Schrum, Adam G.

Abstract Description Diffuse large B-cell lymphoma (DLBCL) is an aggressive malignancy derived from transformed B cells of germinal center (GC) or post-GC origin. DLBCL incidence is elevated in immunodeficient persons with reduced T cells, such as among HIV+ patients. Yet a causal link between T cell deficiency and DLBCL is incompletely understood. Using a mouse model of spontaneous DLBCL, we show that suboptimal T cell function, but neither zero nor full T cell function, can be a driver of B cell lymphomagenesis. In CD3δ-/- mice, which exhibit reduced number, signaling, and function of T cells, we observed progressive disruption of splenic architecture and eventual DLBCL development with ∼100% incidence. In contrast, lymphoma did not develop in wild-type mice, or CD3e-/-z-/- mice with complete T cell absence. This showed that poor-but-positive suboptimal T cells were required for GC B cell transformation and outgrowth. Mechanistically, CD3δ-/- mice exhibited altered GC dynamics at steady state and in response to immunization. Compared with wild-type mice, CD3δ-/- mice had a diminished T follicular regulatory (TFR) compartment and exhibited defective T follicular helper (TFH) cell differentiation when antigen challenged, indicating compromised regulation of the GC reaction. Indeed, defective GC B cell entry and differentiation was also observed. Our data thus suggest that suboptimal function of T helper subsets can lead to dysegulated GC reactions and B cell lymphomagenesis. Funding Sources Supported by AAI Intersect Fellowship Topic Categories Immune Response Regulation: Cellular Mechanisms (IRC)

Abstract The αβ T cell antigen receptor (TCR)/CD3 complex is currently viewed as a monovalent receptor with the subunit composition αβγδε2ζ2. This model has emerged from among at least five different potential stoichiometries that were proposed in previous years. Recently, the case for the accepted stoichiometry has been supported by experiments involving co-immunoprecipitation (IP) followed by Western blot, and by blue native (BN)-PAGE. We have revisited the issue of TCR/CD3 stoichiometry by examining immunoprecipitated complexes by flow cytometry (IP-FCM), a technique which allows a quantitative estimate of the subunit constituents in native multi-protein complexes. Together with size-exclusion chromatography, the data supports the conventional stoichiometry, but provides evidence that the copy number may be two, making the TCR/CD3 a precisely bivalent complex, [αβγδε2ζ2] x 2. The expression of a bivalent TCR/CD3 complex has implications regarding possible mechanisms by which the interaction with antigen may trigger signaling. Research support: Adam G. Schrum, start-up funds from the Mayo Foundation; Laurence A. Turka, grants AI41521 and AI 43620 from NIH; Ed Palmer, grants from the Swiss National Science Foundation.

Abstract Upon positive selection of conventional αβ T cells, CD4+ CD8+ double-positive (DP) cells transition to single-positive (SP) and reduce their responsiveness to weak peptide/MHC ligands (affinity threshold conversion, ATC). We report that ATC actually begins prior to the completion of positive selection in the DP stage, and this checkpoint is where the TCRα/CD3δ signaling axis controls positive selection. We dissected ATC in the OT1 TCR transgenic system in conjunction with several different genes whose null mutations block thymic development at DP stage: β2m, CD3δ, and/or TCRα-Connecting Peptide Motif (αCPM). In the absence of MHC class I (β2m0), CD3δ or αCPM mutants were blocked at pre-selection DP stage and displayed immature, high-sensitivity to weak TCR ligands. In contrast, when MHC class I was present (β2m+), CD3δ or αCPM mutants were blocked at DP stage but underwent partial ATC; they began positive selection but did not complete it. Upon exogenous TCR engagement of partial-ATC thymocytes, the proximal signaling protein interaction network showed overall weakened activity, notably decreasing SOS1 participation linked to the downstream Ras/Raf/ERK selection pathway. Provision of high affinity peptide/MHC in FTOC rescued completion of positive selection for partial-ATC mutant thymocytes. Finally, comparing nontransgenic, polyclonal B6 vs. CD3δ0 DP thymocytes, although most cells showed equal maturation state, a subset of cells in B6 progressed in maturation relative to CD3δ0. We propose the existence of an intermediate checkpoint inside DP stage, in which ATC already begins to reduce responsiveness to selection ligands, yet further TCR signaling is required to complete positive selection.

Abstract TCR mediated antigen recognition results in a signaling cascade leading to T cell activation. Antigen binding to its cognate TCR drives receptor aggregation and conformational changes that trigger signaling from the receptor. Conformational change in the CD3 complex (CD3Δc) is required for robust signaling and productive T cell responses. In contrast, weakly recognized antigens do not induce CD3Δc and the resulting T cell response is poor. Thus, T cell responses to weak antigens may be improved by induction of CD3Δc. We previously described an anti-mouse CD3ɛ monovalent Fab (Mono-7D6-Fab) that induced CD3Δc and synergized with weak antigens to enhance the T cell response, a process we termed co-potentiation. However, Mono-7D6-Fab did not induce classical T cell signaling in the absence of antigenic ligation. Mono-7D6-Fab effectively co-potentiated T cells against murine B16-F10 melanoma cells and reduced tumor burden in vivo in a T cell dependent manner. We exploited a model of human cytomegalovirus (CMV) to generate proof of concept that co-potentiation may enhance T cell antigen recognition in humans. Anti-human CD3ɛ monovalent Fab (Mono-OKT3-Fab) improved the CD8 T cell response to the CMV pp65495–503 peptide as measured by increased T cell proliferation, cytokine production, and the ability to kill target cells. Co-potentiation was dependent on antigen specificity, as 1) the effects were dampened when antigen recognition was impaired and 2) sequencing of the TCRs revealed a restricted number of expanded TCR clones. Our results suggest that Mono-OKT3-Fab may have therapeutic clinical applications to achieve a more robust antigen-specific T cell response in the context of diverse human disease, from cancer to chronic infection.

Abstract The αβ T cell antigen Receptor (TCR) is a multi-protein complex expressed together with CD3 subunits (γδεζ). The conventional model has it as monovalent with a stochiometry of (αβγδε2ζ2); however, some previously published work reported observation of bivalent complexes with two αβ heterodimers. To reevaluate the valency of the TCR, we have subjected T cell digitonin-solubilized complexes to immunoprecipitation detected by flow cytometry (IP-FCM), size exclusion chromatography, and other methods for detecting complex size and subunit composition. Our results indicate that precisely bivalent TCRs are found among surface-expressed complexes independent of cross-linking by strong antigenic ligands. Their detection can be masked in the presence of Coomassie reagent from Blue-native PAGE conditions, or if antibodies used in detection assays are themselves bivalent. Therefore, bivalent TCR/CD3 complexes can be found under the biochemical conditions that have classically defined the minimally expressed TCR/CD3 unit (digitonin solubilization). Further investigation is required to determine the functional relevance of these bivalent TCR complexes, although greater antigenic binding and signaling is predicted over that of monovalent TCRs.

Abstract TCR interactions with agonist pMHCs induce a conformational change in the tails of the associated CD3 complex (CD3Δc) that contributes to efficient TCR/CD3 triggering and productive T cell responses. In contrast, TCR engagement by antagonistic pMHCs does not induce either CD3Δc or full TCR/CD3 signaling in mature T cells. Here, we have used a monovalent Fab fragment (Mono-Fab) specific for the murine CD3 complex to provide induction of CD3Δc to mature T cells engaging pMHC ligands. This Mono-Fab does not interfere with TCR capacity to bind pMHC and fails to elicit signaling in the absence of TCR/MHC interactions. Using OT-I Tg T cells and a collection of altered peptide ligands derived of the agonist peptide OVA, we show that addition of CD3Δc to OT-I T cells using Mono-Fab enables T cell responses to antagonist ligands that fail to induce CD3Δc on their own. Moreover, using a mouse model of metastatic melanoma, we show that Mono-Fab promotes in vivo T cell dependent anti-melanoma responses that are antigen-specific. This Mono-Fab can also be combined with established therapies targeting signals 2 and 3 involved in T cell activation for a synergistic effect. Our data suggests that in-trans CD3Δc provision decreases the activation threshold of T cells. We propose that targeting the CD3 complex for the induction of CD3Δc could be exploited to achieve efficient TCR/CD3 triggering by poorly immunogenic antigens in the context of vaccination against cancer and infectious diseases.

Abstract Signaling networks and the protein interactions that define them are of great interest, but techniques to assess these networks in unmanipulated primary human cells are limited. We have developed and applied a novel technological and analytical platform to measure proteins in shared complexes detected by exposed surface epitopes (PiSCES). Immunoprecipitation reactions are performed in multiplex using microsphere beads to measure 20+ proteins and 210+ distinct pair-wise protein associations that are important for T cell antigen receptor signaling. Jurkat T cells stimulated with superantigen generated an extensive PiSCES signature using this method and provided a dataset from which we optimized an analytical package. The system is currently being utilized to address the hypothesis that T cells isolated from skin lesions of autoimmune alopecia areta (AA) patients will have a unique PiSCES signature compared to those isolated from non-AA donors. To investigate this, we have isolated non-engineered primary cells from AA patient skin lesions and healthy controls, and have begun to identify differences in their PiSCES signatures, either directly after isolation or after stimulation with anti-CD3 and anti-CD28. We conclude that this novel PiSCES approach is capable of generating protein-protein interaction signatures that provide a unique view of TCR signalosome activity solicited from physiologic non-engineered primary human patient samples.

Abstract MHC multimers including tetramers and pentamers provide a powerful way to visualize antigen specific T cell responses in both experimental and clinical immune assays. The multimers can be acquired commercially, from the NIH, or assembled in the laboratory for measuring immune responses to commonly-studied or novel-hypothetical antigens. The multimer reagents can be expensive and time consuming to acquire and in many cases are uncontrolled reagents as robust positive control T cells used to assess the reagents functionally may not be available. This is particularly problematic when using MHC multimers to investigate less well-defined T cell responses, such as those characterized by low T cell frequencies or by specificity to hypothetical antigens relevant to cancer, infection, or autoimmune disease. These problems are magnified when studying irreplaceable clinical trial samples where having confidence that reagents are biochemically intact is critical. To address this problem, we have devised a versatile, flow cytometry-based, capture assay to probe the structural integrity of MHC multimer reagents. In principle, the idea is to immobilize on a bead antibody specific for conformational determinants expressed on a properly folded MHC molecule. The particle becomes labeled and can be visualized and quantified by flow cytometry when a fluorescently tagged tetramer is adsorbed to the particulate upon binding to the antibody.

Abstract Tumor antigen-specific T-cell tolerance is a critical element of tumor escape. It is known that MDSC can induce antigen-specific CD8+ T cell tolerance in an experimental system with adoptive transfer of transgenic T cells. In this study we addressed the critical question whether a similar phenomenon can be observed in a "natural" non-transgenic experimental model. Balb/C mice were subcutaneously inoculated with two different tumor cell lines, a derivative A2L2 cell line, which expresses high levels of HER2/neu and the parental cell line 66.3. HER2/neu specific CD8+T cells were generated by immunization of mice with peptide. MDSC were able to tolerize antigen-specific T cells in vivo if they were collected from A2L2 but not from 66.3 tumor-bearing mice indicating the tumor-specific nature of T cell tolerance induced by MDSC. Next we investigated the type of signaling caused by MDSC in CD8+T cells that could lead to tolerance. MDSC dramatically reduced the ability of CD8+ T cells to respond to the specific peptide. However, study of various signaling molecules using different experimental systems and tests found that MDSC did not induce negative signaling in T cells. This conclusion was confirmed in adoptive transfer experiments with CD8+T cells expressing dual TCR: LCMV glycoprotein peptide gp33-41 and OVA-derived H-2Kb-restricted peptide ova257-264. We found that MDSC loaded with peptide specific for one TCR induced tolerance only against that specific epitope but CD8+ T cells retained the ability to respond to the peptide specific for the other TCR expressed on the same cell.

During normal T cell development in mouse and human, a low-frequency population of immature CD4 − CD8 − double-negative (DN) thymocytes expresses early, mature αβ T cell antigen receptor (TCR). We report that these early αβ TCR+ DN (EADN) cells are DN3b-DN4 stage and require CD3δ but not major histocompatibility complex (MHC) for their generation/detection. When MHC - is present, however, EADN cells can respond to it, displaying a degree of coreceptor-independent MHC reactivity not typical of mature, conventional αβ T cells. We found these data to be connected with observations that EADN cells were susceptible to T cell acute lymphoblastic leukemia (T-ALL) transformation in both humans and mice. Using the OT-1 TCR transgenic system to model EADN-stage αβ TCR expression, we found that EADN leukemogenesis required MHC to induce development of T-ALL bearing NOTCH1 mutations. This leukemia-driving MHC requirement could be lost, however, upon passaging the tumors in vivo, even when matching MHC was continuously present in recipient animals and on the tumor cells themselves. These data demonstrate that MHC:TCR signaling can be required to initiate a cancer phenotype from an understudied developmental state that appears to be represented in the mouse and human disease spectrum.