Systemic lupus erythematosus (SLE) is an autoimmune disease involving the inflammatory damages of multiple organs. Lupus nephritis (LN) as the manifestation in the kidney occurs in more than 50% of SLE patients and is a major cause of morbidity and mortality. Current treatments consist of immunosuppressants that always lead to compromised immune responses with increased risks of infections as the major side effect. To minimize this side effect, it is crucial to develop new treatments that are more natural and specific. Vitamin A, particularly in the form of its functional metabolite, retinoic acid, has shown some beneficial effects against LN in both lupus-prone mouse models and clinical cases. However, a more systemic evaluation of vitamin A treatment in lupus had not been investigated. In our study, we found paradoxical effects of all-trans-retinoic acid (tRA) on lupus-like disease in MRL/lpr lupus-prone mice. Starting at 6 weeks old when the inflammatory environment had been established in MRL/lpr mice, tRA administration reduced immune cell numbers in the secondary lymphoid organs and improved glomerulonephritis. However, circulating autoantibodies and inflammation in renal tubulointerstitium and other organs were increased. The detrimental effects of tRA were not present in MRL control mice, which didn't have an established inflammatory environment at 6 weeks old as shown in MRL/lpr mice, suggesting that the pro-inflammatory effects of tRA are dependent on the pre-existing inflammatory environment. Therefore, to successfully apply vitamin A-based treatment, it is important to avoid the detrimental effects of tRA on lupus by identifying and then specifically eliminating the critical pro-inflammatory immune cell types in lupus. As treatments usually start after the onset of apparent symptoms in patients at the effector stage of autoimmune responses, targeting the inflammatory contributors at this stage appears to be more practical and critical. Among different types of leukocytes, we chose to focus on dendritic cells (DCs), because they are highly diverse and critical in the immune responses as a bridge between the innate and adaptive immune systems. Plasmacytoid DCs (pDCs) as a candidate target have been demonstrated to be crucial for the initiation of lupus development by producing IFNα. However, we demonstrated that although pDCs produced a large amount of IFNα during disease initiation, those from late-stage lupus mice were found to be defective in producing IFNα, suggesting that pDC-targeted treatments should be performed at the initiation stage. This will depend on the progress in early diagnosis in the future. Besides pDCs, we identified a CD11c+ cell population absent at the early-stage but gradually accumulating at the late-stage in the kidneys of lupus mice. These cells have a phenotype of mature monocyte-derived DCs, with particularly high CX3CR1 expression on the surface. Consistent with their pathogenic cytokine profile, in vivo administration of anti-CX3CR1-saporin conjugates to dysfunction these cells in MRL/lpr mice significantly reduced proteinuria scores. Ex vivo activation of renal-infiltrating CD4+ T cells showed increased survival rate, proliferation and IFN-γ production of activated CD4+ T cells when they were cultured with these renal-infiltrating CD11c+ cells. These results suggest that the renal-infiltrating CD11c+ cells are pathogenic and promote inflammation in the kidney at the later effector stage of lupus by interacting with renal-infiltrating CD4+ T cells. In conclusion, although vitamin A showed anti-inflammatory effects on reducing glomerulonephritis, its use in lupus treatment should be guarded due to the other potential pro-inflammatory effects induced by the pre-existing inflammatory environment. IFNα-producing pDCs and CX3CR1highCD11c+ monocyte-derived DCs could be specific therapeutic targets to reduce the established inflammation at the early stage and late stage of LN, respectively. Therefore, it is worthwhile to further investigate the comprehensive effects of combination therapy on lupus, with vitamin A administration and pDCs-specific depletion at the early stage, and CX3CR1highCD11c+ monocyte-derived DCs-specific depletion at the late stage. / Ph. D. / Systemic lupus erythematosus (SLE) is an autoimmune disease involving the inflammatory damages of multiple organs. Lupus nephritis (LN) as the manifestation in the kidney occurs in more than 50% of SLE patients and is a major cause of morbidity and mortality in this disease. Current treatments consist of immunosuppressants that always lead to compromised immune responses with increased risks of infections as the major side effect. To minimize this side effect, it is crucial to develop new treatments that are more natural and specific. My first project was to determine whether vitamin A as a supplement could ameliorate SLE. It turned out to be effective at attenuating LN, but at the same time the nutrient caused massive inflammation in other peripheral organs such as the brain and lungs. This suggests that we need to be cautious when recommending vitamin A supplementation to lupus patients. In order to identify more specific targets in the treatment of SLE, my second and third projects focused on dendritic cells (DCs) that are essential for lupus pathogenesis. I found that plasmacytoid DCs (pDCs), known to be pathogenic in SLE, were in fact defective at promoting inflammation at the late stage of disease, suggesting that pDCs might not be a good target of intervention. In contrast, monocyte-derived conventional DCs turned out to be highly pathogenic especially for the development of LN and could be a potential therapeutic target. Altogether, my investigations have increased our understanding of the pathogenesis of SLE.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21659号 / 医博第4465号 / 新制||医||1035(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 河本 宏, 教授 三森 経世, 教授 椛島 健治 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

La thérapie par cellules T à récepteur d'antigène chimérique (CAR) a produit d'incroyables réponses cliniques contre plusieurs tumeurs malignes, mais elle laisse une marge de manœuvre pour l'échappement de l'antigène. Une nouvelle approche pour surmonter ce problème consisterait à combiner la capacité des CAR à cibler les tumeurs avec la capacité des cellules dendritiques (CD) à amorcer les cellules T afin de générer une thérapie cellulaire qui favorise la propagation de l'épitope plutôt que la destruction directe de la cible. J'ai donc émis l'hypothèse que les cellules dendritiques exprimant les CAR (CAR-CD) peuvent renforcer l'amorçage des cellules T contre les cibles tumorales afin de produire un contrôle adaptatif des tumeurs médié par les cellules T. En utilisant des CD dérivées de la moelle osseuse murine, j'ai d'abord développé un nouveau protocole pour générer des CD. En ajoutant de l'IFNα aux cultures de DC GM-CSF, j'ai généré des CD qui expriment des niveaux plus élevés de molécules stimulant les cellules T et qui induisent une plus forte prolifération des cellules T CD8+ in-vitro par rapport aux CD générées avec le GM-CSF seul. In vivo, ces CD favorisent des réponses effectrices plus fortes, les cellules T CD8+ résultantes exprimant des niveaux plus élevés de marqueurs effectrices, notamment KLRG1 et TIM3, mais des niveaux plus faibles de molécules inhibitrices, notamment PD-1 et CD73. L'expression d'un CAR dans ces DC leur permet de tuer directement des cibles tumorales et d'acquérir des antigènes tumoraux. Plus important encore, ces CAR-CD sont en synergie avec les cellules T CD8+ pour contrôler les cellules tumorales in-vitro, en améliorant la prolifération et la capacité de destruction des cellules T. Chez la souris, les CAR-CD agissent comme un vaccin in vivo en favorisant la génération de réponses de cellules T spécifiques de la tumeur lorsqu'elles sont injectées par voie intratumorale, ce qui permet d'améliorer le contrôle de la tumeur. / Chimeric Antigen Receptor (CAR) T cell therapy has produced unbelievable clinical responses against several malignancies however, this therapy leaves room for antigen escape. One novel approach to overcome this would be to combine the tumor targeting ability of CARs with the T cell priming capacity of dendritic cells (DCs) to generate a cell therapy that provokes endogenous adaptive immunity through epitope spreading rather than just direct target killing. Therefore, I hypothesized that CAR expressing DCs (CAR-DCs) can enhance T cell priming against tumor targets to produce adaptive T cell mediated tumor control. Using murine bone marrow derived DCs, I first developed a new protocol for generating DCs using IFNα. The addition of IFNα to GM-CSF DC cultures generated DCs that express higher levels of T cell stimulatory molecules and induce stronger CD8+ T cell proliferation in-vitro compared to DCs generated with GM-CSF alone. In-vivo, these DCs promote stronger effector responses with the resulting CD8+ T cells expressing higher levels of effector markers including KLRG1 and TIM3 but lower levels of inhibitory molecules including PD-1 and CD73. The expression of a CAR in these DCs allowed them to directly kill tumor targets and acquire tumor antigens. More importantly, these CAR-DCs synergized with CD8+ T cells to control tumor cells in-vitro, enhancing the proliferation and killing capacity of the T cells. In mice, CAR-DCs act as an in-vivo vaccine promoting the generation of tumor specific T cell responses when injected intratumorally, producing enhanced tumor control.