murine genome contains a sequence homolog of pvr. As a first step toward using the murine pvr homolog (mph) to study the function of Pvr, murine genomic and cDNA clones encoding mph were isolated. mph encodes a polypeptide with extensive sequence similarity to the extracellular domains of the human PVR. mph mRNAs of 2.0 and 3.0 kb are transcribed in the adult mouse brain, the spinal cord, the spleen, the kidney, the heart, and the liver. The Mph protein does not function as a receptor for poliovirus. However, substitution of domain 1 of the Mph protein with the corresponding sequence from pvr produced a chimeric receptor that could bind poliovirus and lead to productive infection. By constructing pvr-mph chimeras, it will be possible to identify the contact points of poliovirus within domain 1 of Pvr. Identification of the ligand and the cellular function of the Mph protein may help us understand the role of Pvr in the cell.     

Host range phenotype induced by mutations in the internal ribosomal entry site of poliovirus RNA.

Most poliovirus strains infect only primates. The host range (HR) of poliovirus is thought to be primarily determined by a cell surface molecule that functions as poliovirus receptor (PVR), since it has been shown that transgenic mice are made poliovirus sensitive by introducing the human PVR gene into the genome. The relative levels of neurovirulence of polioviruses tested in these transgenic mice were shown to correlate well with the levels tested in monkeys (H. Horie et al., J. Virol. 68:681-688, 1994). Mutants of the virulent Mahoney strain of poliovirus have been generated by disruption of nucleotides 128 to 134, at stem-loop II within the 5' noncoding region, and four of these mutants multiplicated well in human HeLa cells but poorly in mouse TgSVA cells that had been established from the kidney of the poliovirus-sensitive transgenic mouse. Neurovirulence tests using the two animal models revealed that these mutants were strongly attenuated only in tests with the mouse model and were therefore HR mutants. The virus infection cycle in TgSVA cells was restricted by an internal ribosomal entry site (IRES)-dependent initiation process of translation. Viral protein synthesis and the associated block of cellular protein synthesis were not observed in TgSVA cells infected with three of four HR mutants and was evident at only a low level in the remaining mutant. The mutant RNAs were functional in a cell-free protein synthesis system from HeLa cells but not in those from TgSVA and mouse neuroblastoma NS20Y cells. These results suggest that host factor(s) affecting IRES-dependent translation of poliovirus differ between human and mouse cells and that the mutant IRES constructs detect species differences in such host factor(s). The IRES could potentially be a host range determinant for poliovirus infection.     

Human poliovirus receptor gene expression and poliovirus tissue tropism in transgenic mice.

Expression of the human poliovirus receptor (PVR) in transgenic mice results in susceptibility to poliovirus infection. In the primate host, poliovirus infection is characterized by restricted tissue tropism. To determine the pattern of poliovirus tissue tropism in PVR transgenic mice, PVR gene expression and susceptibility to poliovirus infection were examined by in situ hybridization. PVR RNA is expressed in transgenic mice at high levels in neurons of the central and peripheral nervous system, developing T lymphocytes in the thymus, epithelial cells of Bowman's capsule and tubules in the kidney, alveolar cells in the lung, and endocrine cells in the adrenal cortex, and it is expressed at low levels in intestine, spleen, and skeletal muscle. After infection, poliovirus replication was detected only in neurons of the brain and spinal cord and in skeletal muscle. These results demonstrated that poliovirus tissue tropism is not governed solely by expression of the PVR gene nor by accessibility of cells to virus. Although transgenic mouse kidney tissue expressed poliovirus binding sites and was not a site of poliovirus replication, when cultivated in vitro, kidney cells developed susceptibility to infection. Identification of the changes in cultured kidney cells that permit poliovirus infection may provide information on the mechanism of poliovirus tissue tropism.     

Expression of the poliovirus receptor in intestinal epithelial cells is not sufficient to permit poliovirus replication in the mouse gut.

Although the initial site of poliovirus replication in humans is the intestine, previously isolated transgenic mice which carry the human poliovirus receptor (PVR) gene (TgPVR mice), which develop poliomyelitis after intracerebral inoculation, are not susceptible to infection by the oral route. The low levels of PVR expressed in the TgPVR mouse intestine might explain the absence of poliovirus replication at that site. To ascertain whether PVR is the sole determinant of poliovirus susceptibility of the mouse intestine, we have generated transgenic mice by using the promoter for rat intestine fatty acid binding protein to direct PVR expression in mouse gut. Pvr was detected by immunohistochemistry in the enterocytes and M cells of transgenic mouse (TgFABP-PVR) small intestine. Upon oral inoculation with poliovirus, no increase in virus titer was detected in the feces of TgFABP-PVR mice, and no virus replication was observed in the small intestine, although poliovirus replicated in the brain after intracerebral inoculation. The failure of poliovirus to replicate in the TgFABP-PVR mouse small intestine was not due to lack of virus binding sites, because poliovirus could attach to fragments of small intestine from these mice. These results indicate that the inability of poliovirus to replicate in the mouse alimentary tract is not solely due to the absence of virus receptor, and other factors are involved in determining the ability of poliovirus to replicate in the mouse gut.     

Interaction of Poliovirus with Its Receptor Affords a High Level of Infectivity to the Virion in Poliovirus Infections Mediated by the Fc Receptor

Poliovirus infects susceptible cells through the poliovirus receptor (PVR), which functions to bind virus and to change its conformation. These two activities are thought to be necessary for efficient poliovirus infection. How binding and conformation conversion activities contribute to the establishment of poliovirus infection was investigated. Mouse L cells expressing mouse high-affinity Fcγ receptor molecules were established and used to study poliovirus infection mediated by mouse antipoliovirus monoclonal antibodies (MAbs) (immunoglobulin G2a [IgG2a] subtypes) or PVR-IgG2a, a chimeric molecule consisting of the extracellular moiety of PVR and the hinge and Fc portion of mouse IgG2a. The antibodies and PVR-IgG2a showed the same degree of affinity for poliovirus, but the infectivities mediated by these molecules were different. Among the molecules tested, PVR-IgG2a mediated the infection most efficiently, showing 50- to 100-fold-higher efficiency than that attained with the different MAbs. A conformational change of poliovirus was induced only by PVR-IgG2a. These results strongly suggested that some specific interaction(s) between poliovirus and the PVR is required for high-level infectivity of poliovirus in this system.     

Mouse Homolog of Poliovirus Receptor-Related Gene 2 Product, mPRR2, Mediates Homophilic Cell Aggregation

Poliovirus receptor (PVR) is a cell surface glycoprotein that belongs to the immunoglobulin superfamily. Although MPH was initially reported as the mouse homolog of human PVR, recent data strongly suggest that MPH is the mouse homolog of human PRR2, a PVR-related gene 2 product, and not that of human PVR. Thus MPH is renamed mPRR2 in this study. Physiological functions of the PVR-related gene products have not been elucidated, although PVR has been well characterized as the poliovirus receptor. In this study, a possible function of mPRR2 (MPH), which is not a functional receptor for poliovirus, was investigated. Mouse L cells expressing mPRR2 were prepared. Those mouse cells showed a higher activity of cell aggregation than the parental mouse L cells. Enhancement of cell aggregation was also observed for insect Sf9 cells infected with recombinant baculovirus carrying mPRR2 cDNA. On the other hand, L cells expressing human PVR or monkey PVR (AGMα1 or AGMα2) did not show increased cell aggregation. The cell aggregation activity of L cells expressing mPRR2 was inhibited by the addition of anti-mPRR2 monoclonal antibodies or a soluble mPRR2 molecule produced by the baculovirus expression system. An immunofluorescence study revealed that mPRR2 protein was localized to the cell–cell contact sites between cells expressing mPRR2. A similar localization of mPRR2 was observed for intrinsic mPRR2 molecules of the mouse neuroblastoma cell line NS20Y. The contact site-specific localization of mPRR2 was not observed on the border between mPRR2-expressing and nonexpressing HeLa cells. Furthermore, mPRR2 proteins directly bound to each otherin vitro.mPRR2 was detected on various types of cultured cells of mouse origin and in various mouse tissues. These results suggest that mPRR2 is an intercellular adhesion molecule with a homophilic binding manner.     

Neutralization of poliovirus by cell receptors expressed in insect cells.

To examine the interaction of the poliovirus receptor (PVR) with virus and the role of the PVR in virus entry, the PVR was expressed in insect cells. Poliovirus bound to insect cells infected with a recombinant baculovirus (AcPVR) carrying cDNA encoding the PVR. Antibodies raised against PVR expressed in bacteria immunoprecipitated a 67-kilodalton polypeptide from cytoplasmic extracts of AcPVR-infected cells. Treatment of AcPVR-infected cells with tunicamycin revealed that the PVR is a glycoprotein containing N-glycosidic linkages and that carbohydrate accounts for nearly 50% of its molecular weight as estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. When PVR was solubilized from AcPVR-infected insect cells and incubated with poliovirus, viral infectivity was neutralized. Sedimentation analysis revealed that irreversibly altered 135S particles were formed after incubation of poliovirus at 37 degrees C with solubilized extracts of AcPVR-infected insect cells. These results demonstrate that poliovirus eclipse may result from interaction with the cell receptor at neutral pH in the absence of membranes and suggest that soluble receptors may be effective antiviral agents against picornaviruses.     

Interaction of Poliovirus with Its Purified Receptor and Conformational Alteration in the Virion

Polypeptides of amino acids 1 to 241 (PVR241) and 1 to 330 (PVR330) of the human poliovirus receptor (hPVR) were produced in a baculovirus expression system. PVR241 contained extracellular domains 1 and 2 of hPVR, and PVR330 contained extracellular domains 1, 2, and 3. These peptides were purified by immunoaffinity column chromatography with an anti-hPVR monoclonal antibody (MAb). After the purification, PVR241 and PVR330 appeared to retain their native conformation as judged by reactivity with an anti-PVR MAb that recognized domain 1 of hPVR in a conformation-dependent manner. The virulent Mahoney strain of poliovirus type 1 was mixed with the purified PVRs in various concentrations. An average of at least 43 PVR330 molecules were able to bind to one virion particle under the conditions used. The equilibrium dissociation constant between the PVR330 molecule and the PVR binding site (canyon) on the virion was determined to be 4.50 ± (0.86) × 10⁻⁸ M at 4°C. Higher rates of conformational change of the virus (160S) to 135S and 80S particles were observed as the concentration of PVR330 was increased. In this in vitro system, the ratio of the amount of the 135S particle to that of the 80S particle seemed to be always constant. After the disappearance of the 160S particle, the amount of the 80S particle was not increased by further incubation at 37°C. These results suggested that the 80S particle was not derived from the 135S particle under the conditions used in this study.     

Homolog-scanning mutagenesis reveals poliovirus receptor residues important for virus binding and replication.

Poliovirus initiates infection of primate cells by binding to the poliovirus receptor, Pvr. Mouse cells do not bind poliovirus but express a Pvr homolog, Mph, that does not function as a poliovirus receptor. Previous work has shown that the first immunoglobulin-like domain of the Pvr protein contains the virus binding site. To further identify sequences of Pvr important for its interaction with poliovirus, stable cell lines expressing mutated Pvr molecules were examined for their abilities to bind virus and support virus replication. Substitution of the amino-terminal domain of Mph with that of Pvr yields a molecule that can function as a poliovirus receptor. Cells expressing this chimeric receptor have normal binding affinity for poliovirus, yet the kinetics of virus replication are delayed. Results of virus alteration assays indicate that this chimeric receptor is defective in converting native virus to 135S altered particles. This defect is not observed with cells expressing receptor recombinants that include Pvr domains 1 and 2. Because altered particles are believed to be an intermediate in poliovirus entry, these findings suggest that Pvr domains 2 and 3 participate in early stages of infection. Additional mutants were made by substituting variant Mph residues for the corresponding residues in Pvr. The results were interpreted by using a model of Pvr predicted from the known structures of other immunoglobulin-like V-type domains. Analysis of stable cell lines expressing the mutant proteins revealed that virus binding is influenced by mutations in the predicted C'-C" loop, the C" beta-strand, the C"-D loop, and the D-E loop. Mutations in homologous regions of the immunoglobulin-like CD4 molecule alter its interaction with gp120 of human immunodeficiency virus type 1. Cells expressing Pvr mutations on the predicted C" edge do not develop cytopathic effect during poliovirus infection, suggesting that poliovirus-induced cytopathic effect may be induced by the virus-receptor interaction.     

Construction of viable deletion and insertion mutants of the Sabin strain of type 1 poliovirus: function of the 5'' noncoding sequence in viral replication.

A number of deletion and insertion sequences were introduced into the 5' noncoding sequence (742 nucleotides long) of the genome of the Sabin strain of type 1 poliovirus by using an infectious cDNA clone of the virus strain. The genomes of all three poliovirus serotypes contained highly homologous sequences (nucleotide positions 509 to 639) as well as highly variable sequences (positions 640 to 742) in the 5' noncoding region. The viability of mutant viruses was tested by transfecting mutant cDNA clones into African green monkey kidney cells and then estimating the plaque sizes displayed on the cells. The results suggested that the highly variable sequence next to the VP4 coding region did not play an important role, at least in the in vitro culture system used, that the loci of highly conserved nucleotide sequences were not always expected to be the genome regions essential for viral replication, that the sequence between positions 564 and 599 carried genetic information to maintain the efficiency of certain steps in viral replication, and that the sequence between positions 551 to 563 might play an essential role in viral replication. Four-base deletion or insertion mutations were introduced into relatively variable sequences in the genome region upstream of position 509. The results suggest that variable sequences do not always indicate that the corresponding genome regions are less important. Apparent revertants (large-plaque variants) were easily generated from one of the viable mutants with the small-plaque phenotype. The determination of nucleotide sequences of the revertant genomes revealed the second mutation site. The results suggested that the different loci at around positions 200 and 500 might specifically interact with each other. This interaction may result in the formation of a functional structure that influences the efficiency of certain steps in the viral replication.     

A recombinant virus between the Sabin 1 and Sabin 3 vaccine strains of poliovirus as a possible candidate for a new type 3 poliovirus live vaccine strain.

Biological tests including the monkey neurovirulence test performed on recombinants between the virulent Mahoney and attenuated Sabin 1 strains of type 1 poliovirus indicated that the genome region encoding mainly the viral capsid proteins had little correlation with the neurovirulence or attenuation phenotype of the virus. The results suggested that new vaccine strains of type 2 and type 3 polioviruses may be constructed in vitro by replacing the sequence encoding the antigenic determinants in viral capsid proteins of the Sabin 1 genome by the corresponding sequences of the type 2 and type 3 genome, respectively. Accordingly, we constructed recombinants between the Sabin 1 and Sabin 3 strains of poliovirus in which genome sequences of the Sabin 1 strain encoding most or all capsid proteins were replaced by the corresponding genome sequences of the Sabin 3 strain. One of the recombinant viruses thus constructed was fully viable and showed antigenicity and immunogenicity identical to those of type 3 poliovirus. The monkey neurovirulence tests and in vitro phenotypic marker tests (temperature sensitivity of growth, sodium bicarbonate concentration dependency of growth under agar overlay, and size of plaque) were performed on the recombinant virus. The stability of the virus in regard to the temperature sensitivity phenotype was also tested. The results suggested that the recombinant virus is a possible candidate for a new type 3 poliovirus vaccine strain.     

Construction of recombinant DNA molecules by the use of a single stranded DNA generated by the polymerase chain reaction: its application to chimeric hepatitis A virus/poliovirus subgenomic cDNA.

In order to study the importance of VP4 in picornavirus replication and translation, we replaced the hepatitis A virus (HAV) VP4 with the poliovirus (PV1) VP4. Using a modification of oligonucleotide site directed mutagenesis and the polymerase chain reaction (PCR), we created a subgenomic cDNA chimera of hepatitis A virus in which the precise sequences coding for HAV VP4 capsid protein were replaced by the sequences coding for the poliovirus VP4 capsid protein. The method involved the use of PCR primers corresponding to the 3' and 5' ends of the poliovirus VP4 sequence and that had HAV VP4 3' and 5' flanking sequences on their 5'ends. Single stranded DNA of 240 and 242 nt containing the 204 nt coding for the complete poliovirus VP4 were produced by using a limiting amount of one of the primers in a PCR reaction. These single stranded PCR products were used like mutagenic oligonucleotides on a single stranded phagemid containing the first 2070 bases of the HAV genome. Using this technique, we precisely replaced the HAV VP4 gene by the poliovirus VP4 gene as determined by DNA sequencing. The cDNA was transcribed into RNA and translated in vitro. The resulting protein could be precipitated by antibody to poliovirus VP4 but not to HAV VP4.     

Genetic variation of the poliovirus genome with two VPg coding units.

Amongst the picornaviruses, poliovirus encodes a single copy of the genome-linked protein, VPg wheras foot-and-mouth disease virus uniquely encodes three copies of VPg. We have previously shown that a genetically engineered poliovirus genome containing two tandemly arranged VPgs is quasi-infectious (qi) that, upon genome replication, inadvertently deleted one complete VPg sequence. Using two genetically marked viral genomes with two VPg sequences, we now provide evidence that this deletion occurs via homologous recombination. The mechanism was abrogated when the second VPg was engineered such that its nucleotide sequence differed from that of the first VPg sequence by 36%. Such genomes also expressed a qi phenotype, but progeny viruses resulted from (i) random deletions yielding single VPg coding sequences of varying length lacking the Q*G cleavage site between the VPgs and (ii) mutations in the AKVQ*G cleavage sites between the VPgs at either the P4, P1 or P1' position. These variants present a unique genetic system defining the cleavage signals recognized in 3Cpro-catalyzed proteolysis. We propose a recognition event in the cis cleavages of the polyprotein P2-P3 region, and we present a hypothesis why the poliovirus genome does not tolerate two tandemly arranged VPg sequences.     

Dendritic cells and macrophages are productively infected by poliovirus

Expression of the poliovirus receptor (PVR) on cells is a major host determinant of infection by poliovirus. Previously, the only immune cell type known to express PVR was the blood-derived monocyte, which is susceptible to infection at very low frequency. We demonstrate that professional antigen-presenting cells-macrophages and dendritic cells, generated upon differentiation of monocytes-retain expression of PVR and are highly susceptible to infection by type 1 Mahoney strain of poliovirus. Maximal cell-associated titers of virus are obtained within 6 to 8 h postinfection, and cell death and lysis occurs within 24 h postinfection. Similar kinetics are observed in cells infected with the Sabin 1 vaccine strain. Although protein synthesis and receptor-mediated endocytosis are inhibited upon poliovirus infection of these critical antigen-presenting cells, we demonstrate for the first time that functional presentation of antigen occurs in these infected cells via the HLA class II pathway.     

Transgenic mice carrying the human poliovirus receptor: new animal models for study of poliovirus neurovirulence.

Recombinant viruses between the virulent Mahoney and attenuated Sabin 1 strains of poliovirus type 1 were subjected to neurovirulence tests using a transgenic (Tg) mouse line, ICR-PVRTg1, that carried the human poliovirus receptor gene. The Tg mice were inoculated intracerebrally with these recombinant viruses and observed for clinical signs, histopathological lesions, and viral antigens as parameters of neurovirulence of the viruses. These parameters observed in the Tg mice were different for different inoculated viruses. Dose-dependent incidences of paralysis and of death were observed in the Tg mice inoculated with any viruses used. This indicates that values of 50% lethal dose are useful to score a wide range of neurovirulence of poliovirus. The neurovirulence of individual viruses estimated by the Tg mouse model had a strong correlation with those estimated by monkey model. Consequently, the mouse tests identified the neurovirulence determinants on the genome of poliovirus that had been identified by monkey tests. In addition, the mouse tests revealed new neurovirulence determinants, that is, different nucleotides between the two strains at positions 189 and 21 and/or 935 in the 5'-proximal 1,122 nucleotides. The Tg mice used in this study may be suitable for replacing monkeys for investigating poliovirus neurovirulence.