Chapter 224. Viral Pathogenesis

Viruses can only survive by entering and parasitizing the cells of a host. Because viruses cannot live independently of their interaction with the parasitized host cell, all viruses have evolved strategies for coexisting with cells, while these cells continue to carry out physiologic functions as part of specific tissues or organs. It is the complex interplay between viral and host properties that determines the relationship that the virus will experience with its host in vivo, whether resulting in disease or not. The interplay between viral and host features thus determines the pathogenesis of infection.

Pathogenesis has generally been conceptualized as the series of sequential steps starting with entry of the virus into host cells, and progressing through survival of the virus and dissemination of the virus in the host, to shedding of the virus from the host to spread to new individuals where the virus continues its life cycle. In the animal host, every virus must overcome a series of defenses in order to enter, disseminate, and localize in the ideal target tissue; replicate; and shed to infect new hosts. At each stage, the host possesses specific mechanisms to inhibit the survival of the virus, and the virus has evolved ways to counteract these mechanisms. Some of these strategies for evading host defenses are common to different viruses. Considering viral pathogenesis in this way as the stepwise unfolding of events within the host is useful and simple because thus far we know far more in most cases about the virus itself than we do about the host side of the complex interplay. When applying this model of viral pathogenesis, however, often the steps may not occur sequentially, but rather several steps may be occurring at the same time, or several alternate pathways may be taken simultaneously.

As new understanding of the host contributions to viral pathogenesis emerges, the impact that an individual's genetic makeup has on susceptibility to infectious disease has become clear. In fact, recent data suggest that an error in a single gene is enough to dramatically alter an individual's susceptibility to viral infections.1 It is becoming apparent that pathogenesis should likely, in the future, be conceptualized as a balance between the viral agent and the host’s immunologic and genetic makeup.

Genetic Susceptibility to Infection

For respiratory syncytial virus (RSV), several abnormal underlying conditions that predispose to severe forms of disease have been enumerated, and include prematurity, preexisting lung disease, and various forms of immunodeficiency. However, it is not yet known why some apparently normal infants and children proceed from initial infection to severe lower respiratory tract disease, whereas others experience a relatively mild, self-limited illness. Recent evidence now points to a major role for genetic susceptibility. Certain alleles of IL-4,10 and of the IL-4 receptor,11 are associated with more severe disease, and promoter variants of IL-10, IL-9, and TNF-α genes likely also influence disease severity.12 Variations at the IL-10 gene locus are associated with a severe form of disease,13 and a relationship probably exists between a locus on the IL-8 gene and disease severity.14,15 Variants of the chemokine receptor CCR5 also seem to predispose to severe RSV bronchiolitis,16 and a correlation exists between specific alleles of the genes for surfactant A and D and an increase in disease severity.17,18 In light of this mounting evidence for specific genetic contributions to RSV disease severity, and the possibility that similar or related genetic variation may underlie predisposition to asthma (and the link between respiratory virus infection and asthma19), it is important to understand the mechanisms whereby these gene alterations influence pathogenesis.

Entry

Entry into the host is the step that allows a virus to establish infection in host cells, after overcoming environmental, anatomic, or innate immune mechanisms to gain access to its target. The virus can enter in infected cells that can be injected for example via the mouth parts of infected arthropods; can be contained in droplets or fomites shed from an infected host, or can be inhaled or ingested as free virus. The six portals through which viruses can enter the human body include five epithelial surfaces: skin, conjunctiva, respiratory tract, gastrointestinal tract, and genitourinary tract. Viruses can also enter via the interface between the mother and germ cell or developing fetus.

The barriers to viral entry include general anatomic barriers, defenses based on the immune system’s recognition of patterns (eg, via Toll receptors),2 other innate immune mechanisms, and adaptive immune responses. Epithelial tissues are now recognized to be active immune organs, containing specialized dendritic cells that serve as sentinels for invasion, which when activated by viruses can initiate the induction of immune responses. Lymphocytes are also present in epithelia, generating protective immune responses at this barrier. The epithelial cells themselves can also produce a cellular antiviral response, all taking place before the virus even reaches the stage of inactivation by antibodies and complement.

Spread

When a virus passes through the superficial epithelial barriers, and confronts the intrinsic cellular resistance to infection it then can spread. Most viruses spread locally from cell to cell; some remain at or near the site of entry, whereas others cause viremia and spread widely. Spread of viruses can occur vertically via the placenta or the germ line. Transplacental spread is an important mechanism for several human pathogens, including the herpesviruses human cytomegalovirus (HCMV), varicella-zoster virus (VZV) and herpes simplex virus (HSV), hepatitis viruses B and C, and HIV. Spread via the germ line is the mechanism whereby human endogenous retroviruses enter the human genome; the consequences of this spread to humans are only beginning to be understood.3 Systemic spread of viruses through the infected host can occur via the blood, the lymphatic system, or the nerves. The most important route of dissemination for non-CNS viruses is the bloodstream during viremia. Viruses may circulate either free or cell associated, depending on the individual virus, and may invade tissues from the bloodstream. A number of important human viruses spread through the axons of peripheral nerves and are thus carried to the CNS.

Tropism

Cell and tissue tropism of viruses is determined by the interplay between the specific viral feature, the required conditions for replication, and host features. These include cellular receptors, innate antiviral immunity, intrinsic cellular resistance, the presence of immunoprivilege,8 and the state of differentiation of the cell or tissue.9 Innate immune detection of virus infection is mediated by host pattern recognition receptors including RNA helicases RIG-I and MDA5 and several members of Toll-like receptors (TLR) such as TLR3, TLR7 and TLR9. Of these, RIG-I, MDA5 and TLR3 function as sensors for double-stranded RNA species that are produced by many viruses during replication, and trigger interferon-β induction.4,5 Examples of intrinsic cellular resistance include proteins such as the APOCBEC antiviral proteins that corrupt the genome of invading viruses by inducing mutations, and TRIM5α interferes with the uncoating of retrovirus, thereby preventing the successful reverse transcription and transport to the nucleus of the viral genome.6,7 Immunoprivilege refers to the ineffectivity of the immune system in clearing virus infection from specific tissues; for example, CD8 T cells may be more effective at eliminating infection from MHC class I expressing hepatocytes than from neurons that do not express MHC class I molecules. As suggested previously, the list of host features that determines tropism lengthens as we learn more.

Fate

The successful establishment of a virus in its target tissue may, but does not necessarily, result in damage or disease. The variables that determine the outcome of infection lie in the interplay between features of the virus, including the many factors that determine virulence, and features of the host. One determinant of susceptibility to viral infections that is particularly relevant in pediatrics is age. Many viral infections are far more severe in younger than older hosts. One example of this difference is hepatitis B virus (HBV) exposure. Most adults infected with HBV clear the infection, whereas exposed neonates have a very high level of chronic infection. The actual determinants of disease are not fully understood for most viruses. In large part, the outcome of infection depends on whether the virus is cleared from the infected tissue; whether the immune system, in efforts to clear the virus, damages host tissue; and whether the virus induces autoimmune responses.

Infected cells may be killed either directly by the virus in its subversion of cell metabolism, or by triggering cell-intrinsic programmed cell death pathways (apoptosis). Virus-infected cells can also be killed by lysis of infected cells by the complement system, or via the effector molecules released by NK cells or cytotoxic T cells. Some viruses, despite rapid clearance by the immune system, manage to cause significant tissue damage (eg, variola). Other viruses are less likely to be cleared by the immune system, and result in latent, chronic, or progressive infection (eg, hepatitis B). Some viruses can cause either type of infection—acute or persistent—in different patients (eg, measles causes both acute disease and subacute sclerosing panencephalitis [SSPE]). Even during beneficial immune responses, the immune system can cause significant tissue damage, and in many cases, virus-induced pathology is actually caused by the immune response and is referred to as immunopathologic.

One variant of immunopathology is virus-induced autoimmunity, which can arise either through molecular mimicry or during persistent virus infections without evidence for cross-reactive antigens. Many viruses can, under certain conditions, cause persistent infections that last for the lifetime of the host. Such viruses can only accomplish this feat by maintaining their own genome and evading host immune responses, without destroying the host. Herpesviruses, for example, cause latent infections in the nervous or lymphoid system that may remain silent for the individual’s lifetime or reactivate, especially in the setting of immunosuppression, to cause another disease such as HSV rash or encephalitis, and zoster, due to VZV. Certain herpesviruses such as Epstein-Barr virus (EBV) and human herpesvirus 8 (HHV8) can also reactivate and cause malignancy such as lymphoma and Kaposi sarcoma, respectively. Other viruses associated with persistent infection and development of cancer include hepatitis B and C viruses, and human papilloma viruses 16 and 18.

Shedding and Transmission

Viruses that cause acute infection generally are shed heavily from the host in bodily secretions, to allow for spread to the next host. Persistently infecting viruses may shed very little, but nonetheless enough for eventual transmission. The strategies used by viruses for shedding depend on their ability to survive for brief periods under various environmental conditions. Common strategies for transmission include contamination of hands by bodily secretions; inhalation of aerosolized virus; direct person-to-person contact, including sexual or skin contact; and indirect person-to-person contact via blood. Transmission can also occur via contaminated water, food, or biologics, for viruses that can survive these environments. Efficient transmission is a key to survival of the virus and perpetuation of the viral life cycle.

Acute Viral Infection: Parainfluenza and Respiratory Syncytial Viruses

Respiratory tract infections caused by the human parainfluenza viruses (HPIVs) and respiratory syncytial viruses (RSVs) cause many of the most significant childhood diseases in both developed and underdeveloped areas of the world. These viruses, together with human metapneumovirus, are the predominant causes of bronchiolitis, croup, and viral pneumonia, acute respiratory infections in young children.

HPIV and RSV are members of the Paramyxoviridae, and contain a negative-stranded nonsegmented single-stranded RNA genome packaged within a lipid envelope that is composed of host cell lipids and viral glycoproteins derived from the plasma membrane of the host cell during budding. For HPIV, the F and HN genes encode the two integral membrane glycoproteins, the fusion protein (F) and the hemagglutinin-neuraminidase (HN). The F glycoprotein mediates fusion between the viral and host cell membranes during infection, whereas the HN glycoprotein, possessing both receptor-binding and receptor-destroying activities (ie, hemagglutinating and neuraminidase functions), is responsible for attachment of the virus particle to receptors on the cell, activation of the F protein to mediate fusion between the viral and cell membranes, and release of progeny virus particles from infected cells. For respiratory syncytial virus, the receptor-binding protein is not HN but G, which has receptor-binding but no receptor-cleaving activity.

Viral Entry and Membrane Fusion: The Roles of Viral Surface Glycoproteins

For the paramyxoviruses, infection of cells is initiated by attachment of the virus to the host cell through interaction of the receptor-binding glycoprotein with a receptor molecule on the host cell surface. On interaction with receptor, the receptor-binding glycoprotein triggers or activates the viral fusion protein (F) to its fusion-ready state, promoting fusion of the virus into the cell and initiation of infection.20 The fusion glycoprotein F is synthesized as a single polypeptide chain that forms a trimer before being cleaved by host cell proteases to yield a membrane-distal and membrane-anchored subunit. The new N-terminal region of the membrane-anchored subunit, termed the fusion peptide, contains hydrophobic residues that insert into target membranes during fusion at neutral pH (reviewed in ref. 21). Penetration and uncoating of the virus occur by fusion of the viral envelope with the plasma membrane of the cell, resulting in the release of the viral genetic material into the cytoplasm. Infection also results in fusion between cells, which involves the interaction of receptor-binding proteins and fusion proteins expressed on the surface of an infected cell with the membrane of an adjacent uninfected cell, and leading to “syncytia” or giant multinucleated cells. The role of the receptor-binding surface glycoprotein in promotion of fusion during viral entry has been elucidated in recent years, via the study of the role of the HPIV HN in this process.20,22-26 Unlike influenza viruses, in which separate surface glycoproteins are responsible for the neuraminidase (NA) and receptor-binding (HA) activities, the paramyxovirus HN is a dual function molecule, carrying out both receptor binding during entry and receptor cleavage during budding and release of virus. More recently, the third function of HN has been identified; on binding to receptor, parainfluenza HN plays a critical role in “triggering” F to undergo its final conformational alteration that permits fusion.20 One bifunctional active site on HN possesses both binding and neuraminidase activities, whereas a recently identified second site on HN possesses binding and F-protein–triggering activities.25,27

In the case of influenza virus (an orthomyxovirus, also a cause of acute respiratory infection), the balance of NA and HA activities is important for the retention of in vivo viability. If virus binds receptor avidly, then entry will be efficient and release will be hindered, resulting in loss of viability; if receptor binding is reduced drastically, then entry of the virus is compromised, resulting in a similar loss of viability. Balance of the dual functions of paramyxovirus HN is also important for the functional characteristics of HPIV, even though the distinct biologic activities of receptor binding and receptor destruction reside on a single molecule.23,28 The balance between HN’s two functions—binding and release from receptor—ultimately determines the third key HN function, the promotion of fusion and thus viral entry,20,29 and influences the outcome of viral infection.

Transcription and Replication of the Viral Genomes

Infecting virions fuse with the plasma membrane of the cell, releasing the viral nucleocapsid into the cytoplasm. The nucleocapsid contains the genome RNA in tight association with the viral NP protein, and this RNA/protein complex is the template for both transcription and genome RNA replication. The first step in the viral growth process, termed primary transcription to denote transcription directly from the infecting nucleocapsid template, initiates at the 3'-end of the genome and results in each of the mRNAs being transcribed and present in infected cells as a function of its relative location on the negative sense viral genome. For five of the six genes of the human parainfluenza viruses, transcription generally leads to generation of a single mRNA species that encodes a single protein molecule. However, the P gene of the parainfluenza viruses (along with a number of other pathogenic paramyxoviruses) encodes a number of proteins and, in several members of the family, more than one mRNA is synthesized, and the mRNA also encodes a number of smaller protein products in an alternate reading frame from the P protein. These “C” proteins, as they are commonly called, are a nested set of carboxy-coterminal proteins, with each protein encoded in the same reading frame but differing by its start site on the mRNA. The functions of the C proteins are incompletely understood but are now known to function in pathogenesis in the infected host, playing key roles in evasion of the host immune response.30

Replication of the viral genomic RNA is a process both linked to and distinct from transcription of the genomic RNA. Unlike transcription, replication of the viral RNA genome depends on ongoing protein synthesis, and therefore is linked to prior transcription of the viral genome. The second stage of genome RNA replication occurs when the newly replicated and encapsidated full-length plus-stranded RNA (antigenome)-containing nucleocapsids are copied into negative-strand RNA-containing progeny nucleocapsids identical to the nucleocapsid that is packaged into progeny viral particles. These progeny viral nucleocapsids are templates for transcription and can greatly amplify the amount of viral mRNA and protein produced in an infected cell.

Assembly and Budding of the Viruses

New viruses are formed by budding of newly replicated and assembled nucleocapsids containing the viral RNA genome along with the P and L proteins through modified areas of the plasma membrane that contain the transmembrane glycoproteins and the M protein. In polarized epithelial cells, the viruses bud from the apical surface of the cell. It is believed that the nonglycosylated M protein binds to the nucleocapsid and also likely interacts with the cytoplasmic tails of the transmembrane glycoproteins, thereby mediating the alignment of the nucleocapsid and the areas of the plasma membrane containing viral glycoproteins in preparation for budding.31-36

Evasion of Host Defense Mechanisms

Host cells have developed a variety of countermeasures in the face of viral infection. In some of these, an intracellular environment is created that is less hospitable for virus replication (the interferon [IFN]-induced antiviral state); some cell types undergo apoptosis, which limits viral replication. Many viruses possess strategies for evading these host immune responses, particularly the innate immune response, which plays a key role in controlling virus infection. Many viruses encode gene products that are capable of inhibiting the host interferon response37; for the paramyxoviruses, evasion of this response is mediated by the C and V proteins. The Sendai virus (mouse HPIV1) C gene was one of the earliest examples of an overlapping gene,38 and the study of this virus has yielded much understanding about this pathogenetic mechanism. For Sendai virus, evasion of the interferon system is accomplished in several ways by the C protein.39 Sendai virus C protein binds STAT1, preventing its activation in response to IFN, and the carboxyl part of the C protein mediates this activity. In addition, Sendai C protein targets STAT1 for degradation, and the amino-terminal residues of the C proteins mediate this activity.39 The “V” proteins, also read from the P genes, require the pseudotemplated addition of a single G residue to switch from the N terminal P to C terminal V open reading frame (ORF).40,41 The V proteins are involved in a wide range of mechanisms for evading the immune response. These have been found thus far to include prevention of apoptosis,42,43 cell cycle alterations,44 inhibition of double-stranded RNA signaling,42,45 and prevention of IFN biosynthesis.42,43,45 V protein expression has been shown for several paramyxoviruses to inhibit IFN-responsive STAT proteins. The strategies whereby different V proteins inhibit STAT proteins have been found to be highly diverse. For example, SV5 V proteins target STAT1 for degradation,42,46 whereas HPIV type 2 V protein targets STAT2.47,48 Mumps virus V protein can target both STAT149 and STAT3,50 and can also interact with cellular RACK1, potentially modifying IFN receptor activity.51

Use of Cellular Machinery for the Viral Life Cycle

The potential role of cytoskeletal molecules in the process of viral entry,52,53 as well as the question of how the various viral constituents are directed during the parasitic phase of the life cycle to the correct cellular locations, have yet to be clarified. It is likely that entry, fusion, and/or budding involve the participation of the cellular actin cytoskeleton and microtubule function.54-57 Host cellular proteins that are involved in cell mobility may be involved; one possible component of the entry/fusion mechanisms involves interaction between paramyxovirus F proteins and Rho A.52,53

Disease Induction and Immunity

Bronchiolitis is the result of a balance between cellular damage mediated by the viral pathogen (most commonly, respiratory syncytial viruses [RSV]) and injury caused by the immune response of the host.58 The viruses first replicate in the nasopharyngeal epithelium and spread to the lower respiratory tract. Viral replication in the small airways causes inflammation, sloughing, and necrosis of epithelium. Edema and increased mucous secretion cause plugging of the small airways, atelectasis, airway narrowing, and obstruction. RSV primary infection does not confer permanent immunity; in fact, repeated reinfection with RSV within a year of the previous infection is common in young children, although subsequent infections are usually milder, suggesting some protection against severe disease after primary infection, or possibly immunologic immaturity or simply the small size of the airways of the infected infants. In adults, secretory neutralizing antibodies (but not serum antibodies) correlate with protection against upper respiratory tract infection, whereas circulating serum antibodies, particularly against F and G glycoproteins, have been shown to protect from infection and decrease progression to the lower airways.59,60 The limited degree of protection offered by maternal antibody is underscored by the fact that the peak incidence of serious RSV disease is seen in infants ages 2 to 5 months, a time of life when maternal antibody is still circulating within the infant.

RSV-induced lower respiratory tract disease (bronchiolitis and pneumonia) is the result of a balance between cellular damage mediated by the viral pathogen and injury caused by the immune response of the host.58,61,62 Although the immunopathogenesis of RSV infection is not fully understood, it is clear that both humoral and cellular components of the immune system contribute not only to protection from disease, but also to pathogenesis of disease. Mouse models of RSV disease have been used to dissect the contribution of different T-cell subsets and RSV proteins to the pathology of RSV infection and have shown different disease outcomes, depending on the RSV protein used to prime and the cellular response.63 Recent studies have revealed that the pattern recognition receptor CD14 (Toll-like receptor 4) participates in the innate immune response to RSV,64-66 a response triggered by the F protein. As in animal models, it is likely that in children cell-mediated immunity contributes to host defense against RSV but at the same time causes much of the pathologic process, and inappropriate immune responses may drive pulmonary inflammation during naturally acquired infection.67 Regulation of the response of T lymphocytes to RSV may be critical in determining the clinical outcome of RSV infection. Abnormal T-cell regulatory mechanisms may be related to a hyperactive IgE response, which contributes to an enhanced lung infiltrate .68,69 A number of the proinflammatory cytokines detected in respiratory secretions from RSV-infected individuals, including IL-8, RANTES, and MIP-1-α, mediate neutrophil and eosinophil chemotaxis, and these cell types can promote both host defense and host tissue damage.

Parainfluenza viruses also replicate in the epithelium of the upper respiratory tract, spread to the lower respiratory tract within 3 days, and croup results from inflammatory obstruction of the airway. Epithelial cells of the small airways may become infected with resultant necrosis and inflammatory infiltrates. In many cases, it is likely that, as for RSV, disease severity is increased, and the pathology of clinical disease is actually caused by the inflammatory response rather than by the cytopathic effects of the virus. This fundamental concept is highlighted by the fact that virus titers in the infected host are generally waning by the time disease symptoms become apparent70 and that virus titer does not correlate with the severity of lower respiratory disease. HPIV primary infection also does not confer permanent immunity; however, although reinfection occurs, immunity is usually sufficient to restrict virus replication from the lower respiratory tract and prevent severe disease. Mucosal IgA levels correlate with protection from replication of parainfluenza viruses in humans.71,72 Cell-mediated immunity also figures importantly in prevention of disease: for example, HPIV-3 infection of T-cell–deficient infants can cause a fatal giant cell pneumonia,71,72 and in bone marrow transplant recipients, HPIV pneumonia has a 30% mortality.73

Clearly, even in these acute infections in humans caused by paramyxoviruses, pathogenesis results from a complex interplay between viral properties, including strategies for entry, replication and immune evasion, and host factors, including age, genetic susceptibility, and balance of immunity. The next section uses the example of herpesviruses to show how this complex interplay is engaged in a model of chronic viral infection.

Pathogenesis of Lytic and Latent Viral Infection: Varicella-Zoster Virus

Infections with varicella-zoster virus (VZV) may be acute, latent, and/or persistent. Infection begins with entry of the virus into a virus-naive cell. In the infected individual, the host reacts to the virus initially through innate immunity, which is hypothesized to lengthen the incubation period of varicella, partially keeping the host in control of the virus prior to the development of adaptive immune responses.74 However, VZV reacts to the host by a number of strategies whereby the immune reactions of the host are evaded, giving the virus increased opportunity to replicate.75,76 After varicella develops, the host is almost always protected against future invasions by new VZVs to which the individual is exposed, by the host’s specific antibodies or cell-mediated immunity (CMI), or both.77 The virus is not eliminated, however, and remains sequestered in a latent form in sensory neurons. In 20% of individuals who have had varicella, it reactivates from latency and resumes its lytic cycle for a period of time ranging from days to weeks, and then usually comes under host control again by VZV CMI.77 It is an evolutionary advantage for VZV not to overwhelm its host, so that it can return to its infectious state years to decades later and be transmitted to new, younger susceptibles. This section concentrates on two aspects of VZV, its entry and synthesis in cells and mechanisms by which latency occurs and may temporarily escape control of the immune host.

Entry of cell-free VZV particles appears to use several receptors on the cell surface. These include heparan sulfate proteoglycan (HSPG), mannose 6-phosphate receptors (MPRci) ,64,78 and the cellular insulin-degrading enzyme (IDE). IDE is a receptor for gE.79,80 Entry of cell-free VZV occurs mainly by clathrin-mediated endocytosis and is highly cholesterol dependent.81

After entry and uncoating, VZV reaches the nucleus and begins to replicate, synthesizing its DNA, which is then encapsidated. The capsid is temporarily enveloped from the inner lamella of the nuclear membrane; it is then deenveloped as it passes through the rough endoplasmic reticulum to reach the cytoplasm as an unenveloped capsid.82 Viral capsids receive a final envelope at the trans-Golgi network (TGN). Vial glycoproteins are transferred independently to the TGN, where the tegument is added and final envelopment occurs. At least three of the VZV glycoproteins contain the unusual sugar mannose-6-phosphate (Man-6-P).83 The presence of Man-6-P on the viral envelope results in an interaction between VZV and MPRci that results in the inclusion of virions into endosomes. Here they are exposed to an acid environment that decreases viral infectivity, resulting in an absence of extracellular infectious VZV, although inactivated particles are released from cells.84 It appears that this cell-restricted form of viral spread takes place in the human host as well as in cell cultures in the laboratory.85,86 Exclusive cell-to-cell spread within the body would explain why host defense against this virus is mainly due to cell-mediated immunity rather than antibodies. This mechanism of cell-to-cell spread within endosomes leading to inactivation of much of the replicating virus would also seem to control the viral load and protect the host from overwhelming infection.84 In contrast, in the upper layers of the human epidermis, MPRcis are not present, due to loss of the lysosomal pathway in keratinized skin. In patients with acute disease due to VZV, therefore, cell-free, highly infectious virions are produced in cells of the superficial epidermis, which are released from cells into skin vesicles. These cell-free enveloped virions are hypothesized to play roles in airborne transmission of VZV and also in establishment of latent infection in neurons.84 Enveloped cell-free virions infect axons present in the area of the epidermis, leading to latent infection of neurons.84 In an in vitro animal neuronal model, infection of neurons by cell-associated VZV leads to neuronal death, whereas infection with cell-free VZV leads to latent infection and survival of neurons.

Varicella-zoster virus (VZV) is thus able to spread by two routes.78 One is from one infected cell to a neighboring uninfected cell by fusion of the infected and uninfected cells. Cell fusion is probability mediated by VZV gps B, E, I, and H, which are present not only on the enveloped virion, but also on the plasma membrane of VZV-infected cells. Cell-to-cell spread does not require envelopment of virions.87 Cell-to-cell transmission is probably the major method of spread of VZV in the body. The second form of spread uses cell-free enveloped virus. The major, if not the only, source of cell-free VZV in the body is in skin vesicles, MPRs are not present and the virus is therefore not inactivated in the acid environment of endosomes.84 Latency can be established in an in vitro animal neuronal model only with infection with cell-free VZV.78

Latency of VZV is different than that of HSV, although, like HSV, latent VZV infection occurs in sensory neurons.88-90 During HSV latency only one region of the genome, latency-associated transcripts (LATS), is expressed.77,78 In contrast, VZV expresses at least six genes (and thus their proteins) during latency, including three immediate early (IE) genes (ORFs 4, 62, and 63). The IE ORF 61, however, is not expressed. Based on an in vitro guinea pig neuron model, when ORF 61 is vectored into a latently VZV-infected neuron, lytic infection ensues, with expression of all 71 VZV genes, and neuronal death.78 Thus, ORF61 may act as a “switch” between latency and lytic VZV infection. ORF 63 protein, the predominant latency protein, inhibits apoptosis of neurons infected with VZV in vitro, and thus may play a role in maintaining latent infection.91 Certain early (E) genes are also expressed in VZV latency (ORFs 21, 29, and 66), but late (L) genes are not expressed. Expression of L VZV genes (eg, glycoproteins) indicates lytic infection.88,92-98 In VZV latency, it appears that the cascade of viral synthesis is interrupted by a block in the normal cascade of gene expression, involving failure of expression of ORF 61p.96 Of probable importance is that the VZV virion, which is critical to establishment of latent infection, lacks ORF 61p, although the particle carries the DNA for this protein. In the experimental in vitro model of VZV latency in guinea pig enteric neurons alluded to previously, there is no viral replication, and the neurons survive in vitro for weeks. Infection of neurons with cell-associated VZV, however, results in replication with rapid neuronal death, despite the presence of ORF 63 protein.78,92

Although the neuronal model described previously explains a portion of the pathogenesis of VZV latency, many aspects of latent VZV infection in humans still remain unclear. Patients with impaired cell-mediated immunity to VZV, such as the elderly and the immunocompromised, have an increased incidence of herpes zoster (HZ), which is consistent with the hypothesis that at least some aspects of suppression of VZV infectivity in humans are under immunologic control.99 Exactly how this process occurs, however, is not yet understood.

There are also many clinical data to support the concept seen in the neuronal model, that a varicella-zoster virus (VZV) rash, either due to the vaccine (Oka) strain or the wild-type virus, precedes latent infection and subsequent reactivation.78 In leukemic children vaccinated against chickenpox, the presence of a rash either due to the Oka vaccine strain or the wild-type virus (in vaccinees who develop mild varicella even if they were immunized) is highly linked to development of zoster.100 Immunocompromised vaccinees with no rash rarely develop zoster and may be spared latent infection, although the Oka strain is quite capable of establishing latent infection and causing zoster, as long as they have no VZV rash.78,101 If VZV can be prohibited from infecting the skin, avoiding exposure of sensory neurons to cell-free VZV at that site, then latent VZV infection may not occur.