Antigen Targeting To Different Dendritic Cell Subsets For The Induction Of Anti-Mycobacterial Immunity

One-third of the Earth's population is infected with Mycobacterium tuberculosis, making the pulmonary tuberculosis the most widely spread infectious disease, leading to 1.6 million deaths annually. The only vaccine in use against infection with M. tuberculosis, the live attenuated M. bovis BCG (Bacillus Calmette-Guerin), is not able to protect efficiently against the adult pulmonary tuberculosis in endemic zones. The limited efficiency of BCG in the control of M. tuberculosis infection can be explained by the following hypothesis:(ⅰ) BCG could be attenuated, due to the deletion of genes coding for protective immunogens like the deletion of the "Regions of Differences" (RD). (ⅱ) Despite the strong anti-mycobacterial Thl responses induced by BCG, intracellular M. tuberculosis sequestered inside the granuloma, may be inaccessible to these effector T cells. (ⅲ) Following BCG immunization, induction of regulatory T cells (Treg), in parallel to the induction of anti-mycobacterial Thl responses, could reduce the efficiency of the protective effect against M. tuberculosis infection.With the resurgence of tuberculosis in immuno-compromised individuals and the rapid expansion of multi-drug resistant and extensively drug-resistant tuberculosis, the need of a better rational design of new strategies of anti-tuberculosis vaccines is reinforced. Despite intense research on live attenuated and/or sub-unit anti-tuberculosis vaccines, a very few vaccine candidates display improved protective effect and their efficiency remains limited. Here, we drived the extensive knowledge available on the properties of dendritic cells (DC) and antigen targeting to DC subsets, towards the practical in vivo antigen delivery to DC in anti-tuberculosis vaccination. Indeed, so far, addressing M. tuberculosis-derived protein antigens to DC subset(s) and/or DC surface receptor(s), for the induction of protective anti-mycobacterial immunity, has not been investigated.Despite their shared morphology, abundance in T-cell areas of lymphoid tissues, high MHC-Ⅱexpression and outstanding potential to continuously probe the environment, process and present antigens to T cells, DC are divided into different subsets, according to their ontogenic origin, phenotype, maturation programs and specialized functions. Although the well-established classification of the mouse DC subsets can't be transposed to the human DC populations, plasmacytoid DC, blood-derived lymphoid tissue resident DC, peripheral migratory DC and monocyte-derived inflammatory DC have been distinguished in both mice and humans. Numerous evidences argue that the magnitude of adaptive immune responses, as well as differentiation and specialization of CD4+ T cells into Th1, Th2 or Th17, are dictated by different DC subsets with specialized activities. The mobilization of different DC subsets by their direct in vivo targeting through specific surface markers represents a promising pathway to design well-controlled immunization strategies for the development of preventive and/or therapeutic vaccines.So far, the antigen targeting strategy is limited by the requirement of individual chemical coupling or genetic insertion of each immunogen of interest to mAbs specific to each of the numerous DC surface receptors, candidate for antigen addressing. It is largely admitted that DC translate information from different surface receptors into an activation program that orients the Th cell differentiation. However since all the rules governing functions of DC subsets are not yet understood, it remains difficult to predict which DC subsets/DC surface receptors are the most appropriate to be targeted in order to optimize the protective immunity against a given pathogen. Therefore, comparison of the properties and impacts of various DC subsets on the generation of pathogen-specific adaptive responses may help identify the most adapted DC subset(s), able to tailor the most adapted and protective adaptive immunity. To this end, we have designed a versatile antigen targeting approach, prominent mycobacterial immunogens are genetically fused to streptavidin (SA), the resulted fusion proteins are tetramerized to optimize their high affinity interaction with biotin (biot). Such SA fusion tetramers are then complexed to biot-conjugated mAbs, specific to diverse DC surface receptors. Therefore, in this flexible model, once such antigen-SA fusion proteins are produced, they can be readily carried and delivered to different mouse and human DC subsets by simple use of individual biot-mAbs of a large panel of specificities against DC surface receptors, with expression profiles restricted to given DC subsets. Potent mycobacterial antigens included in this study were selected among highly-conserved, low-molecular weight immunogens belonging to the Early Secreted Antigenic Target,6 kDa (ESAT-6) protein family (ESX) of M. tuberculosis. These proteins, actively secreted by the typeⅦsecretion system of mycobacteria, are known for their marked immunogenicity in mice, guinea pig and in ethnically different human populations, and for their protective potential in animal tuberculosis models. Moreover, the presence of CD4+ and CD8+ effector T cells specific to such proteins is directly correlated to the natural anti-mycobacterial protection in M. tuberculosis-infected humans. Based on this verstaile antigen targeting approach, we aim to identify the most appropriate DC receptor(s) to which the targeting of M. tuberculosis-derived ESX immunogens can induce optimized immune responses with anti-tuberculosis protective potential.PartⅠ. In vitro mAb-mediated M. tuberculosis-derived ESX antigen targeting to DCWe first showed in vitro that this Ab-based antigen targeting system allows delivery of ESX immunogens to various antigen-presenting cell surface receptors, either integrins, C-type lectins DCIR-2, MHC-Ⅱmolecule or CD317 (plasmacytoid DC Ag-1, PDCA-1). Importantly, ESX antigens bound to CDllb, CDllc, DCIR-2, MHC-II on conventional DC or to PDCA-1 on plasmacytoid DC were efficiently endocytosed, most probably due to the cross-linking of the targeted surface receptors by the biot-mAbs, leading to their capping together with the bound ESX-SA cargo. To evaluate the presentation of ESX antigen by targeted APC, we developed MHC-II-restricted T cell hybridomas specific to ESX antigen (TB10.4, ESX-H) from BCG immunized mice. By use of such T-cell hybridomas in in vitro presentation assays that we set up, we demonstrated that ESX antigens delivered to conventional or plasmacytoid dendritic cells or to macropahges, were highly efficiently addressed to MHC-II presentation machinery and processed. The ESX-derived epitopes were then loaded on MHC-II molecules and presented, in a highly sensitive manner, to ESX-specific, MHC-Ⅱ-restricted T-cell hybridoma or the polyclonal splenocytes from M. tuberculosis infected mice.PartⅡ. mAb-based ESX antigen targeting to different DC subsets to induce anti-mycobacterial immunityWe characterized the immunogenicity of several ESX proteins fused to SA (ESX-SA), targeted to different DC surface receptors by complexing them to biot-mAbs specific to:MHC-Ⅱmolecules, CD11b or CD11cβ2 integrins, CD317 (plasmacytoid DC Ag-1, PDCA-1) or C-type lectin receptors of:(ⅰ) mannose receptor family, i.e., CD205 (DEC205), (ⅱ) asialoglycoprotein receptor family, i.e., CD207 (Langerin, Clec4K), or CD209 (DC-specific ICAM3-grabbing non-integrin, DC-SIGN), or (ⅲ) DC immunoreceptor (DCIR) subfamily of asialoglycoprotein receptor, i.e., DCIR-2 (Clec4A). We explored this model to select the most appropriate DC subsets or DC surface receptors to target in anti-tuberculosis vaccination on the basis of:(ⅰ) in vivo capture/processing and presentation of ESX antigens by MHC molecules, (ⅱ) in vivo outcome of the ESX-specific Th1, Th2, Th17 or Treg adaptive responses, (ⅲ) in vivo cross presentation of ESX antigens and cross priming of specific CD8+ T cells, (ⅳ) boost effect of such immunization subsequent to BCG priming.This antigen delivery system was highly specific in vivo, as only the DC subsets, targeted with ESX-SA complexed to selected biot-mAbs displayed ESX antigens at their surface, and when positively sorted ex vivo, were able to present ESX antigens to the ESX-specific, MHC-II-restricted T-cell hybridomas.By use of different biot-mAbs specific toβ2 integrins, diverse C-type lectins or PDCA-1, in the presence of the DC maturation signal, Poly inosinic:Poly cytidylic acid (PolyⅠ:C), we directly compared the in vivo efficiency of ESX targeting to different DC subsets/DC surface receptors in the induction of T-cell responses. Remarkably, a single injection of only 1μg (=50 pmoles)/mouse of ESX-SA complexed to biot-mAbs specific to CDllb, CDllc or CD205, induced specific, intense and highly sensitive lympho-proliferative, Thl and Th17-but not Th2-responses. ESX-SA complexed to biot-mAbs specific to CD207 or PDCA-1 induced less intense and less sensitive, yet still marked Thl responses, while ESX targeting to CD209 failed to induce such responses. Among the C-type lectins evaluated in the present study, CD205 was the most efficient at inducing Thl cells in primary responses to ESX antigens.ESX targeting to DC surface receptors allowed substantial reduction of the effective dose of antigen for immunization without impairment of T-cell immunity, as exemplified by the low dose of 5 pmoles (=0.1μg)/mouse of ESX-SA, complexed to biot-anti-CDllb mAbs which induced highly significant ESX-specific Thl and Th17 responses. Importantly, the facts that:(ⅰ) FcγR -/- and WT mice mounted comparable adaptive immune responses to mAb-mediated ESX targeting to DC and (ⅱ) ESX-SA fusion proteins complexed to biot-control Ig did not induce detectable adaptive immune responses, show that the mechanism responsible of targeting, endocytosis and further antigen presentation does not involve FcR.As priming with live attenuated mycobacteria followed by boosting with subunit vaccines, is of the most promising prophylactic anti-tuberculosis vaccination strategies, we focused particular interest in the boosting potential of ESX antigen targeting to DC subsets. We selected for this study the TB10.4 (Rv0288, ESX-H) antigen, a promising protective ESX antigen with highest interest in development of innovative sub-unit vaccine candidate. In mice primed with BCG and then boosted with TB10.4-SA targeted to CD205, CD207, CD209 or DCIR-2, a comparable boost effect of IFN-γresponses was obtained. The best boost effect at the level of Th17 response was obtained with TB10.4 targeting to CD205, followed by CD207 and PDCA-1. Using the strategy of TB10.4 antigen targeting to DC subsets, among all the conditions evaluated and the DC surface receptors targeted, we only detected efficient TB10.4-specific CD8+ T-cell cross priming in mice which were primed with BCG and then boosted with TB10.4 targeted to CD205.Thus, we have developed a new and versatile approach to target promising mycobacterial immunogens to different DC subsets in order to efficiently initiate or boost anti-mycobacterial CD4+ and CD8+ T-cell immunity. So far, CD205 C-type lectin seems to be the best DC surface marker to induce such T-cell responses.