Unit: Biology of Immune Regulations - INSERM E 352
Director: LECLERC Claude
The activity of our laboratory is focused on the understanding of the mechanisms that control the activation and regulation of T cell responses and on the development of new strategies of vaccination. We have recently developed several strategies of activation of CTL responses as well as fully synthetic glycopeptide carrying saccharidic epitopes to induce anti-tumoral immune responses. We are also investigating the biology of dendritic cells, in particular in neonates.
A. The adenylate cyclase toxin (CyaA) of Bordetella pertussis: a new vector targeting dendritic cells (C. Fayolle, L. Mascarell, X. Préville et G. Schlecht, in collaboration with D. Ladant, IP and P. Sebo, Prague)
The adenylate cyclase toxin (CyaA) of Bordetella pertussis is a major virulence factor, which can invade eukaryotic cells. We previously demonstrated that CyaA uses the integrin CD11b/CD18 as a cell receptor. Using CyaA mutants and CyaA fragments, we have identified the region of CyaA interacting with CD11b. These results will help us to design a highly efficient vector capable of targeting a wide range of antigens to CD11b+ antigen presenting cells, leading to a more efficient immune responses.
This new vector was used to develop a therapeutic cancer against HPV associated cervical cancer. In collaboration with BT Pharma, a start-up incubated at the Pasteur Institute, we constructed recombinant CyaA containing either the full sequence or various subfragments from the HPV16-E7 protein. When injected to mice, these HPV16-recombinant CyaAs are able to induce specific Th1 and CTL responses. Furthermore, when injected into mice grafted with HPV16-E7-expressing tumor cells (TC-1), these recombinant proteins were able to provide up to 100% protection. These results pave the way for the testing of this vector in clinical trials.
The Tat protein from HIV-1 was also inserted in recombinant CyaA and its ability to induce specific humoral responses against Tat as well as cellular responses was assessed. Mice immunized with CyaA-E5-Tat developed a Tat specific neutralizing humoral response as well as CD8+ T cell responses. In addition, the induced cellular polarization is of Th1 type.
We also established that CyaA can deliver a CD4+ T cell epitope to MHC class II presentation pathway 100 to 1000 more efficiently than the native protein containing this epitope. This potentiation of MHC-II-restricted presentation is blocked when CyaA does not interact with CD11b.
We have developed a new approach in which CD8+ T cell epitopes are chemically linked to CyaA. We showed that high CTL responses were induced in mice immunized with CyaA bearing CD8+ T cell epitopes from ovalbumin and LCMV. The expected advantage of this novel strategy is its versatility as we can easily couple any defined epitope to CyaA.
CyaA targets CD8α-, CD11c+, CD11b+ DC with a high efficiency, and dendritic cells (DC) are the only APC capable of presenting CyaA to CD8+ T and CD4+ T cells. After the encounter with various stimuli, DC increase the expression of MHC molecules and several co-stimulatory molecules, which are necessary to trigger an immune response and in particular, a CTL response. We thus analyzed if this high immunogenicity of CyaA was linked to its capacity to induce DC maturation. We have shown that in vitro very low doses of wild type CyaA, but not a catalytic inactive mutant, trigger DC maturation as probed by increase in surface levels of maturation markers: MHC I and II, costimulatory molecules CD80, CD86 and IL-2 receptor. In parallel, intravenous injection of CyaA, but not its inactive mutant, triggers splenic DC maturation. We are currently investigating the mechanisms responsible for the high immunogenicity of detoxified CyaA.
B. Analysis of immunological properties of viral pseudo-particles (D. Briard, G. Morón and S. Hervas in collaboration with P. Rueda, Madrid)
Chimeric porcine parvovirus virus-like particles (PPV-VLPs), prepared by self-assembly of the VP2 capsid protein of this virus and carrying heterologous epitopes at its N terminus is an efficient antigen delivery system that elicits strong CD4+ and CD8+ T cells responses specific for the foreign epitopes in the absence of adjuvant. As usually exogenous antigens cannot enter into the MHC Class I pathway, PPV-VLPs represent a very interesting antigen carrier to trigger CTL response. We have shown in mice that DC capture these particles very efficiently in vivo and are the only cells capable to present PPV-VLPs to specific CD8+ hybridoma. DC stimulation by PPV-VLPs induces the expression of CD8α and CD205 and of co-stimulatory molecules. Furthermore, DC from mice injected with PPV-VLPs are capable to secreting IL12p70 and IFNγ. After in vivo capture, PPV-VLPs are found in late endosomes. MHC Class I presentation of PPV-VLPs requires intracellular processing, following a non classical pathway involving macropinocytosis, vacuolar acidification and lysosomal proteases, but also processing by the proteasome complex and translocation of epitopes from the cytosol to the endoplasmic reticulum using TAP molecules. In parallel, we also study the capture of PPV-VLPs by human DC (in collaboration with O. Schwartz) and we seek to identify the PPV-VLPs receptor by binding studies and confocal microscopy.
We also currently analyze the mechanisms of antigen presentation during several schemes of immunization in which two or more immunizations are done using different antigen delivery vectors. These studies are performed using two non replicative delivery vectors developed in our laboratory, CyaA and PPV-VLPs. We study the capture of PPV-VLPs in different lymphoid organs either in naive mice or following immunization with these vectors in order to determine if the immunity against a vector affects its capacity to be delivered to DC.
C. Analysis of CTL and Th responses induced by dendritic cell subpopulations (G. Dadaglio, G. Schlecht and J. Mouries)
It is now generally accepted that fully mature DC are the only professional antigen presenting cells able to induce primary T cell-mediated immune response. Multiple DC subsets have been defined on the basis of cell surface markers expression and function. In mice, two distinct CD11chigh DC subpopulations can be distinguished in the spleen by the expression of the CD8α marker and one CD11clow population corresponding to the plasmacytoid DC. Polarization of Th responses induced by DC subsets could be controlled by several factors such as the cytokine microenvironment and the antigen dose. Th polarization could also depend on the time point of their activation kinetics. To challenge the latter hypothesis, we evaluated the maturation status of freshly purified untreated DC versus in vivo CpG-activated DC following their i.v. injection into naïve mice. Our results show that the efficiency of untreated DC to monitor T cell activation and proliferation was similar to early and late CpG-activated DC. Furthermore, strong peptide OVA specific IFN-γ production was always detected regardless the activation status of the DC injected suggesting that in each group a Th1 response is induced. Altogether these observations demonstrated that purification procedure of DC induce unspecific maturation leading to induction of Th1 response. Furthermore, in our model, polarization of CD4+ T cell response is not driven by the maturation status of the DC.
Plasmacytoid DC (pDC) have recently been identified in mouse lymphoid organs. To assess their in vivo function on adaptive responses, purified immature or activated pDCs were loaded with a synthetic peptide containing a CD8+ T cell epitope and then transferred by i.v. route into naive syngeneic hosts. We show that immature and CpG-activated pDC do not induce specific CTL response nor regulatory activity, suggesting that they cannot prime CD8+ T cells. Nevertheless, CpG-activated pDC were able to recall memory T cells, in contrast to immature pDC. However, after stimulation by heat-inactivated influenza virus, pDC produced IFN-α and induced antigen-specific effector/memory T cells. Thus, pDC differentiate into professional APC in the context of viral infection, demonstrating that they can play a role in both innate and adaptive anti-viral immunity.
D. Ontogeny, functions and regulation of neonatal dendritic cells (R. Lo-Man, CM. Sun and X. Zhang)
We are investigating how the dendritic cell compartment may contribute to the high susceptibility to infections observed in newborns and to biased Th2 responses that are often induced. We recently described the ontogeny of mouse DC and showed that they are fully functional at the level of innate responses and in their capacity to prime T cells in vitro and in vivo. However, if neonatal DC are fully competent, we were able to demonstrate for the first time that B cells can control the capacity of neonatal DC to prime Th1 responses and therefore dampen their immune functions. We are currently investigating the ins and outs of this regulation process.
E. Investigation of mechanisms of anti-mycobacterial immunity (L. Majlessi, M. Roajs and S. Hervas in collaboration with S. Cole and collaborators)
Induction of Th1 responses to mycobacterial antigens is essential in protection against infection with Mycobacterium tuberculosis, the ethiologic agent of tuberculosis. We study the capacity of recombinant CyaA, bearing mycobacterial antigens, as sub-unit vaccine, to induce T-cell responses and to confer protection against infection with M. tuberculosis. The antigens inserted into CyaA for this investigation are proteins known for their strong immunogenicity in both mice and human or a protein we identified as a potent immunogen for CD4+ or CD8+ T-cell subsets in mice. The potential of these vectors as sub-unit vaccine candidates against infection with M. tuberculosis is currently under investigation in our Unit.
In collaboration with the Unit of Génétique Moléculaire Bactérienne, we recently demonstrated that introduction of a virulence-associated chromosomal region of M. tuberculosis into M. bovis BCG markedly modifies the interactions between the host immune system and mycobacteria. Indeed, BCG complemented with this chromosomal region induces preferential recruitments of dendritic cells and of CD4+ or CD8+ activated/effector T cells. In contrast to the parental BCG, this BCG is able to induce in vivo at the level of dendritic cells an inflammation program as it activates production of inflammatory chemokines and cytokines.
Identification of a mycobacterial immunogen with the rare capacity to be recognized by CD8+ CTL of mycobacteria-infected mice and generation of MHC-I-restricted T-cell hybridomas specific for this immunogen enable us to investigate the cellular and molecular mechanisms of the presentation of mycobacterial antigens via MHC-class I pathway for the generation of CD8+ T-cell responses.
F. Elaboration of a fully synthetic immunogen bearing a carbohydrate tumor marker for immunotherapy (R. Lo-Man and E. Dériaud in collaboration with S. Bay and S. Vichier-Guerre from Unité de Chimie Organique)
We have developed multiple antigenic glycopeptides (MAG) based on a lysine core extended with peptidic arms displaying a carbohydrate tumor antigen (Tn antigen). The resulting dendrimeric MAGs were able to induce anti-Tn IgG antibodies in a T cell dependent manner that recognized murine as well as human tumor cell lines that express Tn. In mice, therapeutic vaccination using these MAG provided a 70% survival rate of tumor-bearing mice. Together, these results demonstrate that the MAG represents a safe and highly efficient system to induce anti-carbohydrate antibodies and is a potent alternative strategy to the traditional carbohydrate-protein conjugates, which are developed for vaccine and therapeutic purposes. We recently developed new MAGs potentially active in humans and the immunogenicity of these molecules has been demonstrated in non-human primates.
Keywords: Vaccines, CTL, dendritic cells, cancer, T cells, immunotherapy