Pathogenesis of HSV-1/2 induced vaginitis/vulvitis of the mouse: dependence of lesions on genetic properties of the virus and analysis of pathohistology

Summary A scoring system for herpes simplex virus (HSV) induced vaginitis/vulvitis in Balb/c mice was delineated from vaginal infections. Four degrees of vaginitis/vulvitis could be distinguished after infection with suitable strains of HSV despite nearly identical replication rates. The time course of replication, inflammation and pathohistology was compared further.Grade 0 was defined by lack of symptoms despite presence of strong replication, which was detectable at days 3–6. Focal necrotic lesions of the epithelial layer were present containing HSV-specific antigens. DNA could be detected by hybridization only in the outer zone of these areas. At day 6 these zones began to be re-epithelialized. In the vaginal lumen abundant detached epithelial cells and granulocytes were already present by day 2. Grade 1 was macroscopically characterized by a slight inflammation commencing on days 5–6. Replication and antigens in the epithelium were found on days 2–6. HSV-antigens were only detected above the basal membrane, and some infiltration with granulocytes and lymphocytes was observed below the basal membrane at day 4. Grade 2 showed strong redness and inflammation as well as hyperemia. Cellular infiltrates were present in the large antigen containing epithelial lesions and below the basal membrane. From day 4 on, neurons were HSV-antigen and DNA positive and macrophages in the stroma contained antigen. The vulva was also shown to be involved. Grade 3 exhibited prolonged severe hyperemia, and destruction of the epithelium and the stroma with necrosis and infiltration, especially of the vulva.This grading system was shown to depend on certain unknown genetic properties of HSV-strains. Neither thymidine-kinase activity, replication in macrophages, fusion activity of strains nor presence or absence of the Hpa I P-fragment were shown to be of importance for severity of vaginitis/vulvitis. Vaginitis/vulvitis was shown to be an all or none response to HSV independent of the rate of replication. The set of virus genes responsible for neuroinvasiveness after vaginal or i.p. inoculation was found to be different. The time course of replication (mainly days 3–6) and inflammation (days 5–10) indicates that inflammation seems to be a secondary immunological phenomenon induced later by the replication phase of HSV. Our system could be useful for separately testing drugs with antiviral and anti-inflammatory properties.In part presented at the 16th International Herpes Virus Workshop, July 7–12, 1991, Asilomar, U.S.A.     

Persistent Theiler's murine encephalomyelitis virus infection in mice depends on plaque size

Theiler's murine encephalomyelitis virus (TMEV) is an enteric pathogen of mice which causes acute and chronic neurological disorders in the natural host. When brain-derived stocks of TMEV isolates are adapted to cell culture they predominantly form either large or small plaques. In this study the type of central nervous system (CNS) infection (acute versus chronic) and the associated disease occurring in mice inoculated intracerebrally with large and small plaque strains of TMEV was investigated. Large and small plaque strains of TMEV were found to vary in virulence, type of neurological disease produced and ability to establish persistent CNS infection in mice. Two large plaque strains, GDVII and FA viruses, were highly virulent, produced acute encephalitis, but were cleared from the nervous systems of surviving animals. Therefore, it appears that these large plaque variants do not cause persistent CNS infection in mice. In contrast, five small plaque strains, DA, WW, TO4, Yale and BeAn8386 viruses, were relatively avirulent, usually produced no illness during the first month after inoculation, but readily established persistent CNS infection in mice. Persistently infected mice later developed demyelinating disease. Having identified strains of TMEV that differ regarding their ability to persist, we now hope to be able to exploit this difference in elucidating the basic mechanism(s) of TMEV persistence.     

Vaginal infection of mice with HSV type 2 variant ER-: a new animal model for human primary genital HSV type 2 infections.

Studying the pathogenesis of vaginal infections in mice with two variants of Herpes simplex virus type 2 (HSV-2) strain ER we observed that both variants ER+ and ER- caused severe vaginitis but only ER+ invaded the CNS leading to lethal neurological disease. In contrast, mice infected with ER- cleared the virus from the vagina and recovered from infection. ER+ and ER- expressed equal levels of thymidine kinase (TK) indicating a TK-independent difference in neurovirulence. Using the non-neurovirulent variant ER-, we were able to investigate humoral immune responses later after infection. Vaginal infection with ER- suppressed serum antibody formation after a secondary systemic HSV-1 infection. Fresh isolates of HSV-1 and HSV-2 caused uniformly a lethal neurological disease after vaginal inoculation of mice. However, some animals survived an intraperitoneal infection with these isolates. Infection with HSV-1 isolates stimulated a strong antibody production, whereas infection with HSV-2 isolates suppressed antibody formation, thus supporting earlier results from our group obtained with laboratory strains. Since suppression of antibody formation could be demonstrated with clinical HSV-2 isolates and likewise after vaginal infection with HSV-2 variant ER- we consider this phenomenon to be of relevance in human genital HSV-2 infections. Vaginal infection of mice with variant ER- represents a new model for primary genital HSV-2 infections; this model could be useful for histopathological, virological, immunological and drug testing studies.     

Central nervous system and genital infection with reactivation of latent herpes simplex virus type 2 in mice

To establish a reactivation model of genital and central nervous system infection, 3- to 12-week-old outbred or BALB/c mice were inoculated vaginally with the HG-52 strain of herpes simplex virus type 2 (HSV-2). Primary infection was confirmed by serially positive vaginal cultures. Mortality in 6- and 12-week old infected mice was about 20%. In survivors, clearance of infectious virus was confirmed in serially negative vaginal cultures. At 6 weeks, immunosuppression of survivors with cyclophosphamide and antilymphocyte serum was begun. Recurrent virus shedding, monitored by daily vaginal cultures, was detected in the majority of animals. All mice became moribund or died, usually during the third to fifth weeks of immunosuppression. Brains and spinal cords from which all sensory ganglia had been removed were homogenized and inoculated onto cultures. One or both central nervous system (CNS) samples were virus-positive in nearly half of these mice, and cell-free virus was isolated from most positive brain and cord supernatants tested. Three-fourths of mice had evidence of virus reactivation with immunosuppression, as indicated by vaginal or CNS isolations, and by failure to isolate virus by identical means in matched infected, non-immunosuppressed controls. Vaginal, spinal cord and brain isolates occurred independently of one another in many immunosuppressed mice, and could not be predicted from presence or absence of external genital lesions during primary infection. These experiments show that with immunosuppression, reactivations of latent HSV-2 infections in mice can be detected in the genital tract and CNS, and provide a model to study productive, recurrent CNS infection and disease.     

Asymptomatic vaginal herpes simplex virus infections in mice: virology and pathohistology

Summary One of the causes of genital tract infections in humans are herpes simplex virus types 1 and 2 (HSV-1, HSV-2). Although primary and recurrent infections can be clinically apparent and in part very serious, many infections are asymptomatic and result only in temporary genital shedding of virus (recurrences). During our investigations of vaginitis, strain IES of HSV-1 produced an asymptomatic infection. Replication in the murine vaginal (vag.) epithelium as well as antibody formation after vag. infection was comparable to those of survivors after infection with highly virulent strains. Titration of liver, spleen, ovaries, adrenal glands, spinal cord, or brain after vag. IES infection revealed no virus, whereas after i.p. infection virus could be demonstrated in many organs examined. Histological examination with a DNA probe (in situ hybridisation), HSV antibodies (immunohistochemistry), and haematoxylin and eosin (HE) staining showed only small focal HSV lesions of the vaginal epithelium in early stages of the infection, never exceeding to the subepithelial tissue. Severe infiltrations and ulcerations after infection with highly virulent strains (17syn⁺, ER^–) could never be demonstrated after IES vag. infection. Identical replication rates of both groups of HSV despite much greater areas of epithelial necrosis with the virulent strains may be explained by the large number of virus inactivating granulocytes induced by the virulent strains, thus inactivating the hypothetical higher virus load.     

Protection of mice against lethal challenge with herpes simplex virus by vaccination with an adenovirus vector expressing HSV glycoprotein B

Increasing attention has been focused on the use of recombinant mammalian viruses as potential vaccines. Recombinant human adenoviruses are one of the more promising vaccine vectors because they can be easily constructed and because live adenovirus vaccines have been administered orally to large numbers of military recruits without adverse reactions. In order to examine the efficacy of human adenoviruses as vaccines we have studied the immunity induced by a recombinant adenovirus vector, AdgB2, which induces high level expression of herpes simplex virus (HSV) glycoprotein B (gB) in human and murine cells. Mice inoculated with AdgB2 produced antibodies specific for gB which neutralized HSV in the presence of complement. Although mice inoculated with AdgB2 showed no ill-effects after AdgB2 inoculation and we were unable to detect replication of human adenoviruses in mice, the mice were protected from a lethal challenge with HSV after a single inoculation with AdgB2.     

Herpes simplex virus infections in male and female mice following pinna inoculation: responses to primary infection and artificially induced recurrent disease

The pathogenesis of herpes simplex virus (HSV) infection is now well understood. After humans or experimental animals recover from primary infection, the virus remains latent in sensory ganglia of the peripheral nervous system. Latency, for the most part, remains unresolved, and elucidating the mechanisms involved with latency has proved difficult. The requirement for unravelling HSV latency is the availability of a reproducible animal model. Previously, the mouse ear model has been extensively characterized; however, many studies using this model have involved female mice only despite evidence that recurrent HSV infection in humans may vary by gender. We inoculated male and female mice subcutaneously in the pinna with varying amounts of one of four strains of HSV and monitored the mice for signs of primary infection. Following recovery from primary infection, mice were induced to develop recurrent disease. In addition, we attempted to isolate virus from dorsal root ganglia of mice suspected of harboring latent virus. There were no differences in the response of male and female mice to either primary infection or artificially induced recurrent disease when inoculated with the same virus. Differences were noted when female mice were inoculated with different strains of virus.     

Impact of mutations in the hemagglutinin gene of H5N1 influenza virus on antigenicity and virulence as revealed by site-specific mutagenesis

In our earlier studies, we have shown that amino acid changes in the hemagglutinin (HA) of influenza H5N1 virus escape mutants conferring resistance to monoclonal antibodies (MAbs) may correlate with a decrease of virus virulence for mice and that the virulence can be restored to the initial level by serial passages. In the present study, the mutations identical to those observed in the HA of a low-virulent escape mutant and its readapted variant were introduced into the HA gene by site-specific mutagenesis. The viruses produced by plasmid transfection and containing the HA gene either of A/Vietnam/1203/2004 (H5N1) virus with a deletion at the cleavage site, or of a low-virulent escape mutants, or of its readapted variant, in the presence of 6 genome segments of A/Puerto Rico/8/34 (H1N1) virus and the NA gene of A/Vietnam/1203/2004 (H5N1) virus, were assayed for virulence. Determination of virulence for mice indicated that amino acid substitution in the HA gene of a low-virulent escape mutant produced a decrease of virulence whereas the additional mutation identical to that acquired by the escape mutant in the course of readaptation restored the virulence to initial level. The findings are the first strong evidence for lower H5N1 virus virulence resulting from the amino acid substitution changing the antigenic specificity of HA and for restored virulence arising from compensating mutation in the HA gene.     

Neurovirulence of herpes simplex virus types 1 and 2 isolates in diseases of the central nervous system

Herpes simplex virus (HSV) isolates derived from the central nervous system of ten patients with HSV-1-induced encephalitis, one patient with multiple sclerosis, and 14 patients with HSV-2-induced meningitis were investigated for neurovirulence by assaying the LD₅₀ after nose and intracerebral (i.c.) inoculation of mice. HSV-1 encephalitis strains were significantly more virulent after nose inoculation (i.e. neuroinvasive) when compared with HSV-1 isolates from patients with oral lesions only, whereas HSV-2 meningitis strains were significantly more virulent after i.c. inoculation when compared with HSV-2 isolates from patients with genital lesions only. No correlation between high neurovirulence (defined as low LD₅₀ for both routes of infection) and replication in cell cultures of neuronal and non-neuronal cell lines was found, but the weakly neurovirulent HSV-1 strain isolated from a patient with multiple sclerosis gave low replication yields. After nose inoculation, a highly neuroinvasive HSV-1 laboratory reference strain replicated to high titers in nose tissue, the trigeminal ganglia and brainstem, while a strain with low neuroinvasiveness but high i.c. virulence replicated less well in the brainstem. Neuroinvasiveness of the virus strain might be one factor of relevance in the pathogenesis of HSV-1 encephalitis in man.     

Gene expression in mice infected with West Nile virus strains of different neurovirulence

West Nile virus causes febrile illness in humans with a proportion of cases progressing to meningoencephalitis, encephalitis, hepatitis, and death. Isolates of the virus fall into two genetic lineages, with differences in neuroinvasiveness for mice occurring between strains within both lineages. We used DNA microarrays to compare gene expression in mice infected peripherally with seven lineage 1 and 2 strains confirmed to be of either high or low neuroinvasiveness in mice and associated with severe or benign infection in humans and birds. The 4 strains with highest neuroinvasiveness induced increased expression of 47 genes in the brain, 111 genes in the liver, and 70 genes in the spleen, relative to the 3 least neuroinvasive strains. Genes involved in interferon signaling pathways, protein degradation, T-cell recruitment, MHC class I and II antigen presentation, and apoptosis were identified that may have both pathogenic and protective effects, but increased expression of certain acute proteins, central nervous system specific proteins and proteins associated with T-cell hepatitis, implicate mechanisms related to exalted virulence.     

Pathogenesis of genital herpes simplex virus infection in mice

This study was undertaken to establish the role of virulence of various herpes simplex virus (HSV) strains in the course of infection when applying the virus to the non-injured mucous membranes of mice.Wild-type HSV-type 1 (HSV-1) strains with marked differences in their neurovirulence following intracerebral inoculation showed minor differences in virulence after vaginal inoculation, but essentially their neurovirulence in cerebral infection corresponded to their virulence on the mucous membranes.In comparison with the wild-types, however, there were pronounced differences among syn- and TK^–-mutants of HSV-1 and HSV-2 in the degree of virulence at different sites in the course of virus infection. Whereas syn-mutants proved avirulent on the mucous membranes but not in neural tissues, TK^–-mutants were avirulent both on mucous membranes and in neural tissues.Ts-mutants of HSV-2 were not found to establish themselves when administered to the non-injured mucous membranes, nor did they induce neutralizing antibodies, but a later challenge with the wild-type virus at the same site lead only to an attenuated course of infection.Supported by the Deutsche Forschungsgemeinschaft Schn 174/6-2     

CMV gB genotypes and outcome of vertical transmission: study on dried blood spots of congenitally infected babies.

BACKGROUND: The role of the virulence of the infecting cytomegalovirus (CMV) strain in the transmission of the virus from mother to fetus and the outcome of the fetal infection has not received much attention yet. Molecular analysis of the gene coding for the surface glycoprotein B (gB) has been used to investigate the relationship between genotype and virulence in groups of immunosuppressed patients. OBJECTIVES: (1) to assess the prevalence of different gB genotypes in babies with congenital CMV infection; (2) to investigate the possible relationship between genotype and severity of congenital CMV disease; (3) to evaluate the possibility of using dried blood on Guthrie cards (DBS) for genotyping. STUDY DESIGN: CMV DNA was extracted from DBS and from urine/saliva samples collected in the first two weeks of life of 98 congenitally infected babies, half of which were symptomatic at birth. Genotyping was performed through RFLP analysis of the region corresponding to the cleavage site of the gB protein. RESULTS: The most prevalent genotype was gB1 (42%) followed by gB3 (26%), gB2 (19%) and gB4 (13%). Rates of disease and CNS damages were higher among children infected by gB1 (35%, 17%) and gB3 (31%, 28%) than in those infected by gB2 and gB4 (20%, 17% and 13%, 15%, respectively). These differences however did not reach the statistical significance. The parallel typing of DBS and urine/saliva strains gave a full concordance of results. CONCLUSIONS: All four major CMV gB genotypes (gB1-4) can cause a congenital infection but none seems to be associated to the development and the severity of disease. The possibility of using the neonatal DBS for genotyping opens a way to the examination of large numbers of cases of congenital CMV infection.     

Herpes simplex virus type 1 DNA sequences which direct spread of virus from cornea to central nervous system

The virulence of a herpes simplex virus (HSV) intertypic recombinant possessing HSV-1 DNA sequences from map units 0.31 to 0.44 and HSV-2 sequences from map units 0 to 0.30 and 0.45 to 1.0 were compared with the virulence of the two parental strains. Following ocular inoculation, both the intertypic recombinant and the HSV-1 parent replicated at the infection site and spread to the peripheral and central nervous system (CNS) to produce fatal encephalitis. The HSV-2 parent also replicated at the infection site but failed to progress to the CNS. However, when inoculated intracerebrally, the HSV-2 strain was as lethal as the HSV-1 parent. Furthermore, the HSV-2 strain could produce thymidine kinase at 37 and 39 degrees in levels comparable to the HSV-1 strain. The results indicate that transfer of the HSV-1 DNA sequences imparted to the recombinant virus the necessary genetic information to spread from the cornea into the central nervous system.     

Herpes simplex virus type 1 DNA sequences which direct spread of virus from comes to central nervous system

The virulence of a herpes simplex virus (HSV) intertypic recombinant possessing HSV-1 DNA sequences from map units 0.31 to 0.44 and HSV-2 sequences from map units 0 to 0.30 and 0.45 to 1.0 were compared with the virulence of the two parental strains. Following ocular inoculation, both the intertypic recombinant and the HSV-1 parent replicated at the infection site and spread to the peripheral and central nervous system (CNS) to produce fatal encephalitis. The HSV-2 parent also replicated at the infection site but failed to progress to the CNS. However, when inoculated intracerebrally, the HSV-2 strain was as lethal as the HSV-1 parent. Furthermore, the HSV-2 strain could produce thymidine kinase at 37 and 39° in levels comparable to the HSV-1 strain. The results indicate that transfer of the HSV-1 DNA sequences imparted to the recombinant virus the necessary genetic information to spread from the cornea into the central nervous system.     

Relationship of the virulence of Trichomonas vaginalis and the major histocompatibility complex in murine trichomonad infection

 A biological assay based upon the induction of abscesses in mice injected subcutaneously with Trichomonas vaginalis was shown to be a valid method for comparing the virulence of two isolates of T. vaginalis cultured from two patients, one suffering from severe vaginitis and the other exhibiting only mild disease. The data showed excellent correlation between the physical dimensions of abscesses in mice injected with each trichomonad isolate and the severity of vaginitis produced in the women from whom the isolates were obtained. The assay employed in our study incorporated measurement of the mean abscess volumes from day 1 to day 6 post-inoculation with T. vaginalis. We found that the abscess assay was clearly superior to a murine intraperitoneal assay for virulence evaluation of trichomonad isolates. We then used the murine abscess assay to determine the susceptibility of different strains of mice to infection with a virulent T. vaginalis isolate so as to test whether the genetic constitution of the host would influence the pathogenesis of the disease. BALB/c (H-2^d) mice were susceptible to infection with T. vaginalis, but both CBA/CaH (H-2^k) and BALB/c-H-2^k mice were shown to be resistant. The quantitation of abscess formation in these inbred and congeneic resistant mouse strains demonstrates that the severity of infection with T. vaginalis is governed by genes mapping within the major histocompatibility complex. Received: 16 October, 1995 / Accepted: 15 March, 1996     

Recombinants are isolated at high frequency following in vivo mixed ocular infection with two avirulent herpes simplex virus type 1 strains

Summary Mixed infections with different strains of herpes simplex virus type 1 (HSV-1) may result in more severe disease than infection with either strain alone. This phenomenon is important because it may facilitate the identification of virulence genes through the transfer of virulence determinants between complementing strains, and it may pose a problem in the use of attenuated HSV strains for vaccines and gene delivery vectors. In this study, we have compared the percentage of recombinants present after mixed infection with HSV-1 strains OD4 and 994 in vitro and in vivo. After corneal inoculation, we found that 74% of randomly picked isolates from the trigeminal ganglia were recombinants, compared with 59% from the cornea. Twenty-six percent of randomly picked isolates were recombinant following mixed infection of Vero cells in vitro. Seventeen recombinant strains isolated from the in vivo mixed infections were assayed for ocular virulence, and they were found to exhibit a wide range of virulence phenotypes. The presence of virulent recombinants suggests that recombination plays a role in the increased disease observed in this mixed infection, and the broad range of virulence indicates that there may be multiple genetic factors involved in the increased virulence observed after mixed infection with these two strains. The recombinants were also tested for their ability to grow in NIH 3T3 fibroblasts, and though some correlation was observed between growth in vitro and ability to cause ocular disease, improved growth in murine cells does not sufficiently explain the increased virulence observed in some recombinants.     

Pathogenesis of the slow disease of the central nervous system associated with wild mouse virus. III. Role of input virus and MCF recombinants in disease

Inbred mice bearing the FV-1n marker when inoculated at birth with an ecotropic murine leukemia virus (WM 1504-E) obtained from wild mice develop a progressive central nervous system (CNS) disease manifested by hindlimb paralysis and incoordination that begins by 6 to 7 weeks of age. Studies of infected SWR/J mice at 3 days to 4 months of age indicated the following: (1) Expression of MuLV gp69/70 and p30 antigens in CNS rises beginning as early as 3 days after inoculation and increases with time. (2) Neuronal damage is evident by Day 7, and neuronal lesions develop in all mice by Day 14. (3) WM 1504-E virus can be isolated from CNS tissue by 48 hr after initiating infection. (4) Upon passage into susceptible newborn mice, the WM 1504-E isolates cause neuronal disease. (5) "Dual-tropic" mink cell focus forming (MCF) -like virus is found in splenic but not CNS tissues by 8 weeks after initiating infection. (6) MCF viruses that arise by env gene recombination between WM 1504-E and endogenous xenotropic viruses do not cause CNS disease upon inoculation into susceptible newborn mice. Similarly inoculated WM 1504-amphotropic virus (WM 1504-A) does not cause CNS disease (7). Results in SWR/J mice can be duplicated in C3H/St and C57Br/cdj mice. These observations define the wild mouse ecotropic virus as the sole infectious agent responsible and sufficient for the development of this neurologic disease. Evidently, the disease from this "slow virus infection" begins early in life shortly after introduction of the infectious agent, but becomes clinically apparent only as neuronal destruction accumulates over the lifetime of the host.     

Immunogenicity of purified glycoprotein gB of herpes simplex virus

Summary The efficacy of a herpes simplex virus (HSV) component vaccine consisting of viral glycoprotein gB was examined in a mouse system. Immunization of mice with HSV type 1 (HSV-1) gB emulsified in Freund's complete adjuvant or with HSV-1 gB adsorbed to aluminum gel was fully protective against subsequent challenge with HSV-1 or HSV type 2. Latent infection in the trigeminal ganglion was also prevented by immunization with gB.With 6 Figures     

Apparent stability of herpes simplex virus genomes isolated from primary and recurrent infections and dorsal root ganglia in the mouse

Following natural or experimental primary infection, herpes simplex virus (HSV) becomes latent in sensory ganglia. Reactivation of latent virus may lead to recurrent disease. If HSV DNA remains stable during primary, recurrent and latent infections, that stability would enable us to trace the transmission of HSV from one individual to another. We inoculated mice in the ear pinna with HSV and collected virus at various intervals during primary infection. In mice surviving primary infections, recurrent disease was induced from which virus was isolated. Virus was also recovered from explanted dorsal root ganglia. Virus isolates were characterised by restriction endonuclease digestion and compared with the original inoculate(s). The data indicate that in all cases except two, the isolates from primary and recurrent infections remained identical to the original inoculates.     

Nucleotide sequences responsible for the inability of a herpes simplex virus type 2 strain to grow in human lymphocytes are identical to those responsible for its inability to grow in mouse tissues following ocular infection

A study was undertaken to determine whether genes associated with herpes simplex virus (HSV) neuroinvasiveness in mice influence the growth of HSV in man, the virus's natural host. HSV-2(186), a nonneuroinvasive HSV strain, was found to replicate poorly (less than 3-fold) in cultures of phytohemagglutinin (PHA) stimulated human peripheral blood mononuclear cells (PBMC). In contrast, seven other HSV strains all multiplied 40- to 100-fold. The paucity of HSV-2(186) growth in PBMC was not due to a failure of this strain to grow in primary human cells because high titers (greater than 10(8) PFU/ml) were obtained following infection of human foreskin fibroblasts. The genetic basis for the deficient growth was analyzed by marker rescue experiments. Recombinant HSV-2 strains were generated in marker rescue experiments utilizing HSV-2(186) DNA and plasmids containing a cloned DNA polymerase gene isolated from a neuroinvasive HSV strain possessing the capacity to replicate in human PBMC. Progeny which rescued DNA from the cloned HSV DNA polymerase gene replicated 40- to 100-fold in PHA-stimulated PBMC. Moreover, unlike the HSV-2(186) parent, HSV-2(186) isolates possessing rescued DNA grew well in the eye, trigeminal ganglion, and brain of mice and induced fatal encephalitis. The results indicate that nucleotide sequences responsible for increasing the capacity of HSV-2(186) to grow in PBMC of man are identical to those responsible for increasing the capacity of this strain to grow in mouse tissues and to spread from the eye to the brain.