ALBERT

Description: Transfer of tumor antigen–specific T-cell receptors (TCRs) into human T cells aims at redirecting their cytotoxicity toward tumors. Efficacy and safety may be affected by pairing of natural and introduced TCRα/β chains potentially leading to autoimmunity. We hypothesized that a novel single-chain (sc)TCR framework relying on the coexpression of the TCRα constant α (Cα) domain would prevent undesired pairing while preserving structural and functional similarity to a fully assembled double-chain (dc)TCR/CD3 complex. We confirmed this hypothesis for a murine p53-specific scTCR. Substantial effector function was observed only in the presence of a murine Cα domain preceded by a TCRα signal peptide for shuttling to the cell membrane. The generalization to a human gp100-specific TCR required the murinization of both C domains. Structural and functional T-cell avidities of an accessory disulfide-linked scTCR gp100/Cα were higher than those of a dcTCR. Antigen-dependent phosphorylation of the proximal effector ζ-chain–associated protein kinase 70 at tyrosine 319 was not impaired, reflecting its molecular integrity in signaling. In melanoma-engrafted nonobese diabetic/severe combined immunodeficient mice, adoptive transfer of scTCR gp100/Cα transduced T cells conferred superior delay in tumor growth among primary and long-term secondary tumor challenges. We conclude that the novel scTCR constitutes a reliable means to immunotherapeutically target hematologic malignancies.

Description: Grafting T cells by tumor-antigen specific T cell receptors (TCR) could trigger the initiation of effector function and redirect T cell cytotoxicity towards tumors. We utilized various HLA-A2.1 transgenic mice to bypass human MDM2- and p53-specific self-tolerance. In contrast to the use of HuCD8×A2Kb transgenic mice to generate an MDM2-specific CD8-dependent TCR, we generated a high-affinity, CD8-independent p53-specific TCR in single human A2.1 transgenic mice. The efficiency of double chain (dc) TCR modified T cells could be affected by the incorrect TCR α/β chain pairing between endogenous and transgenic TCR constructs to form hybrid TCR potentially leading to autoimmunity. To address this concern, chimeric A2.1-restricted peptide-specific murine single chain (sc) TCRs were constructed (Vα-Li-VβCβ) and retrovirally transduced into human T cells. Despite detectable surface expression, these chimeric receptors were not able to convey any MDM2- or p53-specific cytolytic activity. Therefore we developed a truncated TCR-alpha domain (Cα) comprising solely the TCRα signal peptide, the ecto-domain, the transmembrane region as well as the cytoplasmic tail and cotransduced these construct with the scTCRs. We anticipated that Cα would stabilize scTCR expression by interacting with the single chain beta chain. Indeed, this approach not only led to increased expression levels of the chimeric scTCRs, but also induced specific lysis of A2.1 positive MDM2 or p53 peptide-pulsed target cells as well as solid tumor cell lines. Recognition of malignant targets by p53 specific scTCR transduced CD4 and CD8-positive T cells was equivalent to that observed with double-chain p53 TCR gene modified effector cells. To test whether this concept is applicable to human TCRs as well, we constructed a human gp100-specific scTCR and a human Cα domain. In contrast to the gp100-specific double chain TCR, only a marginal expression pattern was observed for the human scTCR / Cα constructs. Introduction of an additional disulfide bond within the constant domains in order to stabilize TCR surface expression showed no effect. Since murine TCR are expressed on human T cells to a much higher extent, the human constant β-domain of the scTCR was replaced by murine Cβ. Comparable to the murine scTCR concept, the chimerized scTCR coexpressed with murine Cα demonstrated high cell surface expression and triggered cytotoxicity of malignant A2.1/gp100-positive targets. In summary, our results lay a commonly applicable conceptual basis for the construction of therapeutic scTCR to prevent recombination of natural and transgenic dcTCR alpha and beta chains.