The research done in the Slow Viruses Unit concerns the pathogenesis of several persistent viral infections of animals and of man. Three of the viruses under study are neurotropic: Theiler's virus causes a chronic neurological disease of mouse which is a model for multiple sclerosis. Persistent infection of neurons by Borna disease virus causes behavioral disturbances in rats that are reminiscent of human mood disorders. HTLV-I, a human retrovirus, is the agent of tropical spastic paraparesis. We are also using anti-measles live virus vaccines as vectors to immunize against SIV and HIV.
1-Pathogenesis of the infection of mouse by Theiler's virus (Jean-Francois Bureau, Michel Brahic).
The disease caused by Theiler's virus, a mouse picornavirus, is biphasic. The first phase corresponds to the infection of neurons in brain and spinal cord. This is followed by a persistent infection of glial cells in the white matter of spinal cord with chronic inflammation and primary demyelination. These lesions resemble active plaques of multiple sclerosis closely. All mouse strains are susceptible to the early phase of infection. However, the infection persists only in genetically susceptible strains.
For several years, the group headed by Jean-Francois Bureau has been characterizing genes making mice susceptible to the persistence of the infection. The Tmevp1 locus, located in the MHC region, has a major effect on susceptibility. The group showed that all the properties of this locus are those of a Class I gene. The H-2Db gene brings resistance in a dominant fashion through a cytotoxic response directed at an immunodominant epitope, whereas the H-2Kb gene has no role in resistance. We showed that the peptide binding grove of the Class I molecule bears the determinants of H-2Db restricted resistance.
SJL/J mice are very susceptible to persistent infection whereas B10.S mice are resistant. Both bear the same H-2s haplotype. Susceptibility loci have been mapped by screening the genome of crosses between these two strains and by constructing congenic mouse strains. Two susceptibility loci, Tmevp2 and Tmevp3 have been localized on chromosome 10, on either side of the interferon gamma gene. The Tmevp3 region has been sequenced and a candidate gene is being characterized.
Lethal irradiation and bone marrow reconstitution have been performed between the histocompatible SJL/J and B10.S strains. Following reconstitution, mice acquired the phenotype of the bone marrow donor. Therefore, although it is not MHC linked, hematopoietic cells determine the difference of susceptibility between the SJL/J and B10.S strains.
We mapped two loci that are important for pathogenesis. One, on chromosome 14, determines the extent of demyelination independently of the viral load. The other one, on chromosome 11, controls the severity of clinical disease in animals with a high viral load.
The virus persists mainly in CNS macrophages. Virus replication is restricted in these cells. We studied virus replication in primary cultures of bone marrow macrophages from SJL/J mice. The cells express large amounts of viral antigens during the first 20-30 hrs that follow inoculation. Thereafter, viral replication becomes restricted. This restriction is linked to the production of type I interferon.
2- Pathogenesis of the infection of new born rats by Borna disease virus (Daniel Gonzale-Dunia)
Borna disease virus, a non segmented negative strand RNA virus with a wide host range, is responsible for persistent CNS infections with behavioral abnormalities. Sero-epidemiological and molecular data suggest that BDV can infect humans and could be implicated in the etiology of certain psychiatric diseases.
Infection of newborn Lewis rat with BDV leads to persistent CNS infection without inflammation and a selective loss of defined neuronal populations particularly in the hippocampus and cerebellum. We studied, at different times post inoculation, the expression level in CNS of growth associated protein (GAP)-43 and synaptophysin, two molecules involved in synaptic density and plasticity. We observed a progressive decrease in the expression of these markers in specific neuronal populations. These changes could play an important role in the cognitive impairment associated with BDV infection.
Since BDV is non-cytolytic, the mechanisms underlying its neurotoxicity are not well understood. One hypothesis is that the infection interferes with the response of neurons to neurotrophins, proteins that are instrumental in the survival of neurons and the outgrowth of their processes. We showed that PC12 cells infected with BDV become resistant to NGF-induced differentiation. This resistance is linked to a downregulation of the expression of the NGF receptor at the cell surface, and also to defects in the MEK/ERK signaling cascade triggered by NGF.
We extended these observations to primary cultures of rat hippocampal neurons. We observed that hippocampal neurons were highly susceptible to BDV replication and spread, although the infection was non-cytopathic and did not cause overt morphological changes to the cells. However, we found that infection selectively blocked the expression of molecules involved in neuroplasticity, such as GAP-43, synapsin, VAMP-2 and synaptophysin. Moreover, BDV-infected neurons responded only weakly to neurotrophins such as BDNF and NT3, both in term of ERK signaling cascade activation and of BDNF-induced synaptic remodeling.
There are only very few agents that inhibit BDV. We observed that the mitotic inhibitors 1-ß-D-arabinofuranosylcytosine (Ara-C) was a very effective inhibitor of BDV in persistently infected Vero cells as well as in primary cultures of neurons. We found that Ara-C inhibited BDV RNA and protein synthesis to some extent. More strikingly, however, the expression of viral proteins was almost exclusively restricted to the nucleus in Ara-C-treated cells. The nuclear sequestration of viral nucleocapsid abolished cell-to-cell spread of BDV almost completely. The inhibitory effect of Ara-C appears to be specific for BDV, since two other negative strand RNA viruses, measles and influenza viruses, were not sensitive to the drug.
3- Pathogenesis of HTLV-I infection (Frédéric Tangy).
HTLV-I is responsible, in a small number of seropositive individuals, for a chronic demyelinating neurological diseasenamed tropical spastic paraparesis. The pathogenesis of this disease is poorly understood, in part because of the lack of a good animal model. We constructed a chimeric HTLV-I virus in which the envelope protein has been replaced by that of ecotropic Moloney murine leukemia virus. This virus has acquired tropism for murine cells and lost its tropism for human cells. It replicates, albeit at low levels, in murine cells. We hope that it will allow us to obtain a murine model of HTLV-I infection.
4- Anti-retroviral vaccines (Frédéric Tangy).
Live attenuated measles vaccine strains are widely used in human medicine. It is now possible to turn them into vectors for the expression of foreign genes. Pilot experiments with a measles virus expressing E. Coli beta-galactosidase showed that inoculated saimiri monkeys raised antibody and CTL responses against the bacterial protein.
Recombinant measles viruses expressing various genes from HTLV-I, SIV and HIV have been constructed. The recombinant viruses express their respective transgenes at high level. They are stable and grow normally in tissue culture. Saimiri monkeys inoculated with recombinant measles virus expressing the gag, env or tax gene of HTLV-I raised good antibody responses against the foreign antigen.