Investigating the Structural Requirements of Chimeric Antigen Receptors for T cell Activation

Abstract

Introduction: The application of T cells modified to express Chimeric Antigen Receptors (CAR) has recently gained clinical approval for treatment of a subset of B cell malignancies, having demonstrated impressive clinical responses in otherwise untreatable disease (ORR 82%, CR 54% in patients with relapsed or refractory large B cell lymphoma). One of the key co-stimulatory signalling molecules included in the design of second-generation CARs is CD28. Inclusion of this molecule results in enhanced engraftment and anti-tumour activity compared to first-generation receptors. In addition to the essential cytoplasmic signalling moiety, transmembrane and extracellular receptor components are also important features in receptor design. In the clinically approved Axicabtagene Ciloleucel CAR T cell platform these components are derived from CD28. Despite inclusion in this receptor, the functional implications of certain structural features are not completely elucidated including the ligand binding motif and the capacity for disulfide mediated homodimerisation. Methods: The focus of this study was to determine whether the CD3ζ isoform 1 and 2 can be used interchangeably in CAR design and to investigate the functional implications of specific structural aspects of the extracellular domain of CD28 CARs. These include the role of the endogenous CD28 natural ligand binding domain, inclusion of the full CD28 extracellular domain and disulfide mediated receptor dimerisation in the context of both truncated and non-truncated CD28 CARs. To investigate the functional contributions of these features, a panel of receptors with specific mutations were employed. The functional implications of these modifications were assessed using in vitro activity assays including cytotoxic function, degranulation capacity, cytokine production and expression of activation associated markers. Results: The results of this study demonstrate that in the context of second generation CD28 CARs bearing a full extracellular domain, the natural CD28 ligand binding motif facilitates binding of CD80/CD86 ligands. This binding was prohibited by mutation of the binding motif. CAR specific activation through the CAR in the presence of a functional ligand binding motif specifically resulted in enhanced secretion of IL-17a, IL-8, IL-10 and IL-6, particularly at a high density of target antigen. Inclusion of the extracellular domain conferred CARs with enhanced expression and activity, compared to those expressing truncated counterparts. These features result in an intrinsic enhancement in CAR T cell functional avidity. Abrogation of disulfide mediated dimerisation of CD28 CAR resulted in reduced expression and stability of CAR compared to the parental counterpart. Abrogating disulfide-mediated CAR dimerisation resulted in a reduced capacity to secrete high levels of IL-2 and increased the threshold of target antigen to induce a functional IL-2 response. Conclusion: Use of either isoform 1 or 2 of CD3ζ in CAR design did not result in any detectable differences in T cell activity. Inclusion of the CD28 endogenous ligand binding motif resulted in increased production of cytokine release syndrome related cytokines. The CD28 extracellular domain serves as a suitable platform to alter CAR surface expression and stability, and thus functional avidity depending on the target antigen. Homodimerisation is an important feature of CD28 CARs which can be harnessed in CAR design.

Date of Award	9 Nov 2020
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Anne Armstrong (Co Supervisor), Robert Clarke (Main Supervisor), Eleanor Cheadle (Co Supervisor), Milena Kalaitsidou (Co Supervisor) & Gray Kueberuwa (Co Supervisor)