Varicella virus isolation from spinal ganglion

Varicella-Zoster Virus-Genetics, Molecular Evolution and Recombination

This chapter first details the structure, organization and coding content of the VZV genome to provide a foundation on which the molecular evolution of the virus can be projected. We subsequently describe the evolution of molecular profiling approaches from restriction fragment length polymorphisms to single nucleotide polymorphism profiling to modern day high-throughput sequencing approaches. We describe how the application of these methodologies led to our current model of VZV phylogeograpy including the number and structure of geographic clades and the role of recombination in reshaping these.

Pathogenesis of varicelloviruses in primates

Varicelloviruses in primates comprise the prototypic human varicella-zoster virus (VZV) and its non-human primate homologue, simian varicella virus (SVV). Both viruses cause varicella as a primary infection, establish latency in ganglionic neurons and reactivate later in life to cause herpes zoster in their respective hosts. VZV is endemic worldwide and, although varicella is usually a benign disease in childhood, VZV reactivation is a significant cause of neurological disease in the elderly and in immunocompromised individuals. The pathogenesis of VZV infection remains ill-defined, mostly due to the species restriction of VZV that impedes studies in experimental animal models. SVV infection of non-human primates parallels virological, clinical, pathological and immunological features of human VZV infection, thereby providing an excellent model to study the pathogenesis of varicella and herpes zoster in its natural host. In this review, we discuss recent studies that provided novel insight in both the virus and host factors involved in the three elementary stages of Varicellovirus infection in primates: primary infection, latency and reactivation.

VZV molecular epidemiology

The molecular epidemiology of varicella zoster virus (VZV) has led to an understanding of virus evolution, spread, and pathogenesis. The availability of over 20 full length genomes has confirmed the existence of at least five virus clades and generated estimates of VZV evolution, with evidence of recombination both past and ongoing. Genotyping by restriction enzyme analysis (REA) and single nucleotide polymorphisms (SNP) has proven that the virus causing varicella is identical to that which later reactivates as zoster in an individual. Moreover, these methods have shown that reinfection, which is mostly asymptomatic, may also occur and the second virus may establish latency and reactivate. VZV is the only human herpesvirus that is spread by the respiratory route. Genotyping methods, together with epidemiological data and modeling, have provided insights into global differences in the transmission patterns of this ubiquitous virus.

Pathway of viral spread in herpes zoster: detection of the protein encoded by open reading frame 63 of varicella-zoster virus in biopsy specimens

Reactivation of varicella-zoster virus (VZV) in the dorsal root or trigeminal ganglia causes herpes zoster. The pathway of viral spread from the ganglia to the skin and also within the skin is not yet completely understood. Histological studies have revealed that each skin lesion in herpes zoster progresses sequentially through the stages of erythema, vesicles, pustules and finally ulceration. An immunohistochemical study of the early skin lesions of herpes zoster demonstrated a high incidence of hair follicle involvement and the main localization of the virus at the isthmus. This evidence suggests that VZV initially spreads from the ganglia through myelinated nerves, which predominantly end around the isthmus of hair follicles. To further investigate the viral spread within the skin, we analyzed the sequential appearance of the immediate early proteins encoded by ORF 63 of VZV (IE63), using an anti-IE63 antibody raised by immunization of rabbits with a recombinant protein. This antibody could detect IE63 in a western blot analysis of infected cells and also in immunohistochemical analysis of the skin lesions of herpes zoster. IE63 initially appeared in the nuclei of the follicular epithelial cells and basal or parabasal epidermal cells. Later, the nuclei and cytoplasm of cells in the epidermis and hair follicles became positive. IE63 remained in the virus-infected cells even during their degeneration. When we examined the hair follicles in the early erythematous lesions, cells positive for IE63 were predominantly distributed around the isthmus. In addition, some lymphocytes around the blood vessels were also positive for IE63, but these cells were seldom positive for the structural antigen. Thus, these observations suggest that VZV arriving through myelinated nerves infects not only permissive cells, but also non-permissive cells in the involved skin of herpes zoster.

Varicella zoster viral disease

Herpes zoster is cause of considerable morbidity, especially among elderly patients, with a suggestion of a slight increase in incidence among female patients. Substantial research on the biology of the varicella zoster virus has led to advances in our knowledge of the pathophysiology of the disease along with more successful therapy for the acute episodes of herpes zoster. Ophthalmic zoster is more common than zoster in other cranial nerves and is associated with pronounced suffering. This article reviews the epidemiology, biology, and latency of herpes zoster, discusses the pathophysiology of the disease, and describes treatment options with antivirals and corticosteroids. The pathophysiology and treatment options for postherpetic neuralgia are also addressed. The varicella vaccine is now available, and initial results suggest that this may lessen the effect of herpes zoster in the future.

Hair follicle involvement in herpes zoster: pathway of viral spread from ganglia to skin

Herpes zoster is caused by reactivation of varicella-zoster virus (VZV) persisting in dorsal root or trigeminal ganglia. To clarify the pathway of viral spread from the ganglia to skin, 16 biopsy specimens of early skin lesions of herpes zoster obtained from the face and trunk of 13 patients were studied histologically and immunohistochemically using monoclonal antibodies to the structural proteins of VZV. VZV-infected cells were detected in the hair follicles in 10 of the 16 specimens and in the epidermis in 2 specimens. Infected cells were localized in the isthmus of every involved follicle (12/12), frequently in the stem (8/10) and infundibulum (6/10), and never in the bulb. The high frequency of follicular involvement in herpes zoster suggests that VZV spreads to the area of skin innervated by myelinated nerves, which end around the isthmus of hair follicles and sebaceous glands.

Natural history and treatment of varicella-zoster in high-risk populations

Rigorous clinical trials have established that both acyclovir and vidarabine favourably alter the clinical course of herpes zoster and chicken-pox in immunocompromised patients. In one comparative study, acyclovir was shown to be superior to vidarabine for zoster in bone marrow transplant recipients. These data, plus the fact that acyclovir is easier to administer than vidarabine, and perhaps less toxic, have made intravenous acyclovir the recognized drug of choice for treatment of herpes zoster in immunocompromised patients. Acyclovir sodium sterile powder received Federal Drug Administration (FDA) approval for this indication in 1990 in the United States. Since complications of zoster occur in only a minority of immunocompromised patients, most physicians would prefer to initiate therapy with an orally-administered drug and avoid the cost and inconvenience of hospitalization. Future studies will compare the efficacy and safety of orally administered bromovinyl arabinosyl uracil and acyclovir in treatment of varicella-zoster virus infections.

Immunological diagnosis in viral infections of the central nervous system: course of antibody titres against homo- and heterologous viruses

In clinical cases suspected for viral encephalitis or meningoencephalitis, the estimation of virus-specific antibodies especially in liquor requires high sensitivity as well as specificity. With enzyme immunoassays the sensitivity in detecting antibodies has increased compared to e.g., complement fixation tests. This report concerns the determination of virus-specific antibodies with a commercial enzyme-linked immunosorbent assay (ELISA) in paired liquor/serum samples of four patients with encephalitis or meningoencephalitis. Up to six virus-specific antibodies of the IgG and IgM classes have been determined [herpes simplex virus (HSV), varicella-zoster virus (VZV), cytomegalovirus, mumps virus, measles virus, and rubella virus]. Additionally, serum samples from several patients suffering, or recovered from, diseases caused by HSV and VZV without CNS involvement have been included as controls. The results showed that besides the virus-specific antibody development (IgG and IgM) against the leading virus, i.e., principally concerned in the disease manifestation assumed to be primarily causing the disease, virus-specific antibodies of the IgG and IgM class against a heterologous virus (e.g., VZV) could also be measured with substantial titers. "Cross-reacting" antibodies to both HSV and VZV with the ELISA only appeared and were present in cases where the infection mainly affected the CNS: no such immunological "cross-reactivity" was observed in serum of individuals in "clinically silent" stages of both HSV and VZV infections. The same situation with no measurable "cross-reacting" antibodies was found in cases of acute HSV or VZV diseases where the CNS was not involved. These findings have been discussed with respect to the findings of common antigens, especially between HSV and VZV, and with respect to an unspecific stimulation of immunocompetent cells.

Strain variation of R5 direct repeats in the right-hand portion of the long unique segment of varicella-zoster virus DNA

We located a region of interstrain size variability in a short segment in an area at the right-hand end of the long unique sequence of the varicella-zoster viral genome. Varicella-zoster virus strains isolated in a district of Japan were classified into three groups on the basis of the size of this segment. Sequence comparison of the variable segment among strains from different groups revealed that the tandem direct repeat, R5, in the segment was variable among strains. R5, which was first discovered in a European strain (Dumas), contained a direct duplication of 88-base-pair (bp) elements separated by a 24-bp element (A.J. Davison and J.E. Scott, J. Gen. Virol. 67:1759-1816, 1986). We found that one 88-bp element and one 24-bp element constitute a repeating unit whose copy number varied from one to three among strains. The simplest R5 we detected was similar to that of Dumas, but there were a few base mismatches between these two R5 structures.

Variation of R1 repeated sequence present in open reading frame 11 of varicella-zoster virus strains

We molecularly cloned the tandem direct reiteration (R1) present in open reading frame (ORF) 11 from three independent strains of varicella-zoster virus. Comparison of the R1 sequences among varicella-zoster virus strains revealed that, although the portion of R1 near the 5' terminus of ORF 11 was conserved among strains, the 3'-terminal portion varied remarkably. This variation was due to the different arrangement of two elements (A and B) and a segment produced by fusion of A and B and to a single-base change in the A element. Since the difference in the size of R1 among strains was a multiple of 3 base pairs, the variation in R1 caused no frame shift in ORF 11.

Detection of human cytomegalovirus DNA in human tonsillar lymphocytes

The human cytomegalovirus (HCMV) was first isolated in cell cultures from the oropharynx, which is thought to be a site of primary infection. Although HCMV can be recovered from the oropharynx during reactivation phases, its exact site of latency is not known. In the present study we demonstrated evidence suggesting the presence of latent HCMV in this anatomic region--in the palatine tonsils. Samples from 30 tonsils obtained by tonsillectomy were screened for the presence of HCMV. Out of the 30 tonsil donors, 23 were seropositive for HCMV. Three methods were used in attempts to demonstrate HCMV's presence in the tonsils: (1) viral isolation attempts on various cell cultures, (2) immunohistochemical staining--immunoperoxidase method--designed to detect viral antigens, and (3) DNA dot hybridization with a HCMV-DNA probe designed to detect viral DNA. Neither infectious HCMV nor other viruses were isolated in cell cultures. No viral antigens were detected by immunoperoxidase staining in the tonsillar tissue. Four out of the 30 tonsils studied were found to contain viral DNA. In one case in which the tonsillar mononuclear (MN) fraction was separated from the polymorphonuclear (PMN) fraction, only the first fraction contained the viral DNA.

Genome variation among varicella-zoster virus isolates derived from different individuals and from the same individuals

We have used 12 restriction enzymes to analyse the DNA of 24 clinical isolates of VZV derived from 12 patients in order to compare isolates derived from different individuals and derived serially from the same individual. As reported previously, only a small proportion of the isolates differed with respect to the presence or absence of restriction sites. However, we found that the size of DNA fragments generated from all the isolates derived from different patients varied in any of four regions, one of which was first recognized in this study. In one case, where multiple isolates recovered from the same individual were analysed, each was distinguished from the others not only by differences in the variable regions but also by the presence or absence of a restriction site in a nonvariable region. This suggests that multiple strains of VZV can be present in the same human host.

Varicella-zoster virus infection of human sensory neurons

Primary varicella-zoster virus (VZV) infection of humans may result in latent infection of sensory neurons in the peripheral nervous system. To examine the interaction of VZV with the sensory neuron we infected immunochemically defined human neurons with cell-associated VZV. Utilizing double-label immunofluorescence technology, a VZV-specific glycoprotein and a nonglycosylated phosphoprotein were detected in human fetus dorsal root ganglion (DRG) neurons, as defined by the presence of the neuron-specific enolase isoenzyme and the A2B5 ganglioside antigen, respectively. In addition to VZV antigen expression, progressive virus-induced cytopathic damage (neuronal enlargement and nuclear granulation of a fraction of the neuron population) was evident. As determined by transmission electron microscopy, VZV-infected human fetus DRG neurons contained empty and complete nucleocapsids with numerous pleomorphic virus particles in the cytoplasm, often in association with vacuoles. Although virus-specific antigen expression, particle synthesis, and cytopathic effects were observed in the human neuron population, neurons were less susceptible to VZV-induced cytopathic damage than supporting nonneuronal cells, suggesting neuronal modulation of VZV infection in vitro. This system provides the first model to examine the neuron- and virus-specific gene(s) and gene product(s) pertinent to the interaction of VZV with the human neuron.

Epstein-Barr virus replication in oropharyngeal epithelial cells

Despite the well-established tropism of the Epstein-Barr virus (EBV) for human B lymphocytes, the cell type within the oropharynx capable of allowing EBV replication has never been conclusively identified. Using in situ cytohybridization, we demonstrated EBV DNA in oropharyngeal epithelial cells from 10 of 12 patients with infectious mononucleosis. In duplicates of specimens found to contain cell-associated EBV DNA, we detected EBV RNA in two of four samples, using a biotin-labeled EBV DNA probe, thereby confirming the intracellular location of the viral genome. In 20 of 28 throat washings analyzed, cytohybridization results and assays for cell-free infectious virus were in agreement. In seven of the eight remaining specimens, cytohybridization identified intracellular EBV DNA in the absence of detectable extracellular virus. We conclude that the oropharyngeal epithelial cell may be the target cell type that is productively infected in infectious mononucleosis.

Comparison of human cytomegalovirus growth in MRC-5 human fibroblasts, brain, and choroid plexus cells in vitro

Cell cultures derived from human brain, choroid plexus, and human lung fibroblasts (MRC-5) were infected with the Towne strain of human cytomegalovirus (CMV). The cytopathic effect, beginning 24-48 hours after infection, was characterized by foci of enlarged rounded cells that spread slowly and eventually coalesced to destroy the entire monolayer within one week. Cowdry type A inclusion bodies and herpes virus nucleocapsids were seen in infected cells. CMV-specific antigen was demonstrated by immunofluorescence in fibroblasts and astrocytic cells of brain cultures and in cells of choroid plexus cultures as well as in MRC-5 fibroblasts. Despite these morphologic and immunochemical similarities the growth of CMV differed in cells of brain and choroid plexus origin as compared with MRC-5 cells. In brain and choroid plexus cell cultures most of the virus remained cell-associated throughout the observation period of one week, whereas in MRC-5 cells the CMV was found in both cell-associated and cell-free fractions of harvested material.

Herpes zoster mandibularis

A case of Herpes zoster of the mandibular nerve in an 80 year old man is reported. Herpes zoster infection affecting only the mandibular branch of the trigeminal nerve is rare. It is proposed that the distribution of lesions is a consequence of a unique latent site of the virus within neurons of the gasserian ganglion innervating the mandibular division of the trigeminal nerve.

Viral vaccines under development: a third generation

In summary then, my purpose has been two-fold: on the one hand, I have tried to highlight the kinds of basic science advances in both cellular and virologic research that can (and should) be focussed both on vaccines under development and, retrospectively, on those whose origins were strictly empiric. On the other hand, I have attempted a partial survey of some of the prominent members of a potential new generation of vaccines to point out areas where these advances can and should contribute either to progress or to a sense of caution about the further reliance on pure empiricism. It is clear that we are not finished with new viral vaccines. It is equally clear that narrowing the persistent gap between basic science and its application to public health needs will require much energy and attention as vaccine development progresses.

Herpes zoster at the NIH: a 20 year experience

One hundred and seven cases of herpes zoster in a hospitalized population with a variety of illnesses during a 20 year period were reviewed. Zoster occurred throughout the year, without seasonal predominance, and was most frequent in lymphoproliferative malignancy. In the majority, lesions were confined to the skin in one or more adjacent dermatomes (localized zoster) and were most frequent in the thoracic region. In 15 per cent of the cases, cutaneous dissemination of the lesions developed; this occurred four to 11 days after the onset of dermatomal lesions, and in one-third of these there was central nervous system involvement. Dissemination of zoster, however, directly resulted in only one death. Predisposing factors for zoster included local irradiation and, occasionally, surgery in subsequently involved areas. There were trends for more frequent splenectomies in patients with Hodgkin's disease in whom zoster subsequently developed, and for more frequent corticosteroid therapy in patiens with disseminated zoster. Advanced stage of Hodgkin's disease, in itself, was not associated with development of zoster, and the onset of zoster did not herald a poor prognosis for the underlying disease. Herpes zoster was, thus, largely a source of increased morbidity rather than mortality in the population studied, and multiple factors appeared to predispose to the development of zoster in this group of patients.

Varicella-zoster virus infection of human brain cells and ganglion cells in tissue culture

The growth of varicella-zoster virus (VZV) in cultures of human brain (HB) and human ganglion (HG) cells was compared to VZV growth in human fibroblasts. Infected cultures were monitored by histologic, electron microscopic (EM), and virologic techniques. Two to three days after VZV infection of all cell cultures at a multiplicity of infection (MOI) of 0.1, a multifocal cytopathic effect (CPE) developed. CPE was characterized by multinucleated cells and virus-specific intranuclear inclusions as determined by immunofluorescence and EM. In VZV- infected HB and HG cells only, large vacuoles were also seen in the cytoplasm of dying cells. Some vacuoles were almost devoid of structures. Within and at the limiting membranes of other vacuoles, aggregates of VZV particles (measuring 210--230 nm) were seen enveloped in osmiophilic material. VZV infection of HB and HG cultures was strongly cell-associated. Clarified tissue culture medium removed at maximum CPE failed to infect homologous HB or HG cells. When an inoculum of VZV-infected HB or HG cells was transferred to homologous uninfected cultures for 10--15 passages, the incubation period for CPE remained constant, and the titer of VZV in cells sampled randomly corresponded to the amount of virus that was used for original infection.

Pathology of the human spinal ganglia in varicella-zoster virus infection

Spinal ganglia from a patient who died on the 6th day of varicella infection were examined by immunofluorescence and electron microscopy, and were compared with spinal ganglia from a patient dying on the 17th day of herpes zoster infection. In herpes zoster, typical intranuclear inclusion bodies were found in neurons, satellite cells and fibroblast-like cells of the ganglia, which contained numerous naked virus particles. In varicella, few changes were found by light microscopy but viral antigen was detected in a few neurons and satellite cells by immunofluorescence. Electron microscopy revealed scattered virus particles near the nuclear membrane of a neuron, satellite cells and capsular cells and enveloped particles in the cytoplasm of satellite cells. The particles in the nuclei were mostly naked virions with specific crescent-like inner-nuclear structure; those in the cytoplasm had complete and incomplete envelopes and showed pleomorphism. A "virus-like" intranuclear filament found in mononuclear cells in herpes zoster and a "plexiform vermicellar array" found in the nuclei of neurons in varicella are at present considered to be non-specific nuclear changes caused probably by viral infections.