A. Randrup Thomsen

Lymphocytic Choriomeningitis Virus Infection is Associated with Long-Standing Perturbation of LFA-1 Expression on CD8+ T Cells

Flow cytometric analysis of splenocytes from mice infected with lymphocytic Choriomeningitis virus revealed marked and long‐standing up‐regulation of LFA‐1 expression on CD8+, but not on CD4+ T cells. Appearance of CD8+ T cells with a changed expression of adhesion molecules reflected polyclonal activation and expansion which was demonstrated not to depend on CD4+ T cells or their products. Cell sorting experiments defined virus‐specific CTL to be included in this population (LFA‐1hiMEL‐14lo), but since about 80% of splenic CD8+ T cells have a changed phenotype, extensive bystander activation must take place; this is indicated also by the finding that CD8+LFA‐lhi cells transiently express several markers of cellular activation, e. g. transferrin receptor, IL‐2Rα and β. Analysis of cells from the cerebrospinal fluid of mice infected intracerebrally showed that virtually all T cells present belonged to the CD8LFA‐lhi subset and, correspondingly, the ligand ICAM‐1 was found to be up‐regulated on endothelial cells in the inflamed meninges. Preincubation of LCMV‐primed donor splenocytes with anti‐LFA‐1 markedly inhibited the transfer of virus‐specific delayed‐type hypersensitivity to naive recipients. Together, these findings indicate that up‐regulation of LFA‐1 expression is a critical factor involved in directing activated CD8+ T cells to sites of viral infection.

Sensitization to Lipopolysaccharide in Mice with Asymptomatic Viral Infection: Role of T Cell-Dependent Production of Interferon-γ

The interplay between viral infection and lipopolysaccharide (LPS) was studied. Infection with a noncytopathogenic virus, lymphocytic choriomeningitis virus (LCMV), was found to sensitize mice to low doses of LPS. In vivo, this hypersensitivity correlated with hyperproduction of tumor necrosis factor-alpha (TNF-alpha), and in vitro, LPS-stimulated splenic adherent cells produced increased amounts of TNF-alpha. Hyperproduction of TNF-alpha was temporally correlated with virus-induced production of interferon-gamma (IFN-gamma); only marginally increased IFN-gamma and TNF-alpha production was observed in LCMV-infected, T cell-deficient mice and in mice infected with vesicular stomatitis virus, a virus that induces much less T cell activation than does LCMV. Finally, LCMV infection was much less efficient in priming IFN-gamma knockout mice for hyperproduction of TNF-alpha. These findings indicate that clinically silent viral infections may induce hypersensitivity to LPS through T cell activation and subsequent production of IFN-gamma; this sensitizes monocytes/macrophages for hyperproduction of TNF-alpha.

Fatal Meningitis following Lymphocytic Choriomeningitis Virus Infection Reflects Delayed‐Type Hypersensitivity Rather than Cytotoxicity

Fatal meningitis following intracerebral inoculation of lymphocytic choriomeningitis virus (LCMV) reflects an immunopathological lesion believed to be mediated by cytotoxic T cells. The results presented here demonstrate that pretreatment with cyclophosphamide (Cy; 150mg/kg body weight) 2 days before intracerebral infection significantly reduced the lethality of the infection. However, this treatment did not impair the antiviral cytotoxic response as measured in the spleen. On the other hand, virus‐specific delayed‐type hyper‐sensitivity (DTH) was significantly reduced. This reduction seems to be the result of a Cy‐induced lack of non‐committed ancillary cells since: (1) virus‐primed spleen cells from Cy‐pretreated donors conferred normal LCMV‐specific DTH to naive recipients; (2) transfer of virus‐primed spleen cells from untreated donors did not increase the suppressed DTH response of the Cy‐pretreated mice: and (3) inoculation of irrelevant antigen and antigen‐primed spleen cells into the footpads of Cy‐pretreated, infected mice resulted in a significantly reduced footpad swelling as compared with untreated, infected controls. Taken together, these results indicate that LCMV‐induced meningitis does not solely represent T‐cell‐mediated cytotoxicity in vivo but that a fatal outcome of the infection critically involves not only effector T cells but also ancillary cells.

Virus Elimination in Acute Lymphocytic Choriomeningitis Virus Infection Correlation with Virus-Specific Delayed-Type Hypersensitivity rather than Cytotoxicity

The immunological effector mechanism responsible for the elimination of virus in murine acute non‐fatal extracranial lymphocytic choriomeningitis virus infection was studied. In this infection virus clearance is generally regarded as the result of a direct action of virus‐specific cytotoxic T cells (Tc cells) on virus‐producing target cells in the infected mouse. However, by manipulating the antiviral immune response by pretreatment with various doses of cyclophosphamide, we found lack of correlation between Te‐cell activity and the clearance of virus. In contrast, we observed a conspicuous correlation between the hosts ability to mount a virus‐specific delayed‐type hypersensitivity (DTH) response and its capacity to combat virus. Moreover, pretreatment with silica and carragheenan prolonged viraemia without impairment of the peak Tc‐cell response. These findings indicate that Tc cells have little or no capacity to eliminate virus, at least in the absence of an inflammatory response, and our findings suggest that virus clearance reflects a DTH‐like process.

Inducible nitric-oxide synthase plays a minimal role in lymphocytic choriomeningitis virus-induced, T cell-mediated protective immunity and immunopathology

By using mice with a targetted disruption in the gene encoding inducible nitric-oxide synthase (iNOS), we have studied the role of nitric oxide (NO) in lymphocytic choriomeningitis virus (LCMV)-induced, T cell-mediated protective immunity and immunopathology. The afferent phase of the T cell-mediated immune response was found to be unaltered in iNOS-deficient mice compared with wild-type C57BL/6 mice, and LCMV- induced general immunosuppression was equally pronounced in both strains. In vivo analysis revealed identical kinetics of virus clearance, as well as unaltered clinical severity of systemic LCMV infection in both strains. Concerning the outcome of intracerebral infection, no significant differences were found between iNOS-deficient and wild-type mice in the number or composition of mononuclear cells found in the cerebrospinal fluid on day 6 post-infection. Likewise, NO did not influence the up-regulation of proinflammatory cytokine/chemokine genes significantly, nor did it influence the development of fatal meningitis. However, a reduced virus-specific delayed-type hypersensitivity reaction was observed in iNOS-deficient mice compared with both IFN-gamma-deficient and wild-type mice. This might suggest a role of NO in regulating vascular reactivity in the context of T cell-mediated inflammation. In conclusion, these findings indicate a minimal role for iNOS/NO in the host response to LCMV. Except for a reduced local oedema in the knockout mice, iNOS/NO seems to be redundant in controlling both the afferent and efferent phases of the T cell-mediated immune response to LCMV infection.

Studies on the Role of Mononuclear Phagocytes in Resistance to Acute Lymphocytic Choriomeningitis Virus Infection

The role of mononuclear phagocytes in various phases of the acute lymphocytic Choriomeningitis virus (LCMV) infection was studied. The anti‐macrophage agent carrageenan delayed virus clearance. Carrageenan was most effective when given before virus inoculation, suggesting that it interfered with early events in the host response to the virus. Correspondingly, carrageenan enhanced early virus multiplication. Pretreatment with carrageenan apparently did not inhibit induction of the T‐cell response and had little or no direct effect on T‐cell‐dependent anti‐viral activity. The LCMV‐induced natural killer response was also unimpaired by this treatment. Taken together, these findings suggest that resident macrophages constitute a barrier to the initial multiplication of LCMV. A breakdown of this macrophage barrier results in a more disseminated infection, which the specific immune response has difficulty in eliminating. Adoptive transfer experiments with pre‐irradiated recipients showed that T‐cell‐dependent virus clearance required interaction between donor‐derived primary immune spleen cells and radiosensitive host cells. T cells did not seem to constitute the radiosensitive host component, since athymic (nude) mice functioned well as recipients. Together with previously published data, this finding strongly suggests that T‐cell‐dependent virus clearance involves cooperation between T cells and non‐committed cells, probably monocytes.

Influence of Host Genes on the Outcome of Murine Lymphocytic Choriomeningitis Virus Infection

One of the most important challenges in pathogenesis is to understand the influence of host genes on disease susceptibility. It is an attractive hypothesis to geneticists that the existence of extensive polymorphisms within populations implies that various alleles confer certain selective advantages for survival under particular evolutionary pressures. Thus, it is not surprising that when the extensive polymorphism at HLA loci (the designation for major histocompatibility complex of genes, MHC, in humans) was discovered, geneticists examined a variety of diseases for possible associations with one or more alleles at this complex of loci. Further impetus for these studies was the observation by Lilly and coworkers (1964) that in mice the H-2 complex (the designation for MHC in mice) controlled susceptibility to viral leukemogenesis. Unfortunately, the most striking finding in the majority of ensuing studies on resistance to infectious disease, in both men and mice, was how little disease resistance is influenced by MHC (Brinton and Nathanson, 1981; Clatch et al., 1987; Klein, 1986). A rare exception is the finding of two HLA types frequent among West Africans being associated with resistance to severe malaria (Hill et al., 1991). This is in contrast to the well-known associations between HLA and many autoimmune diseases (Tiwari and Terasaki, 1985). This correlation is quite understandable since until recently there was virtually no selection against noninfectious diseases, probably because people died before the diseases could develop (Klein, 1986). Although this is not the case today, autoimmune diseases still do not appear to exert any selective pressure on the population because most of them become apparent after the age of reproduction. Even at this time, the development of disease is basically a statistical risk. This could be due to multigenic effects (see below) or the need for exposure to a triggering factor (even in monozygous twins there is discordance, Tiwari and Terasaki, 1985). In addition, in almost no case do we know the etiological agent(s) responsible for an HLA-associated disease. As Klein (1986) states, “In fact, at the present stage, the study of HLA-associated diseases has about the same value to medicine as beetle counting had to zoology 200 years ago.”

Concanavalin A-mediated in vitro activation of a secondary cytotoxic T-cell response in virus-primed splenocytes.

In a recent report it was shown that what appeared to be secondary cytotoxic T cells could be obtained from lymphocytic choriomeningitis virus (LCMV)‐primed splenocytes after stimulation in vitro with the non‐specific T cell mitogen concanavalin A (Con A). The present experiments attempt to characterize further these effector cells and, in particular, to establish whether the Con A‐activated cytotoxic effectors are qualitatively different from the secondary cytotoxic T cells induced by restimulation with the homologous antigen. It was found that: (1) in vitro activation with Con A could be obtained with populations harvested between 13 days (the earliest tested) and at least 300 days after priming: (2) cytotoxicity was independent of the presence of carried‐over Con A in the cytotoxicity assay; (3) cytotoxicity was dependent on close association between activated T cells and target cells, since no evidence was found to indicate a role for other cell types or soluble (cytotoxic or arming) factors; (4) cytotoxicity was specific with regard to both virus and ‘self’. By comparison with previous data on LCMV‐induced cytotoxic T cells, it is concluded that Con A induces the generation of cytotoxic T cells from LCMV‐primed splenocytes, which, by the criteria used, are indistinguishable from virus‐induced secondary cytotoxic T cells. The implications of these findings are discussed.