Molecular Biology of Picornavirus

A breakthrough in the comprehension of poliomyelitis, viz., clinical disease and pathology, was possible by progress achieved in neuropathology and neuropathologic techniques. On the basis of the autopsy of two cases of poliomyelitis acuta, review of the literature, and the unknown etiologic agent, they caution against jumping to conclusions about the causal sequence of events. In the late 1930s, incontrovertible evidence was brought forth for the presence of virus in the stools of patients with paralytic or nonparalytic poliomyelitis and in contacts as well. Some years earlier researchers tried with ingenious techniques to multiply poliovirus in mice in support of Brodie’vaccine trials, but with the vaccine failures this work was neglected. The finally well-established viremia, however, could be interpreted differently from that of Bodian; e.g., Sabin considered viremia as means to amplify virus in various extraneural tissues from where poliovirus subsequently reached the central nervous system via neural, axonal spread. Last but not least, the seminal paper of researchers was the starting point of modern poliovirology, and it launched the scientific revolution rightly called molecular virology.

Picornaviruses have traditionally been defined in terms of serotypes, grouped into genera. Recently, a radical change has been introduced with the advent of the concept of a picornavirus species, generally consisting of several serotypes. This classification has evolved in response to developments in one’s understanding of the biological and genetic properties of picornaviruses, which has accelerated greatly over the past few years. This chapter examines some of the properties that can be used to group, or differentiate between, picornaviruses and some of the complications that arise from attempting to classify viruses, which are potentially highly plastic in terms of sequence and even genome organization. Enteroviruses were originally classified as poliovirus (PV), coxsackievirus A (CVA), coxsackievirus B (CVB), and echovirus on the basis of their pathogenicity in experimental animals. The current number of established picornavirus serotypes is very high, especially among the two closely related genera, enteroviruses and rhinoviruses. The three types of PV are classical examples of easily distinguishable enterovirus serotypes, whereas definite cross-reactivity is readily demonstrable between, for example, several established CVA serotypes. As the sequences of numerous picornaviruses have become available, they have shown that there is essentially one common genome organization. To provide biologically important insights, taxonomy should ultimately be based on genetic relationships, which in turn reflect the evolutionary history of the viruses.

In many cases, structures have been determined, often using the lower resolution cryo-electron microscopy (cryo-EM) technique, of picornaviruses in complex with their cellular receptors, neutralizing antibodies, antiviral compounds, or other, biologically significant ligands. Picornavirus capsids are assembled from 60 protomers, each composed of four structural proteins, viral protein 1 (VP1), VP2, VP3, and VP4. The first three of these proteins have molecular weights of around 30 kDa and form the external surface of the icosahedral shell. Conservation of three-dimensional structure is almost invariably greater than conservation of amino acid homology. Thus, structural comparisons can be used to trace divergent evolution over longer time spans than is possible by amino acid sequence comparisons. Assembly of picornaviruses proceeds from 6S protomers of VP1, VP3, and VP0, via 14S pentamers of five 6S protomers, to mature virions. The final step involves inclusion of the RNA into empty capsids or partially assembled shells with simultaneous cleavage of VP0 into VP2 and VP4. The mutant viruses that were able to grow were mostly single mutations and could be sorted into groups that were neutralized by the same set of antibodies. A variety of additional evidence all points to the ability of simple icosahedral viruses to be in constant flux or “breathing”. This unexpected and structurally difficult-to-understand phenomenon accounts for the virus being able to externalize the internal VP4 and amino-terminal region of VP1 in the initial stages of cell entry.

This chapter discusses the possible mechanisms of antibody-mediated neutralization of human rhinovirus to better understand how antibodies recognize their targets and neutralize viral infectivity. Understanding these fundamental processes is crucial for future vaccine development and new antibody therapeutics. This is especially true for viruses like the human immunodeficiency virus (HIV), where the more traditional approach of using attenuated viral strains appears to be risky and insufficiently efficacious. The rhinoviruses, of which there are more than 100 serotypes, are major causative agents of the common cold in humans. The major difference in human rhinovirus 14 (HRV14) preparation between two crystals is the polyethylene glycol (PEG) 400 that was added as a cryoprotectant. This strongly suggests that the pocket factor found in the HRV14-Fab complex came from PEG 400. Since there has been no direct evidence that pocket factors are derived from the host cell, these results further suggest that pocket factors found in the other viruses might also be compounds used in purification or crystallization. For poliovirus and rhinovirus, interactions with their receptors appear to be essential for the proper release of the genomic RNA into the cytoplasm of the host cell. When antibody-poliovirus complexes enter cells, the viral RNA is quickly digested. Vaccine design strategies might, therefore, benefit by focusing on the production of high-affinity antibodies rather than on a particular in vitro neutralization property.

Antigenic variation in foot-and-mouth disease (FMD) is important for both practical and fundamental reasons. In the first place, it has considerable importance in the control of the disease by vaccination because vaccines providing protection against one of the seven serotypes afford no protection against viruses belonging to the other six serotypes. Equally important, the wide spectrum of antigenic variation within the serotypes provides similar control problems. The second reason is the opportunity that antigenic variation provides for studying the immunochemistry of the virus. It is of particular interest that a major antigenic feature of the virus is coincident with the cell receptor-binding motif. This provides the intriguing challenge of trying to understand how the apparently conflicting requirements of maintaining receptor-binding specificity while allowing antigenic variation to occur in the same structural feature are resolved. The discovery of serotypes stemmed from the observation that animals in the field could become infected on more than one occasion. Cross-resistance analyses involving panels of monoclonal antibodies and panels of resistant viruses are used to map the mutations into nonoverlapping sites. Acquisition of the ability to bind to heparin sulfate at cell surfaces has been seen in foot-and-mouth disease virus (FMDV), as it has in other tissue culture-adapted viruses.

The most surprising result of studies focused on early events in animal virus infection is the diversity of cell surface proteins serving as receptors. Two viruses belonging to two entirely different families, say, the small RNA coxsackievirus B and the large DNA adenovirus 2, have chosen the same small cell surface protein as receptor (CAR). Perhaps Retroviridae assembled the most diverse menu of cellular receptors. Most receptors belong to the immunoglobulin (Ig) superfamily or the integrin receptor family. Considering the many unresolved puzzles concerning enterovirus entry, and the steady expansion of picornavirus genera, the number of picornavirus receptors can increase significantly in the future. The cellular function of integrins includes binding extracellular matrix proteins, cell-cell interactions, and signal transduction. Decay accelerating factor (DAF) (CD55) is a member of the regulator of the complement activity protein family and protects the cell from autologous lysis. Heparan sulfate has been implicated in receptor function for certain strains of foot-and-mouth disease virus (FMDV) and clinical isolates of echovirus 6. The difference in receptor utilization not only yields host range phenotypes in vitro, but it has also a profound effect on pathogenesis in animals. The polypeptide chain of decay-accelerating factor (DAF) consists of four short consensus sequences (SCRs), some of which are involved in virus binding. Pathogenesis is determined by tissue tropism, spread of the virus to target tissues, and virulence.

Poliovirus is an ideal model for understanding how non-enveloped viruses enter cells and initiate infection. The study of entry of a wide variety of viruses reveals common themes that are the consequence of a central problem faced by all viruses in the passage from cell to cell or from host to host. The final stage of assembly for many viruses involves proteolytic processing of a virion protein. The replication cycle is initiated when poliovirus encounters the poliovirus receptor (Pvr), a transmembrane glycoprotein with three extracellular immunoglobulin (Ig)- like domains. Although there is still considerable controversy concerning the role of the two particles, the A particle may be an intermediate in the cell entry pathway, and the 80S empty particle may be the final protein product that accumulates after the RNA is released into the cytoplasm to initiate translation and replication. An attempt has been made to obtain structural “snapshots” of stable intermediates in the poliovirus cell entry pathway and to couple the structural information with the results of genetic, biophysical, and biochemical observations to fill in the gaps in the pathway. On the basis of structural, genetic, and biochemical evidence available to date, a working model for the cell entry of poliovirus, related enteroviruses, and major group rhinoviruses is proposed. An alternative model is proposed in which the transition from the initial binding complex to the tight-binding complex is characterized by movements of VP1, VP2, and VP3 that mimic the umbrella-like movements of the virion to A particle transition.

The first step in the virus life cycle is the interaction between the virus and the cellular receptor. This interaction is a significant determinant of pathogenesis. For picornaviruses, the best studied virus-receptor interaction has been that of the major group rhinoviruses and their cellular receptor, intercellular adhesion molecule-1 (ICAM-1, CD54). Later, following the structure determination of rhinovirus, the receptor for the major group of rhinoviruses was identified as ICAM-1. It should be informative to examine the features of a receptor and a virus that use this canyon strategy for attachment and entry. This chapter looks at a well-studied example of this interaction at a structural level, the interaction of the major group rhinoviruses with their cellular receptor, ICAM-1. Molecular genetic and structural studies have demonstrated that rhinoviruses bind to domain D1 of ICAM-1. A major tenet of the canyon hypothesis suggested that the receptor would bind in the crevice that surrounds each of the fivefold vertices. This was supported by molecular genetic studies in which residues that line the floor of the canyon in rhinovirus 14 (HRV14) were mutated and the resulting virus particles had altered levels of receptor binding. The electron density values for D1D2 were nearly identical to those of the virus, indicating a nearly complete saturation of all of the 60 available binding sites on the virion. The association of ICAM-1 with HRV3 was slow compared with other protein-protein interactions, suggesting that the receptor may have a difficult time penetrating the canyon and making the correct contacts.

More than 25 years ago Lonberg-Holm and colleagues demonstrated that out of 11 human rhinovirus serotypes investigated, HRV3, -5, -10, -14, -15, -39, -41, and -51 competed for the same binding site on HeLa cells whereas HRV1A, -1B, and -2 were found to recognize other sites on the cell surface. The ubiquitous presence of minor group HRV receptors was taken to indicate that it was evolutionarily strongly conserved; this was then confirmed by the discovery of its identity with the low-density lipoprotein receptor (LDLR) and the existence of several closely related molecules with virus-binding activity. A structural hallmark of members of the LDLR family is various numbers of incomplete direct repeats of about 40 amino acids containing six cysteines each, which are all involved in disulfide bridges. Purification of amounts of LDLR or of lipoprotein receptor-related protein (LRP) large enough to carry out cell-protection experiments appeared tedious, and expression of these proteins in bacteria was expected to be difficult due to the large number of disulfide bridges present in the native proteins. To determine the minimal structure requirements of LDLR for viral recognition and the LDLR-binding site on the viral capsid, the authors started to express soluble truncated LDLR in insect Sf9 cells. Receptor derivatives might inhibit virus infection by various mechanisms. In most cases soluble receptors compete for the membrane receptors present on the cell surface.

Studies in the late 1950s demonstrated that homogenates of particular tissues could adsorb picornaviruses, including some echoviruses and coxsackieviruses, and correlated virus adsorption with susceptibility to infection. The understanding of the receptors for group B coxsackieviruses (CVBs) is largely based on work carried out in the beginning of 1960s, and culminating in the identification of two receptor molecules within the past 5 years. Attachment-interference studies, in which saturation of cellular receptors by one virus was found to prevent attachment of a related virus, identified several picornavirus receptor families, whose members were likely to share receptors. Decay-accelerating factor (DAF) is expressed on many cell types and functions to protect cells from lysis by autologous complement. DAF is a member of a family of complement regulatory proteins composed of homologous short consensus repeat (SCR) domains. Consistent with the original observation that all six CVB serotypes compete for a single receptor, Coxsackievirus and adenovirus receptor (CAR) has been shown to mediate infection by laboratory and clinical isolates belonging to all six serotypes, including viruses like CVB3-rhabdomyosarcoma (RD) that also interact with DAF. The observation that CAR-transfected rodent cells become infected, while DAF-transfected CHO cells do not, suggests that DAF cannot perform some postattachment function essential for virus infection. Recent experiments confirm that ICAM-1 is in fact a receptor for group A coxsackieviruses (CVA)21. Infection by a variety of echovirus serotypes, as well as by CVA9, was reportedly inhibited by a monoclonal antibody to a 44-kDa cell surface protein.

This chapter examines the early events that occur upon infection of cultured cells with foot-and-mouth disease virus (FMDV) and defines the known virus-receptor interactions. In addition, the authors try to relate what is known about these early interactions to disease pathogenesis. The first identification of the integrin receptor for FMDV was made by comparing its receptor specificity with that of the human enterovirus, coxsackievirus A9 (CAV9), which contains a 17-amino-acid C-terminal insertion in VPl containing an arginineglycine- aspartic acid (RGD) sequence. Other important functional domains of integrins include the cytoplasmic domains of the α and β subunits. The authors examined the role of the cytoplasmic domains of the bovine integrin αv β3 in FMDV infection of cultured cells. While they have learned much about the early interactions of FMDV with its receptors in vitro, the role these receptors play in the pathogenesis of the disease is still unclear. Studies on the pathogenesis of FMD have shown that initial sites of viral replication are the lung and pharyngeal areas followed by rapid dissemination of the virus to the oral and pedal epithelia. Application of knowledge of the detailed mechanisms of FMDV-receptor interactions in vitro to the disease in the whole animal should provide insights into viral pathogenesis and may provide new information on how to control this important disease. Thus, future research should concentrate on determining which of the RGD-binding integrins found in susceptible hosts are capable of serving as receptors for FMDV.

This chapter provides an overview of the organization of the picornavirus genome. It highlights that it is reasonable to consider the structure and functions of the picornaviral genome as an integrated system designed to ensure efficient self-reproduction. To achieve this goal, the genetic system should store information about how to synthesize the new generation of viral RNA and protein molecules endowed with the ability to assemble into virions capable of infecting new host cells. The capacity of poliovirus RNA to code for protein has been documented by its translation in a cell-free system. Demonstration of infectivity of the poliovirus cDNA heralded the advent of a new era in picornavirus genome research. Translation initiation of picornaviral RNAs is accomplished through a cap-independent mechanism involving an RNA segment hundreds of nucleotides long called the internal ribosome entry site (IRES) located within the 5' UTR. RNA folding is intrinsically dynamic and is influenced by many variables such as local ionic conditions and, in particular, RNA-binding proteins. Picornavirus infection is accompanied by changes in the permeability of both intracellular and plasma membranes. Appropriate specific alterations occur early and late in the reproduction cycle. An important function of the viral genome is to prevent or neutralize the host defensive measures. The genes encoding ancestors of picornavirus proteins were likely existent long before the appearance of the first picornavirus.

Genome connoisseurs are always looking for new methods to tweak the data, especially in sequence regions where good fits are harder to achieve or recognize. The previous capsid alignments were based on superimposition of virion crystal structure hydrogen-bonding maps, then extended by multiple, reiterative pairwise comparisons to include similar related sequences. With the HMMER program suite full-length picornavirus genome hidden Markov model (HMM)-profiles (lengths of 8,000 to 9,000 bases) were calculated from the previous alignments. The alignments were checked for consistency, and then refined mathematically and heuristically to (i) maximize the number of matched bases and encoded amino acids, (ii) minimize the location and frequency of indels (insertions/deletions), and (iii) emphasize the conservation of homologous features such as catalytic sites and proteolytic cleavage sites. The current iterations now include genome-length alignments for the seven most populous picornaviral genera (Enterovirus, Rhinovirus, Cardiovirus, Aphthovirus, Hepatovirus, Parechovirus, and Teschovirus) and extend over 173 different strains, providing formats for about 1,000,000 bases. The correlate polyprotein alignments, derived by translation of the aligned RNAs, include about 291,000 amino acids. For eukaryotes, there is very little information about the specific pressures that might shape a given lineage. Certainly, the symmetrical methylation of CG dinucleotides has a unique structural significance in many higher DNA genomes, reducing the value and frequency of this sequence in coding regions.

By the late 1970s, a skeleton of the translation mechanism utilized by the bulk of cellular mRNAs had been elucidated. More elegant proof of utilization of an internal ribosome entry site (IRES) on the viral RNA was provided by analysis of translation of bicistronic constructs engineered to encode two tandem protein sequences, separated by a viral 5' untranslated region (UTR). The RNA sequences that constitute an IRES extend through several hundred nucleotides and fold into complex, multidomain structures. To identify the nucleotide sequences and structures in the picornavirus 5' UTRs that contribute to IRES function, mutations were introduced into cDNAs to generate transcripts whose translation could be evaluated in vitro or in transfected cells. For the picornavirus RNAs, a relatively relaxed structure is predicted, with a long, central axis of shifting base pairs, with relatively stable stems, loops, helices, and branch points extending from the central backbone. The conserved motifs in the secondary structure elements that are essential for IRES activity are likely to facilitate RNA-RNA or RNA-protein interactions required to maintain a higher order structure needed for proper recognition of the IRES element by the translational machinery. The picornavirus IRES elements represent relatively large complex domains composed of multiple subdomains that require correct spatial orientation and interactions to carry out IRES function. Efficient translation of capped mRNAs in eukaryotic cells involves a synergistic dependence on the 5'-terminal cap structure and the 3'-terminal poly(A) tail.

In December 1975, Hugh Pelham invented the micrococcal nuclease-treated (messenger-dependent) rabbit reticulocyte lysate. The sequencing of picornavirus RNAs in the early 1980s revealed that they had rather long 5' untranslated regions (5' UTRs) of between 610 and ~1400 nucleotides [nt], depending on the particular virus species. In due course, evidence for translation by direct internal ribosome entry was provided by the demonstration that the insertion of a picornavirus 5' UTR between the two cistrons of a laboratory-constructed dicistronic mRNA leads to dramatic enhancement of expression of the downstream cistron. Entero- and rhinovirus 2A and foot-and-mouth disease virus (FMDV) L proteases cleave eIF4G into an N-terminal one-third fragment, which has the eIF4E interaction site, and a C-terminal two-thirds fragment, which has the interaction site for eIF3, and both sites where eIF4A binds. Toe-printing and sucrose gradient analyses of initiation complexes formed with highly purified initiation factors have shown that the binding of the 40S subunit to the correct initiation site on the encephalomyocarditis virus (EMCV) internal ribosomal entry site (IRES) absolutely requires eIF2, 3, and 4A and either the complete native eIF4F complex (with associated eIF4A) or recombinant fragments of eIF4G, which include the central one-third domain. In conclusion, therefore, the most plausible hypothesis is that by binding at multiple points in the IRES element, polypyrimidine tract-binding protein (PTB), PCBP-2, and unr serve to help in the maintenance or the attainment of the appropriate three-dimensional RNA structure.

Picornaviruses employ a number of unique intracellular mechanisms and novel processes during their infectious cycles resulting in their being among the most successful of viral pathogens. This chapter begins with a discussion of the features of viral proteinases, continues with an outline of the functions of both precursor and mature viral polypeptides present during a picornaviral infection, and concludes with a brief summary of nonviral substrates cleaved by viral proteinases. Viral proteinases including L protein, 2A proteinase and 3C proteinase have been discussed in the chapter. The aphthoviruses and cardioviruses code for an L protein at the N terminus of their polyproteins. The cleavage activity of the L proteinase from foot-and-mouth disease virus (FMDV), an aphthovirus, has been well characterized. The 3C proteinase activity carries out the majority of the proteolytic processing of the viral polyprotein. The evolution of picornaviruses might dictate that the P1 to PN substrate positions be identical or similar to optimize polyprotein processing and maximize the generation of mature viral proteins. In vitro synthesized viral RNAs containing large inframe deletions within the P1 region are self-replicating in cultured cells, suggesting that the proteins required for viral RNA replication are located primarily within the P2 and P3 (nonstructural) regions of the genome. Since picornaviruses utilize a mechanism of translation that is cap independent, it is advantageous to the virus to inhibit nonessential cap-dependent cellular translation.

This chapter summarizes the implications of three-dimensional structures on the proteolytic mechanisms, the substrate specificities, and the functions of the proteinases in infected cells. The picornaviral polyproteins can be divided into three regions, designated P1, P2, and P3. These correspond to the N-terminal capsid protein precursor (P1, containing the four capsid proteins 1A-1D), the middle of the polyprotein containing three of the nonstructural proteins (P2, the three proteins 2A-2C), and the most C-terminal segment of the polyprotein containing four nonstructural proteins (P3, proteins 3A-3D). In the cardio- and aphthoviruses, a protein known as the leader protein precedes P1. The hepato- and parechoviruses encode only a single proteolytic enzyme and are therefore proteolytically the simplest of the picornaviruses. During the replication of a picornavirus, the physiology and ultrastructure of the infected cells are drastically modified. Thus, cellular RNA and protein synthesis as well as protein trafficking are inhibited. The chapter discusses crystal structures in terms of their mechanisms of action and specificities and how the proteinases have evolved to be able to carry out their specific roles in the replication of the respective viruses. Like Streptomyces griseus protease B (SGPB) and the 3C proteinases, HRV2 2Apro comprises two subdomains, built up by β-strands as found in chymotrypsin.

In the case of the aphtho- and cardioviruses, the primary cleavage in the region of the polyprotein was known to be different, occurring at the C terminus of 2A. Precursor forms spanning the 2A/2B junction are not observed in aphtho- or cardiovirus polyprotein processing. Deletions downstream of 2A did not appear to affect cleavage. Experiments analyzing the endogenous processing properties of recombinant aphthovirus indicated that the cleavage activity could be a property of the 2A oligopeptidic region alone. Consistent with this notion, studies on the endogenous processing properties of domains of the cardiovirus Theiler’s murine encephalomyelitis virus (TMEV) polyprotein localized the 2A/2B cleavage activity within the 2AB region. With artificial reporter polyprotein systems the 2A/2B cleavage activity of both EMCV and TMEV was subsequently mapped to the C-terminal 18 aa of their 2A proteins—these cardiovirus sequences being as efficient as the FMDV 2A in mediating cleavage. The molar excess of the translation product N terminal of 2A over that C-terminal of 2A is a product of inserting the 2A sequence into our artificial polyprotein systems. In summary, the authors and others have shown the aphtho- and cardiovirus 2A/2B cleavage is mediated by an oligopeptidic region, representing either the whole (aphthoviruses) or part (cardioviruses) of the 2A region.

This chapter summarizes all the pertinent experimental evidence that is currently available and proposes a unified model for picornavirus RNA replication. These data are derived from three types of experiments. In the simplest type, purified enzymes are used to study biochemical reactions in vitro. The second in complexity are those studies that use crude replication complexes isolated either from infected cells or from coupled translation/replication reactions of viral RNA. Finally, the most difficult method involves studying reactions in the infected cell itself. The proteins of the P3 domain are those that are most directly involved in the process of RNA synthesis. During translation of poliovirus RNA the P3 precursor is generated from the polyprotein by a fast cleavage event at the amino terminus of the 3A-coding region. The RNA polymerases of poliovirus and human rhinoviruses (HRV)2 are dependent in vitro on an RNA template and on a primer, either RNA, DNA, or VPg. Properties and functions of proteins encoded by the P3 domain of the poliovirus polyprotein are discussed. Mutational analysis of the heteropolymeric sequences in the 3 ' nontranslated region (NTR) of entero- and rhinoviruses indicated that this region is important for RNA replication. Prior to the initiation of minus-strand RNA synthesis, the RNA polymerase has to recognize its own viral RNA in a pool of cellular mRNAs and then select it as the only template for transcription.

During their replication cycles many viruses extensively affect host cell morphology. Pathologists diagnosed viral diseases on the basis of morphological characteristics of the diseased tissue long before the nature and biology of vimses were known. The take-off of modern experimental virology was the observation by Enders that in cultured cells poliovims (PV) induces morphological alterations, termed cytopathic effect (CPE). Subsequently, CPE could be used as an easy marker for virus replication in cell cultures. Picornaviruses, with the possible exception of hepatitis A virus (HAV), induce cell alterations that culminate ultimately in cell death. Picornavirus-induced cell alterations and cell destruction are directly coupled to viral replication. This was also demonstrated in coxsackievirus-infected muscles of mice where viral RNA replication located to the foci of gradual destruction of the contractile material. Expression of individual viral proteins in HeLa cells also pointed to the viral protein 2BC as the protein responsible for triggering the vesicle formation process, possibly assisted by protein 3A. Recent experiments visualized the vesicle budding process at the ER and demonstrated that the formation of vesicles is part of the anterograde transport. For PV RNA replication, translation of an individual RNA molecule has to be down-regulated to allow for transcription. The proposed mechanism for this switch consists of an enhanced 3CD-mediated binding of the cellular poly(rC)-binding protein to the 5'-cloverleaf structure of plus-strand viral RNA.

Replication of the poliovirus genome has been studied for many decades by using a variety of molecular, genetic, biochemical, and structural approaches. These studies have uncovered most, if not all, of the virus encoded proteins and RNA sequences/structures required for genome replication. Proteins encoded by the P3 region of the genome, however, are thought to participate more directly in the genome replication process. The fourth and final protein domain of the P3 region of the viral polyprotein is the RNA-dependent RNA polymerase (RdRP) 3Dpol, the core component of the replication machinery. One of the most important contributions to one’s understanding of 3Dpol function was the solution of the crystal structure of 3Dpol by researchers in 1997. This structure provided the first glimpse into the architecture of an RdRP. The preceding discussion highlights the similarity of 3Dpol with other classes of nucleic acid polymerase. While features unique to 3Dpol may exist, for example, the so-called fingertips, this subdomain likely exists in all RdRPs based on the two RdRP structures available to date. In contrast, two potential interaction/oligomerization domains were also observed in the crystal structure of 3Dpol. These interaction surfaces have no structural homologues in any other polymerases for which structural information is available, including the RdRP from hepatitis C virus (HCV). It is clear that the availability of structural information for 3Dpol has shed light on one’s understanding of 3Dpol function.

This chapter overviews the mechanisms underlying the stability and especially the variability of the picornavirus phenotype and genotype. It also talks about the principles of genetic analysis of viral functions and of manipulations with the viral genome. The pathogenic potential of picornaviruses may vary in two respects: (i) viral variants may be more or less virulent, in the sense that different doses of a virus are required to inflict similar clinical signs (at the extreme, even the highest possible dose may fail to produce any signs); and (ii) the variants may cause different clinical patterns. The picornavirus RNA polymerase, like analogous enzymes of other RNA viruses, lacks the ability to remove the erroneously incorporated nucleotide (the proofreading activity), characteristic of many DNA polymerases. Viral RNA rearrangements of any kind (recombination, deletions, insertions) may contribute to genome plasticity and evolution. There are two general approaches to linking specific phenotypic traits to specific genes or their parts: (i) identification of mutations associated with a distinct phenotype, and (ii) observation of phenotypic effects of engineered mutations in specific genes. The current understanding of the mechanisms of picornavirus genome variability and stability is rather superficial and limited.

This chapter examines key features of the three main stages in virus evolution with examples drawn from several virus groups. It then reviews specific results with picornaviruses, as well as implications of high mutation rates and quasispecies dynamics for this important and diverse group of pathogens. With the levels of heterogeneity and viral load often seen in infected individuals, RNA virus populations include potentially all possible single mutants and decreasing amounts of multiple mutants. There are probably many routes to drug resistance, including amino acid replacements in the wall of the pocket preventing accommodation of the drug into the pocket (termed "exclusion" mutants), and replacements elsewhere in the capsid that affect viral uncoating. Resistant mutants display decreased affinity for the drug or increased affinity for the receptor, and often show low fitness values relative to their parental counterparts. Studies with picornaviral RNA-dependent RNA polymerases (replicases) face limitations derived from the difficulties in obtaining purified enzymes capable of sustaining multiple rounds of template-dependent copying. Viral quasispecies show features of complex adaptive systems such as mobilization of minority components (individual genomes from the mutant spectrum) in response to external stimuli. New developments in biochemistry and structural biology, and a deeper understanding of the principles governing viral evolution can now be combined to produce practical developments following an extensive (and necessary) accumulation of results from basic research.

This chapter discusses the strategy used by picornaviruses belonging to the genera of enteroviruses, rhinoviruses, and aphthoviruses to interfere with host cell protein synthesis. First it describes the mechanism of initiation of capped mRNA and compares it with that of uncapped mRNA. It then discusses the role of the viral proteinases and their cellular targets. Finally it focuses on the effect of specific proteolytic cleavage and its function in the initiation of protein synthesis. There is ample evidence that 2Apro is the viral proteinase responsible for eIF4G cleavage, which results in the inactivation of cap-dependent translation leading to the shutoff of cellular protein synthesis. In addition, based on the known cleavage sites of 2Apro of HRV2 and 2Apro of coxsackievirus B4 in the viral polyprotein, sequence comparisons allowed the prediction of three potential cleavage sites in eIF4G. It has recently been demonstrated using an in vitro translation system that newly synthesized 2Apro of HRV2 is highly efficient in eIF4G cleavage. The discrepancy was particularly dramatic when viral replication was partially inhibited by guanidine-HCl, 3-methyl quercetin, monensin, or nigericin. In the presence of these inhibitors, eIF4G cleavage was found to be essentially complete, whereas cellular protein synthesis was only partially inhibited. Furthermore, generally the same initiation factors are involved in the translation of both types of mRNAs.

The inhibition of host cell translation by poliovirus, also called host cell shutoff, occurs early in the infectious cycle, typically only 1.5 to 2.5 h postinfection in HeLa cells. This chapter explores information regarding cleavage of new initiation factors and then discusses information relevant to whether these factors, particularly eukaryotic translation initiation factor 4GI (eIF4GI), are cleaved in vivo by viral or cellular proteases or both. Researchers have discovered evidence for several eIF4Gase activities that cleave eIF4GI at different sites to produce distinct types of eIF4G cleavage products. In the chapter, researchers define eIF4Gase activities based on the size of cleavage products generated (which results from use of alternate cleavage sites) and the mode of induction of the activity. Researchers have identified three types of eIF4Gase activities (termed eIF4Gase-α, elF4Gase-β, eIF4Gase-γ) that can be generated without poliovirus (PV) infection; two of these appear similar to activities also present in infected cells. The sites, which generate the three major N-terminal cleavage products seen on our immunoblots, are 149 amino acids (aa) upstream of the 2Apro cleavage site and 106 aa upstream of the putative eIF4Gase-β cleavage site. Researchers have screened many S10 extracts with this procedure and found that this specific cleavage activity, which is often weak, appeared in approximately 40% of the extracts tested.

Poliovirus (PV) is known to shut off both host cell transcription and translation. It is believed that the shutoff of host cell transcription in PV-infected cells increases the pool of free ribonucleotides that the PV-encoded RNA-dependent RNA polymerase (Pol) uses to transcribe and replicate the viral genomic RNA. In support of this theory, PV first shuts off Pol I-mediated transcription in the cell that accounts for greater than 50% of all host cell transcription. The authors have used in vitro transcription systems for understanding the mechanism by which PV shuts off host cell transcription catalyzed by RNA Pol I, II, and III. Infection of susceptible cells with PV results in rapid and dramatic changes in macromolecular metabolism, including the shutoff of host cell transcription. Early attempts to identify the cellular components of the transcriptional machinery inactivated by picornavirus infection focused on the polymerases. To examine whether 3Cpro is sufficient to cause inhibition of host cell transcription seen in virus-infected cells, 3Cpro was cloned into the eukaryotic expression vector pCDNA. The role of the 3C protease in host cell transcription shutoff is clear from both genetic and biochemical analyses. PV is an RNA virus, which replicates in the cytoplasm of infected cells. To shut off host cell transcription, one or more viral gene products must enter the nucleus of the infected host cell. The viral precursor 3CD has protease activity and is able to autocatalyze the formation of 3C and 3D polypeptides.

As with the majority of cytolytic animal viruses, picornavirus infection leads to profound alterations in cellular membranes. Three types of changes are observed at late times of infection in cellular membranes: enhanced membrane permeability, proliferation of intracellular membranous vesicles, and inhibition of vesicular trafficking with the consequent blockage of protein glycosylation. This chapter focuses on both the structural and functional modifications that membranes of picornavirus (PV)-infected cells undergo during virus replication. Distinct portions of the vesicle membranes are very densely stained and thicker than typical intracellular membranes. PV 2B, 2BC, and 3A proteins are able to block glycoprotein transport when they are expressed individually in mammalian cells. Sedimentation of cytoplasmic extracts in sucrose gradients yielded viral RNA polymerase activity associated with the smooth membranes. Viral replication complexes from this fraction contain all types of PV RNA and several viral proteins involved in RNA replication. The sequence of protein 2B is one of the least conserved among picornaviruses. The permeabilizing activity of coxsackievirus B3 (CVB3) 2B has been implicated in virus release, since viruses carrying a mutant 2B protein exhibited a defect in virus yield. However, in the case of PV 2B, it has been reported that this protein is located mainly in the central portion of the cytoplasm, associated with the membranous vesicles that surround the viral replication complexes. PV protein 2C is a 329-amino-acid polypeptide that contains a typical nucleoside triphosphate (NTP)-binding domain. Its sequence is one of the most highly conserved among all picornaviruses.

The picornaviruses are a diverse group of human viral pathogens that together comprise the most common causes of infections of humans in the developed world. Within the picornavirus family are three well-known groups of human pathogens—the human rhinoviruses (HRVs), the enteroviruses (EVs) (including polioviruses, coxsackieviruses, and echoviruses), and the hepatoviruses (including hepatitis A). This chapter focuses on the rhinoviruses and enteroviruses. The HRVs include more than 100 serotypes in two main groups based on their cellular receptors. The major risk factor for HRV infection appears to be contact with young children. The majority of symptomatic EV infections in the United States are characterized by minor EV illnesses associated with fever and constitutional symptoms, with or without rashes. These illnesses are of clinical significance because they may mimic other diseases including bacterial sepsis, other viral exanthematous diseases, and herpes simplex infections. A recent review of EV-associated respiratory illnesses found that 46% of cases presented with upper respiratory infections, 13% with respiratory distress/apnea, 13% with pneumonia, 12% with otitis media, and fewer cases with bronchiolitis, wheezing, croup, and pharyngotonsillitis. New developments in the rapid diagnosis and therapy of these infections promise to significantly reduce the disease burden and the associated costs to affected individuals and to society.

This chapter provides a brief synopsis of the natural history of paralytic poliomyelitis, and gives an overview of the status of research concerning the molecular determinants of the pathogenesis of paralytic poliomyelitis. Determinants of the pathogenesis of poliomyelitis are either of viral origin, e.g., non-coding viral sequences, structural or nonstructural viral gene products, or of host origin, e.g., distribution of the cellular receptor and host cell factors required for viral replication. To provide a rational account of the relative contributions of a multitude of factors toward a complex phenomenon, the chapter is subdivided into sections dealing with the main parameters of poliovirus neurological disease. Tropism, neurovirulence, and conditions of the host are discussed separately. The chapter discusses experimental evidence for the genetic basis of neurovirulence in the 5’ non-translated region (5’ NTR) and the coding regions for the structural and nonstructural proteins of poliovirus. Extraneural determinants of neuropathogenicity, such as invasion of or spread within the CNS, combine with intraneural factors, such as IRES-mediated cell type specificity or the efficiency of genome replication. Excellent studies in nonhuman primates in the prevaccine era and recent progress through the advent of genetic engineering and transgenic animal models for human disease have afforded us detailed insight into the pathogenic mechanism of paralytic poliomyelitis.

The pathogenesis of poliomyelitis is central to understanding the effectiveness of the vaccines that are likely to eradicate the wild-type virus from the world over the next few years. Studies of the molecular biology of the Sabin live polio vaccines have so far concentrated on their virulence or attenuation for primates or transgenic mice carrying the human receptor for poliovirus where the virus is given directly into the central nervous system or parentally. The basis of attenuation or reversion of the Sabin vaccine strains has been studied by comparing the vaccine strain of each serotype with a closely related strain, either the precursor of the vaccine strain or an isolate from a vaccine-associated case of poliomyelitis. Monoclonal antibodies can recognize neutralization sites specific for vaccine rather than wild-type strains, and the most strain-specific antibodies for any serotype are directed against site 3 (Nag III), composed of sequences from VP3 and VPl. A study was conducted in 1962 to assess the ability of immune deficient hypogammaglobulinemic patients to mount an immune response to vaccines, including polio vaccine. A strain of virus with a completely stable phenotype excreted for a brief period that was still sufficient to immunize the recipient would be valuable, and there have been attempts to develop such strains, based on both classical methods and current understanding of the molecular biology of the virus.

The coxsackieviruses are the best-studied group of nonpolio enteroviruses. Unlike that of the polioviruses, the immune reactions of the human or murine (in models of human disease) hosts of infections by the coxsackieviruses are an important component of the diseases generally induced by these viruses. The first T-cell epitope of the coxsackieviruses to be recognized is located in the 2C protein and is held in common with other human enteroviruses; the predominant lymphocyte population proliferating in response to the 2C epitope was CD4+ T cells. In a study using the CVB4-induced murine pancreatitis model, CVB4-inoculated nude and severe combined immunodeficient SCID mice had increased viral replication in the pancreas and increased mortality at early stages of disease, but CD4 knockout mice were protected. This work suggests that CD4+ T-cell responses in the pancreas (a tissue which, in intraperitoneally inoculated mice, achieves coxsackievirus titers as high or higher and more rapidly than serum titers observed in viremia) may be essential for survival during early virus-induced disease, but at later times these cells participate in increased inflammatory disease. In conclusion, immune responses against coxsackieviruses have been explored through the use of murine models of virus-induced heart and pancreatic diseases. The diversity of inbred mouse lines and viral strains leaves the significance of studies to be resolved by comparison with human disease or by a methodical dissection of the immune response to coxsackieviruses generated in the various murine strains.

Regarding the pathogenesis of enteroviral heart disease there has been uncertainty whether viral cytotoxicity or immune-mediated processes are crucial for organ pathology during acute and persistent heart muscle infection. This chapter provides experimental evidence for the decisive role of virus replication in the induction and maintenance of chronic myocardial damage. In addition, the capacity of cellular signal transduction pathways to modulate enterovirus replication as well as coxsackieviruses of group B (CVB) receptor interactions and their role in pathogenesis are discussed. It has been found that in all investigated mouse strains CVB3 is capable of inducing acute myocarditis, which is characterized by virus-induced myocytolysis and reactive formation of interstitial mononuclear infiltrates. Importantly, persistent infection of myocytes was found to be related with morphological changes of the myofibrils. Preferential targets of the enteroviral proteinase 2Apro are proteins involved in the Cap-dependent translation of cellular mRNAs, since enteroviruses employ a Cap-independent mechanism of protein translation. Importantly, recognition of cellular targets by viral proteinases may be preceded by virus-induced modifications of these proteins. As recently suggested for infections with encephalomyocarditis virus, activation of p38/MAPK could be induced by the dsRNA intermediates during enteroviral replication.

In 1973, Feinstone and coworkers first visualized hepatitis A virus (HAV) in the feces of an individual with acute hepatitis A. HAV was originally classified as enterovirus 72, but subsequent analysis of the sequence of its genome in comparison with newer information about the genomic sequences of other picornaviruses led to its being reclassified in its own genus, Hepatovirus, which it shares with the closely related simian HAVs. The current hypothesis is that viral hepatitis is both caused and cured by the cytotoxic T lymphocyte (CTL) activity of CD8+ lymphocytes infiltrating the virus-infected liver. In general, wild-type strains of HAV recovered from clinical cases of acute hepatitis replicate extremely slowly and to low titer in vitro and inoculated cell cultures may require months of incubation before maximum amounts of virus are produced. Coincident with this adaptation to cell culture, the virus, in some cases, lost its virulence for primates or humans and acquired an attenuation phenotype. As hepatitis A vaccines are based on the immunogenicity of intact virions, an appreciation of the molecular basis for efficient growth in cell culture or of attenuation has been necessary for the development of such vaccines. Live attenuated hepatitis A vaccines also have been developed by isolation and blind serial passage in cells. The most promising candidate vaccines have been grown in primary or continuous monkey kidney cells. However, virtually all primary monkey kidney cells are contaminated with simian viruses and therefore not useful for vaccine propagation.

Strains of Theiler’s murine encephalomyelitis virus (TMEV) were first isolated by Max Theiler at the Rockefeller Foundation during his investigations of yellow fever virus vaccine. The GDVII strain of TMEV was named because this isolate was the seventh made by George, Theiler’s technician, from an uninoculated mouse that was found to be diseased (paralyzed). The presence of polypyrimidine tract binding protein (PTB) and nPTB binding sites in GDVII affects neurovirulence. The importance of sialic acid binding to virus entry is further supported by studies that showed that a variant of DA that has enhanced binding to L929 cells and has mutations in the VP2 puff B or VP1 loop 2; this region is also one that has been found important in disease pathogenesis. It was found that demyelination can occur when one replaces any part of the DA genome with the corresponding segment from the GDVII genome. In addition, the immune system is believed to mediate the pathology in both diseases. Therefore, it is hoped that investigations of TMEV-induced demyelination may provide insights into the pathogenesis of multiple sclerosis. The identity of the TMEV receptor, or what are probably TMEV receptors, is likely to help clarify the pathogenesis of TMEV-induced central nervous system disease. In addition, the identification of cell-specific RNA-binding proteins that affect translation and regulate the synthesis of L* versus the polyprotein is also likely to help in the understanding of picornaviral gene expression and TMEV disease pathogenesis.

Viral persistence is currently one of the main public health problems. It depends on specific virus and host factors and particularly on the capacity of the virus to escape the immune defense. Some picornaviruses establish persistent infections in their natural animal hosts, while others are suspected to be responsible for chronic human diseases. This chapter focuses on the aspects relating to virus persistence. Theiler's murine encephalomyelitis virus (TMEV) persistence appears to be required for immune-mediated demyelination. During the chronic phase, the number of infected cells is always low, there are only small amounts of capsid proteins in infected cells, and viral RNA replication seems to be blocked at the minus-strand RNA synthesis step. Encephalomyocarditis virus (EMCV), another cardiovirus, establishes persistent infections in human erythroleu-kemic K562 cells. B group coxsackieviruses (CVBs) induce different effects in different types of cultured human renal cells, and interestingly, CVB1, -3, -4, and -5 established persistent infections in glomerular mesangial cells, which failed to develop cytopathic effects (CPE). The authors have started to study the molecular mechanisms of poliovirus (PV) persistence in HEp-2c cells because this cell model can be used for both lytic and persistent PV strains, allowing the identification of viral determinants involved in persistence. Persistent picornavirus infections are valuable models for understanding complicated pathologies resulting from persistent viral infections.

This chapter discusses two processes of the polioviral life cycle that were successfully reproduced in the cell-free system, recombination and complementation. First, recombination, the components of the cell-free recombination system, and the alteration of crossover sites by temperature are discussed, and then the inadvertently discovered phenomenon of complementation are presented. The cell-free system has a limitation in that it works best when either viral RNA (extracted from purified poliovirus particles) or efficiently replicating in vitro-synthesized transcripts are used in conjunction with the HeLa extract. To prove that recombination had indeed occurred in the cell-free system, several controls were used. These experiments showed that only when the two parental RNAs were coreplicated, recombinants were generated. To analyze the recombinants, recombinant plaques were amplified once under the selection conditions and the reverse transcriptase PCR (RT-PCR) product sequenced to confirm the presence of the gr mutation and subjected to restriction analysis. To determine the frequency of recombination, it was important to determine the yield of the parental viruses in extracts programmed by the parental RNAs. When guanidine hydrochloride (Gua-HCl) was added to the cell-free recombination reactions, it resulted in a significant reduction of binants (only one out of six reactions gave a single recombinant plaque). This observation is significant because it may provide insights into the mechanism of recombination. Studies involving intertypic poliovirus recombination have suggested a role for RNA secondary structure and a preference for crossover to occur in loop regions of stemloop structures.

Poliovirus RNA replicates in membrane-associated replication complexes in the cytoplasm of infected cells. By using a reversible inhibitor of poliovirus RNA replication, it is possible to synchronize viral RNA replication. The processing of the viral polyprotein results in the formation of the individual viral proteins along with stable intermediates in the processing pathway. To expand the utility of the in vitro complementation assay, experiments were designed to determine if all of the viral replication proteins could be provided in trans to support the replication of mutant RNA templates. The authors engineered two transcript RNAs (DJB2 and DJB15) that contained large out-of-frame deletions in the polyprotein coding sequence. The results to date using the in vitro complementation assay indicate that the 5’ cloverleaf, the 3’ nontranslated region (NTR), and the poly(A) tail are the minimum sequences required for negative-strand synthesis. Previous studies have shown that the 5’ cloverleaf plays an important role in viral RNA replication. To investigate the role of the 5’ cloverleaf in negative-strand synthesis, the authors determined how cloverleaf mutations affected negative-strand synthesis in preinitiation RNA replication complexes. Results of these experiments showed that 5’ cloverleaf mutations dramatically diminished RNA stability and negative-strand RNA synthesis. The results of recent studies indicate that the 5’ cloverleaf is required for the initiation of negative-strand synthesis. Once the viral mRNA is cleared of translating ribosomes, it could then serve as a template for negative-strand synthesis.

Poliovirus is highly transmissible, with only 1 of every 100 to 1,000 infections among susceptible individuals resulting in clinically recognizable disease. The origin of global eradication of poliomyelitis is conventionally attributed to Albert Sabin and his colleagues in the frequently quoted and often reprinted 1960 report on the effects of rapid mass oral polio vaccine (OPV) immunization in Toluca, Mexico. For many, smallpox eradication would be a fluke in the annals of public health. No other infectious diseases were thought to share the unique characteristics that made it possible to eradicate smallpox: clinically apparent disease; low agent transmissibility; and an effective, inexpensive, and easily administered vaccine. The eradication of smallpox had proven the value of program goals based on disease outcome rather than service coverage. As with smallpox eradication, one of the anticipated major benefits of polio eradication is the cessation of immunization. Unlike the live smallpox (vaccinia) vaccine strain, Sabin OPV strains commonly spread from vaccinees to close nonimmune contacts and are genetically unstable, regaining certain wild virus characteristics upon replication in the human gut. The only way to fully control poliomyelitis in the developing tropical world with limited health infrastructure is to interrupt virus transmission through mass OPV immunization. Global eradication is a natural outcome.

Systematic international efforts to eradicate polio from the developing world began in the Americas in 1985, when the Pan American Health Organization (PAHO) declared a target date of 1990 for the eradication of polio throughout the Americas. Surveillance for wild polioviruses has two arms: (i) acute flaccid paralysis (AFP) case investigations and (ii) virologic studies of polioviruses obtained from clinical specimens. AFP surveillance by itself is neither highly specific nor highly sensitive for detecting individual wild poliovirus infections. Poliovirus eradication has achieved the elimination of individual lineages (equivalent to chains of transmission), different genotypes (groups of related lineages sharing >85% nucleotide sequence identity), and probably wild poliovirus type 2. Polio cases associated with these importations have revealed pockets of unimmunized children in the new host areas, prompting local immunization responses. However, by far the most effective response is to eliminate the source reservoirs. Various supplementary approaches are implemented to monitor ongoing laboratory performance. A serious challenge to the integrity of poliovirus surveillance data is the occurrence of poliovirus contamination of cultures. Recently, polio outbreaks associated with circulating vaccine-derived poliovirus (cVDPV) have been recognized in three different parts of the world. The occurrence of iVDPVs and cVDPVs appears to be rare. Moreover, most chronic poliovirus excretors in developed countries spontaneously stop shedding or die of complications from their immunodeficiency. As an increasing number of highly developed countries have switched to inactivated polio vaccine (IPV), the chances for the occurrence of new iVDPV infections have decreased.