Unit: Neuro-Immuno Virology

Director: Monique Lafon

Our group is studying the interactions of neurotropic viruses with the nervous system to identify molecular basis of neuron survival. An in vivo approach using several model of transgenic or KO mice, consists in studying the mechanisms developed by pathogenic strains of rabies virus to subvert the host surveillance (strengthen of the immune privilege of the nervous system). An in vitro approach consists in listing and subsequently characterizing the neuron genes utilized by pathogenic virus to subvert the T cell response and to promote intrinsic survival. This is performed by microarrays using human post mitotic neurons infected with herpes Simplex virus and rabies virus of different pathogenicity and by analysing virus gene that control survival and pahogenicity.

Upregulation of FasL in the CNS: a mechanism of immune system evasion by rabies virus (Leïla Baloul )

After an injection of mice in the hindlimbs, the highly pathogenic strain of rabies virus CVS invades both spinal cord and brain and causes the death of mice. In contrast, the CNS invasion by the strain of attenuated pathogenicity PV is restricted to the spinal cord and mice survive with paralytic sequels. We found previously that T cells cannot control the neuroinvasive CVS rabies infection. Our data indicate that despite a similar expression of virus and TNF-αmRNAs, the migration of lymphocytes is transitory in fatal rabies, whereas it is sustained in non-fatal rabies. Transitory migration of T cells in fatal rabies, is associated with an increase of cell apoptosis. We found that only fatal rabies upregulates the early expression of FasL mRNAs. FasL is expressed by infected neurons. In mice lacking FasL, (gld) the infection by the neuroinvasive rabies virus strain is less severe, and the number of CD3 T cells undergoing apoptosis is reduced compared to those observed in their normal counterparts. These data strongly support that fatal rabies virus triggers early FasL expression that leads to the destruction of migratory T cells by the Fas/FasL apoptosis pathway. Thus it seems that rabies virus uses an immunosubversive strategy to successfully invade the CNS which takes advantage of the immune privilege status of the CNS. This could explain why without post-exposure vaccination a rabies virus infection is fatal in most cases (Lafon, 2004).

Rabies virus triggers both caspase-dependent and independent apoptosis in Jurkat T cells (Maria-Isabel, Thoulouze, Mireille Lafage)

Strains of live rabies virus (RV) vaccine strains triggers of caspase-dependent apoptosis in the human lymphoblastoid Jurkat T cell line (Jurkat-vect). This process involved caspases, including caspases 3, 8 and 9. Treatment with the pan-caspase inhibitor ZVAD-fmk reduced, but did not abolish, apoptosis, suggesting that a caspase-independent pathway is also induced by RV infection. Indeed, translocation of the apoptosis-inducing factor (AIF) was detected in RV-infected Jurkat T cells. In contrast, strain of RV that does not induce apoptosis did not activate caspases nor translocate AIF. Bcl-2 overproduction in Jurkat T cells (Jurkat-Bcl-2) abolished both caspase activation and AIF translocation, suggesting that Bcl-2 overproduction blocks apoptosis by interacting upstream in both the caspase-dependent and the caspase-independent apoptotic pathway. RV infection and production were similar in Jurkat-vect and in Jurkat-Bcl-2 cells. This indicates that 1) caspase activation is not a prerequisite for RV maturation or budding, and 2), in contrast to what was observed with HIV or Influenza virus infection, Bcl-2 has no direct antiviral effects against rabies virus. Bcl-2 production is naturally upregulated by day 3 in RV-infected Jurkat-vect cultures. This seems to be controlled by the virus infection itself. The Bcl-2 increase results in the establishment of long-term, persistently infected cultures that continue to produce virus particles. Thus, during the course of infection by live RV vaccine strain, Bcl-2-rescued infected cells may therefore be productive reservoirs of virus in the long term (Thoulouze et al, 2003a, 2003b).

Rabies virus pathogenicity inversely correlates with apoptosis induction (Leïla Baloul, Maria-Isabel Thoulouze, Mireille Lafage)

We reported that non-neurotropic rabies virus (RV) strains, currently used to immunize wildlife against rabies, induces not only a caspase-dependent apoptosis in the human lymphoblastoid Jurkat T cell line (Jurkat-vect) but also a caspase-independent pathway. In contrast, strain of neurotropic RV did not induce apoptosis of the neurons they infect in the nervous system, or in vitro in neuroblastoma cell lines. The inverse correlation of the induction of apoptosis and the capacity of a virus strain to invade the brain suggests that blockage of apoptosis could be a strategy selected by neurotropic virus to favor their progression through the nervous system.

(Baloul and Lafon, 2003, Thoulouze et al, 1997, 2003a, 2003b).

Pro-apoptotic property of glycoprotein of non-pathogenic rabies virus strain (ERA) (Stéphanie Lay, Christophe Préhaud, collaboration with Bernhard Dietzschold, Jefferson Institute, Philadelphia, PA, USA)

We showed that the attenuated live rabies virus (RV) vaccine strain ERA triggers apoptosis in the human neuroblastoma SK-NSH cell line along with activation of caspase, nuclear fragmentation and phosphatidylserine exposure.In contrast, infection, with the pathogenic and neurotropic CVS RV strain did not cause apoptosis in Jurkat T cells. Patterns of viral proteins in the cytoplasma and at the membranes are different in CVS and in ERA infection (photo II).To identify which of the two major RV proteins (G and N) is responsible for the triggering of apoptosis, both G and N of ERA were expressed individually in Jurkat T cells by using the inducible Tet-ON expression system. Induction of RV G expression but not RV N expression resulted in apoptosis indicating that the capacity of a particular RV to trigger apoptosis is largely determined by determinants of G protein. To further examine qualitative aspects of RV G that are associated with the induction of apoptosis and to determine whether observations made in Jurkat T cells take place in neurons, we infected human neuronal cells with recombinant RVs in which G from an attenuated strain was replaced by the G from a virulent strain. This experiment revealed that only recombinant RVs containing the G of a non-pathogenic RV strain, but not of a pathogenic strain, are able to trigger apoptosis in neuronal cells. It shows that neither the transcription, nor the replication of the virus can induce major signaling pathways driving the cells to programmed cell death. This suggests that apoptosis is induced by determinants which are only present in the G proteins of non-pathogenic attenuated RV strains. These determinants could be, at least in part, responsible for the unique ability of attenuated RV strains to induce protective immunity. Moreover, the distribution of cytoplasmic membrane-bound G was very different in ERA- and CVS-infected cells. In ERA-infected cells, the G protein was observed in localized areas of the cytoplasmic membrane, where it formed a fairly large ribbon-like protein structure. On the contrary, G-CVS was distributed on the cytoplasmic membrane as bright patches. This distribution is not modified after blockage of apoptosis (ZVAD-treatment), indicating the large-ribbon-like viral protein accumulation does not result of caspase activation. Fluorescent recovery after photobleaching (FRAP) experiments indicate that membrane fluidity is modified, suggesting that perturbation of membrane stability may be a key element in triggering neuronal apoptosis.

(Prehaud et al, 2003, Lay et al, 2003).

Apoptotic bodies as powerful immunogens (, Françoise Mégret, Christophe Préhaud in collaboration with Christophe Batejat and Nicolas Escriou from Unité de Génétique Moléculaire des Virus Respiratoires, Institut Pasteur, Paris, France

Intracellular pathogens, such as viruses or intracellular bacteria, often induce apoptosis of the cells they infect. The demise of the infected cells results in the formation of apoptotic bodies which contain cytoplasm along with microbial antigens. Several lines of evidence indicate that these structures are powerful immunogens that stimulate T lymphocytes very efficiently, after their uptake and degradation by antigen presenting cells (APCs: dendritic cells and macrophages) and presentation of microbial antigens via MHC class I or class II molecules. Our goal was to use the property of G-ERA to trigger the formation of apoptotic bodies from cells infected by a live vaccine to improve vaccine immunogenicity. The ability of the G-ERA protein to increase a B cell response was tested using vaccinia viruses vectorized to express either influenza hemagglutinin protein, (Flu-HA) or rabies virus glycoproteins. Vaccine preparations, made of mouse lymphoblastoid cell lines (EL4 cells) infected with several combinations of recombinant vaccinia virus expressing either G-ERA (pro-apoptotic protein), G-CVS (a non-apoptotic protein as a negative control) or Flu-HA, were injected to C57Bl mice. We showed that co-expression of G-ERA and Flu-HA in apoptotic bodies , is the best engineered vaccine for stimulating an humoral anti-HA response. We although established that the stimulation is apoptosis-mediated, since necrosis was not detected and blockage of apoptosis (ZVAD-treatment) abolishes the immunopotentiation. These data indicate that the fragmentation by apoptosis of cells infected by a live vaccine is an advantageous device to improve the immunogenicity of a live vaccine.

This work opens new ways for the improvement of vaccine with low immunogenicity potential but also for the design of new generation of vaccine for infectious diseases.

Patent filed by Pasteur Institute (WO 03/048198-A2)

Micro-arrays of post-mitotic human neurons infected with HSV-1 and rabies virus (Christophe Prehaud, Françoise Megret)

The complex interaction between host and pathogen can be explored by using microarrays. By following the pattern of gene expression at different times, it is possible to elucidate which host genes are up- or downregulated over the course of infection. Identification of genes that are differentially regulated and the characterization of their functions provide a promising window on the understanding of pathogenicity. Moreover when virus promotes neuronal survival, as rabies virus does, transcriptome analysis can also lead to the identification of neuronal survival genes. Therefore we have undertaken a transcriptome analysis of human neurons, NT2-N infected with HSV-1 or rabies virus (pathogenic and attenuated strains) in collaboration with the Genopole of Louis Pasteur University in Strasbourg. Human neurons were infected either with HSV-1 (18h) or rabies viruses (24h) and total RNAs were extracted. RNA quality control was assessed on Agilent biochip RNA 6000 nano. RT-PCRs were undertaken in order to detect viral amplicons. RNAs were labelled, hybridized on the Affymetrix human microarrays HG U133A and HG U133B and the chips were scanned. Data were collected, normalized and mined by using Micro DB and Data Mining Tool softwares. Comparative analysis identified a limited set of candidate genes. These genes may be key elements in controlling pathogenicity and neuron survival.

Photos :

I: Culture of post-mitotic human neurons (NT2-N)

II:Immunopotentiatioin of apopoptotic bodies. Mouse lymphoblastoid cell (EL-4) expressing a vaccinal antigen, (the hemagglutinin of influenza virus inn red) encouters apoptosis (fragmentation of nucleus in blue) due to the expression of the pro-apoptotic G-ERA.


Keywords: neurons, apoptosis, NT-2N, rabies virus , HSV-1, micro-arrays, AIF Neurology, virology, immunology

Activity Reports 2003 - Institut Pasteur

Page Top research Institut Pasteur homepage

If you have problems with this Web page, please write to rescom@pasteur.fr