Cytotoxic T lymphocytes (CTL) are major effector cells in the adaptive immune response against intracellular pathogens and cancers, killing targets with high precision. Precision is achieved through the specificity of the clonally expressed T cell receptor (TCR). TCRs recognise a specific peptide chain loaded into a major-histocompatability complex, triggering signalling, inducing the CTL to attach and kill target cells. Key stages in this attack are the initial conjugation followed by polarisation and docking of the centrosome to the junction of the two cells, the immune synapse (IS). This focuses secretion of the cytolytic components, perforin and granzyme, from modified lysosomes to kill the target cell. My PhD has utilised amino acid substitutions in the target peptide to alter its signal strength and shown this alters the subsequent killing efficiency of a target population. I developed new imaging and analysis techniques to investigate the effect of TCR signal strength at each step of the killing process. I show the first step, conjugation, is reduced for a percentage of cells with dwell times decreasing as TCR signal strength decreased. The next key step of centrosome polarisation and docking at the IS was also impaired for an increasing proportion of cells as TCR signalling reduced. Impaired centrosome docking reduced efficient granule recruitment to the IS, necessary for target killing. Centrosome docking was linked with the TCR-induced intracellular calcium flux, the duration of which increases with the strength of TCR signalling. This demonstrates how the process of CTL killing can be fine-tuned by the quality of antigen.

Target cell lysis is the hallmark of immune effector responses of cytotoxic T lymphocytes (CTL), natural killer (NK) cells, and monocytes. The most commonly used assay to measure target cell lysis is the 51Cr release assay and is considered the ‘gold standard’. However, this assay has many disadvantages that limit its use by most laboratories. Thus, several alternative assays have been developed. Some of these alternative assays are more sensitive, easy to perform and do not use radioactive elements. In this study, four of these assays were evaluated for their ability to detect antigen- specific CTL responses in equine blood. Three long-term equine arteritis virus (EAV) carrier stallions, two vaccinated stallions and one naïve stallion were included in this study. Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood collected of these stallions to be used as effector cells. The PBMCs were stimulated with EAV in vitro for 7-10 days to generate antigen-specific effector cells. The granzyme B assay, the Carboxyfluorescein succinimidyl ester (CFSE)/7-Aminoactinomycin D (7AAD) assay and the Lactate dehydrogenase (LDH) assay were performed using these effector cells and autologous equine dermal cells (isolated from each stallion) as target cells. The first two assays (i.e., granzyme B and CFSE/7AAD assays) were difficult to optimize for this study because they work well with non-adherent targets and require immediate flow cytometry analysis. The LDH assay, however detected CTL lysis in one of the two vaccinated stallions at day 99 post vaccination and no response was detected in PBMCs isolated from carrier stallions and control stallion. Based on these findings, the LDH assay is the most suitable assay since it works well with adherent target cells, it produces quantitative data, and is ideal for high-throughput screening.

Interest in cellular immunotherapy has increased with the recognition of the pivotal role that dendritic cells (DC) play in the adaptive immune system. The preparation of DC to present tumour antigens and subsequent induction of tumour specific T cells have been widely documented. This thesis studied the ability of cord blood (CB) stem cells to differentiate into functional CD34+DC, followed by the optimisation of electroporation of RNA into these cells. Total RNA derived from a leukaemic cell line and a primary human leukaemic sample was electroporated into CD34+DC DC and we were able to generate anti-leukaemic cytotoxic T lymphocytes (CTL). The CTL specifically targeted leukaemia but not normal cells. While the in vitro data showed promising results of the CTL specificity, a NOD-SCID model of human ALL was established to allow the CTL to be tested in vivo. We established a reproducible model of human ALL in NOD-SCID mouse using four primary human ALL samples. The adoptively transferred anti-leukaemic CTL into the ALL bearing NOD-SCID mice showed that ALL engraftment was significantly delayed. However, the addition of total RNA loaded CD34+DC DC did not enhance the in vivo CTL effect. Lastly, by dissecting the CTL response, we found that the polyclonal CTL were targeting survivin, HM1.24 and CT-7 antigens. The CTL clones generated from these polyclonal CTL showed high specificity for leukaemia but not normal cells. In conclusion, these preliminary data support the use of total RNA electroporated CD34+DC as a means of inducing anti-leukaemic CTL, and have demonstrated the efficacy of the CTL in a NOD-SCID model of ALL. This study has also provided insight into the polyclonal CTL response and future studies will likely continue along this path.

No description available.

No description available.

The aim of the present study was to develop a dendritic cell (DC) vaccine encoding hepatitis C virus (HCV) multiple cytotoxic T lymphocyte (CTL) epitopes that can stimulate T cell responses in vitro, and can be used for immunization in vivo. DCs were infected with recombinant replication-defective adenoviruses (Ads) expressing 2 HCV sequences fused with green fluorescent protein (GFP) and FLAG tags. One sequence (sequence 1) contained the HCV CTL epitopes, NS4B 1793-1801 and P7 774-782, as well as the HCV Th epitope, NS3 1248-1261. A second sequence (sequence 2) was the positive epitope control which contained HCV core 35-44, core 132-140 and NS3 1248-1261. The efficiency of infection was detected by flow cytometry and the expression of HCV epitopes in the DCs was confirmed by RT-PCR and western blot analysis. Ad infection significantly enhanced DC maturation and interleukin (IL)-12p70 production, resulting in T cell proliferation and increased interferon-γ secretion. The CTLs stimulated by Ad-infected DCs specifically killed Huh7.5 human hepatoma cells. The recombinant Ad-expressing multiple CTL HCV epitopes effectively infected the DCs in vitro and promoted T cell antiviral immune responses, thereby laying the foundation for the development of anti-HCV DC vaccines.

The aim of the present study was to develop a dendritic cell (DC) vaccine encoding hepatitis C virus (HCV) multiple cytotoxic T lymphocyte (CTL) epitopes that can stimulate T cell responses in vitro, and can be used for immunization in vivo. DCs were infected with recombinant replication-defective adenoviruses (Ads) expressing 2 HCV sequences fused with green fluorescent protein (GFP) and FLAG tags. One sequence (sequence 1) contained the HCV CTL epitopes, NS4B 1793-1801 and P7 774-782, as well as the HCV Th epitope, NS3 1248-1261. A second sequence (sequence 2) was the positive epitope control which contained HCV core 35-44, core 132-140 and NS3 1248-1261. The efficiency of infection was detected by flow cytometry and the expression of HCV epitopes in the DCs was confirmed by RT-PCR and western blot analysis. Ad infection significantly enhanced DC maturation and interleukin (IL)-12p70 production, resulting in T cell proliferation and increased interferon-γ secretion. The CTLs stimulated by Ad-infected DCs specifically killed Huh7.5 human hepatoma cells. The recombinant Ad-expressing multiple CTL HCV epitopes effectively infected the DCs in vitro and promoted T cell antiviral immune responses, thereby laying the foundation for the development of anti-HCV DC vaccines.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21635号 / 医博第4441号 / 新制||医||1034(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 椛島 健治, 教授 岩田 想, 教授 山田 亮 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DGAM

Au cours de la réponse immunitaire antitumorale, l’instabilité génétique des tumeurs combinée à la pression de sélection du système immunitaire peut conduire, via l’immunoediting, à l’émergence de variants tumoraux résistants à la lyse par les effecteurs cytotoxiques. Une meilleure compréhension de ces mécanismes potentiellement impliqués dans la susceptibilité tumorale à la lyse naturelle et/ou spécifique pourrait permettre le développement de stratégies d’immunothérapie intégratives plus efficaces. Dans ce cadre nous avons étudié un modèle de résistance à la lyse spécifique impliquant un remaniement du cytosquelette d’actine (i). Nous avons pu mettre en évidence que l’inhibition conjointe de protéines interagissant avec l’actine (caldesmone, ézrine, radixine et moésine) générait une réduction de la susceptibilité des cellules tumorales à la lyse par les lymphocytes T cytotoxiques (CTLs). Parallèlement, nous avons identifié les microARNs différentiellement exprimés entre le variant résistant et la lignée parentale et étudié leur implication dans la susceptibilité tumorale à la lyse par les CTLs. Dans le but de déterminer le rôle d’une pression de sélection par les cellules tueuses naturelles (NK pour Natural Killer), de l’immunité innée, sur les cellules tumorales et l’émergence de variants résistants, nous avons aussi établi un modèle de coculture continue de cellules tumorales de mélanomes avec des cellules NK (ii). Les cellules tumorales obtenues sont résistantes à la lyse NK (mais toujours sensibles à la lyse spécifique par un clone lymphocytaire T cytotoxique) et établissent moins de contact et de synapse immunologique avec les cellules NK que la lignée parentale. L’analyse transcriptomique a révélé la baisse d'expression de B7-H6 (ligand d'un récepteur activateur des cellules NK) qui contribue partiellement au phénomène de résistance. De nombreux gènes impliqués dans les phénomènes de migration/invasion/adhérence sont également modulés et certaines propriétés cellulaires (croissance en milieu semi-solide, adhérence, migration) semblent refléter l’acquisition d’une agressivité tumorale accrue suite à la coculture. Nous avons finalement analysé l’impact sur la réponse antitumorale de la connexine-43, impliquée dans la formation des jonctions communicantes (GJ pour Gap Junction) (iii). Nous avons montré que sa présence à la synapse entre cellules tumorales et CTL n'exerce aucun impact sur la susceptibilité à la lyse. Néanmoins, les GJs sont impliquées dans l’émergence par stimulation antigénique de lymphocyte T CD8+ spécifique hautement réactif. / During antitumor immune response, cancer cells genetic instability combined with immune system selective pressure may drive to the emergence of tumor variant resistant to lysis by cytotoxic effector cells through a phenomenon called immunoediting. A better understanding of those mechanisms putatively involved in tumor susceptibility to natural and/or specific lysis would enable new integrative and more effective immunotherapeutic strategies. In this context, we studied a model of resistance to specific lysis linked to actin cytoskeleton remodeling (i). We showed that combined inhibition of actin interacting protein (caldesmone, ezrin, radixin and moesin) reduced tumor cells susceptibility to cytotoxic T lymphocytes (CTLs) lysis. Moreover, we identified microRNAs differentially expressed between parental cell line and resistant variant and are currently studying their impact on tumor susceptibility to CTLs lysis. In order to depict the role of innate immunity Natural Killer (NK) cells selective pressure, on tumor cells and on the emergence of resistant variants, we also established a maintained coculture model of melanoma cells with NK cells (ii). Selected cells obtained were resistant to NK cells-mediated lysis (but still susceptible to CTLs-mediated specific lysis) and formed less contact and immune synapse with NK cells than parental cell line. Transcriptomic analysis revealed the reduced expression of B7-H6 (ligand of an NK cells activating receptor) partially contributing to the resistance phenotype. The expression of several genes involved in migration/invasion/adhesion is also modulated and some cell characteristics (cell growth in semi-solid medium, adhesion, migration) tend to reflect the acquisition through coculture of an increased aggressiveness. Finally, we evaluated the impact of connexin-43 (Cx43), involved in the establishment of Gap Junctions (GJs), on antitumor response (iii). We showed that despite localization at the immune synapse between tumor target cell and CTL, Cx43 and GJs do not modulate susceptibility to CTL-mediated specific lysis. Nevertheless, GJs contribute to the emergence of highly reactive specific CD8+ T lymphocytes following antigen stimulation.

The ability of our immune system to mount a response against non-self-antigens legitimates the semi-allogenic fetus as a target for maternal immune attack. Yet, in a normal pregnancy the fetus stays well protected due to the concerted action of several diverse mechanisms which either suppress the fetal allogenicity or spatio-temporally inhibit maternal immune cells’ growth and functions. One such factor which aids in the establishment, progression and maintenance of pregnancy is the 28 kDa dimeric sialylated glycoprotein Glycodelin-A (GdA). Synthesized by the endometrium and decidua, this protein has myriad functions, the most important being that of immunosuppression. GdA is inhibitory to all hematopoietic cells and also induces programmed cell death in activated T cells and monocytes via the intrinsic mitochondrial pathway. In the Introductory chapter of this thesis, details about GdA and the other isoforms of the glycodelin family of proteins have been presented which highlight the involvement of glycodelins in primate pregnancy, with emphasis on GdA and its pleiotropic functions associated with reproduction in females. Activated T-lymphocytes against paternal antigens are found in the uterine compartment and in the maternal circulation throughout pregnancy. Activated CD8+ T-lymphocytes have been reported to pre-dominate the uterine T-lymphocyte population during pregnancy and unlike the CD4+ T cells, are retained until term. Studies show that activated CD8+ T-lymphocytes are necessary for the establishment and progression of early pregnancy. However, how these lymphocytes harbouring cytotoxic activity are regulated at the later stages of pregnancy is poorly defined. We attempted to uncover a possible mechanism of regulation of CTL (cytotoxic T lymphocyte) activity (if any) during primate pregnancy by GdA. In the absence of established human CD8+ T cell lines, we first standardized the generation of CTLs in-vitro from hPBMCs (human peripheral blood mononuclear cells) by alloactivating them with an ovarian carcinoma cell line OVCAR-3 utilized as a mimic of an allograft. The details of the rationale behind using this method for generating CTLs and the alloactivation methodology have been put together in the Chapter 1 of this thesis. The activation of hPBMCs was confirmed by the surface expression of an early activation marker CD69 and tritiated thymidine incorporation. Differentiation of CD8+ T cells into effector cells was confirmed by the upregulation of perforin and granzyme transcripts by real time RT-PCR analysis. Target-cell specific cytolytic activity of the CTLs was assessed by using a cytotoxicity measurement assay- JAM test, details of which also form a part of chapter 1. Having generated effective CTLs in vitro, we tested the effect of GdA on CTL activity. Our findings, on the effect of GdA on CTLs have also been discussed in the Chapter 1. We observed that the cytolytic activity of CTLs was significantly reduced by GdA treatment albeit at a dose three to four times higher than that required for inhibiting CD8+ T cell proliferation, implying that a mechanism of temporal regulation of CTL activity operated at the feto-maternal interface, thereby contributing to the establishment and progression of pregnancy. Interestingly, in our quest to uncover the mechanism of inhibition of CTL activity by GdA, we found that the inhibition of proliferation was comparable in both CD4+ and CD8+ T-lymphocytes at all dosages of GdA, but unlike CD4 + T cells CD8 + T cells were resistant to GdA-induced apoptosis even at high dosage of GdA. Hence we could rule out that the loss of CTL activity upon GdA treatment was due to CD8+ T cell death. Further, we assessed the functional competence of alloactivated CTLs by quantitating the mRNA transcripts of key cytolytic molecules; perforin and granzyme B, in GdA treated alloactivated hPBMCs and found that there was a significant reduction in the mRNA of these cytolytic molecules. Additionally, we also found that GdA treated CD8+ T cells exhibited impaired release of the cytolytic molecules by the process of degranulation, measured by the surface exposure of LAMPs (Lysosome associated membrane proteins) on the surface of cells by flow cytometry and as seen by the retention of perforin protein in them assessed by intracellular staining and flow cytometry. Intrigued by the observations, we probed for the regulators of perforin and granzymes in CTLs. EOMES (Eomesodermin) and T- Bet are well known transcription factors which control the differentiation of CD8+ T cells into effector and memory cell CD8+ T cell type. Interestingly we found that the expression of EOMES was significantly reduced in activated GdA treated hPBMCs, both at the transcriptional and translational level, however T-Bet did not show any variation in expression upon GdA treatment. All the above findings have been compiled in Chapter 2 along with our studies on the possibility of GdA to induce a tolerogenic phenotype in T cells. We found there was no difference in the mRNA level and surface expression of CD103 and CD28 in alloactivated PBMCs, while FOXP3 mRNA did not show any variation upon GdA treatment, indicating that GdA does not induce a tolerogenic phenotype in T-lymphocytes, further confirming our data that the decreased cytolytic activity of CTLs upon GdA treatment was not due to tolerance but due to impaired function Interestingly, IL-2/IL-2R signaling is known to directly regulate perforin and granzyme expression as well as it plays a role in the expression of T-Bet and EOMES. Therefore, as a read out of IL-2 signaling we checked for the surface expression of the high affinity IL-2R subunit, CD25. As expected, CD25 expression was more pronounced in CD4+ T cells and consistent with published reports in literature that GdA suppresses IL-2 synthesis, we also observed a significant reduction in the CD25bright population in both the T cell subsets (CD4+ and CD8+) upon GdA treatment (addressed in Chapter 3). This finding supports a mechanism of action of GdA, wherein the cytolytic activity of CTLs is compromised by the downregulation of EOMES, triggered by the low IL-2 levels. This translates to aberrant synthesis of key cytolytic molecules perforin and granzyme B, leading to low efficiency CTLs, which are further disabled by defective degranulation machinery induced by GdA. We did not look into the mechanistic aspects of how GdA suppresses degranulation, which can be addressed later as a part of another study. Building up on our observations, and taking cues from existing literature, that IL-2 regulates the expression of pro and anti-apoptotic protein levels within activated cells, we looked at the expression profile of Bcl-2 (anti-apoptotic) and Bax (pro-apoptotic) in activated PBMCs upon GdA treatment. There was a significant reduction in the total mRNA and protein level of Bcl-2, while a very significant increase in Bax mRNA and protein was observed. Chapter 3 of the thesis also presents this data and explains a plausible mechanism of the inhibitory effect of GdA on T-lymphocytes. In Chapter 2, we have also addressed the probable reasons for the differences in the responses of CD4+ and CD8+ T-lymphocytes to GdA. Interestingly, surface glycan profile of CD4+ and CD8+ T-lymphocytes upon activation and the surface expression of the most probable receptor for GdA i.e. CD7 was comparable in both the T cell subsets, indicating that possibly the downstream signaling events leading to GdA-induced apoptosis and not the surface binding of GdA may vary in CD4+ and CD8+ T-lymphocytes, due to which we observed a difference in the extent of apoptosis induced in these cell types by GdA although the inhibition of proliferation in both the subsets was comparable. In summary, this study is the first to provide evidence for a possible mechanism of temporal regulation of CTL activity at the feto-maternal interface, where activated CD8+ T cells are abundantly present. We can say with much confidence that binding of GdA to T-lymphocytes causes sub-optimal IL-2 signaling which translates into reduced expression of EOMES and hence downregulation of perforin and granzyme B, leading to impaired CTL activity in CD8+ T-lymphocytes, which is further weakened by the impaired release of the cytolytic molecules from them. Insufficient IL-2 signaling in the presence of GdA can also be a cause of inhibition of proliferation in T-lymphocytes, while the resulting decrease in anti-apoptotic protein Bcl-2 and increase in pro-apoptotic protein Bax seem to contribute to the induction of apoptosis in CD4+ T cell. It will be interesting to explore the mediators involved in the IL-2 signaling pathway that are differentially regulated in CD4+ and CD8+ T cells which confer resistance in CD8+ T cells to GdA-induced apoptosis and also the mechanism by which GdA regulates the degranulation of cytolytic vesicles in CTLs needs to be worked out.