ESTABLISHMENT OF A CRISPR/CAS9-BASED GENOME EDITING PLATFORM FOR THE GENERATION OF AN OFF-THE-SHELF ALLOGENEIC CAR-T CELL MODEL

Abstract

Introduction: Chimeric antigen receptor (CAR)-T cell therapy has a remarkable clinical success. However, its autologous nature reduces accessibility to the therapy. Thus, one of the next frontiers in the field is to develop an allogeneic, ‘off-the-shelf'therapy. This can be achieved by knocking out the T cell receptor (TCR) from CAR-T cells using genome-editing tools to prevent the development of graft versus host disease. In addition, depletion of the human leukocyte antigen I (HLA-I) on the surface of CAR-T cells would dramatically reduce the chance of CAR-T cell rejection by the host immune system. Objective: Establish a methodology to produce TCR− and HLA-I− CAR-T cell models by knocking out the TRAC (T Cell Receptor Alpha Constant) and B2M (Beta-2-Microglobulin) genes, respectively, using CRISPR/Cas9 genome editing. Methods: We tested four guide RNAs (gRNAs) sequences, two against the TRAC gene (TRAC1 and TRAC2) and other two against the B2M gene (B2M1 and B2M2). The gRNAs sequences were cloned individually into lentiviral vectors upstream the anti-CD19 CAR gene, and the corresponding lentiviral particles were used to transduce Jurkat cells. We used these cells as a model for TCR and HLA depletion after the electroporation of a plasmid for a high-fidelity Cas9 transient expression. Results: After transduction, CAR+ Jurkat cells were enriched by magnetic selection to obtain homogeneous expression of CAR and gRNAs. We achieved above to 94% of CAR expressing cells for all populations. Next, TRAC/CAR+ cells were electroporated with a plasmid encoding Cas9 and GFP using two different media: Solution V and Opti-MEM. After 24h of electroporation in Solution V, Jurkat cells displayed high viability (̃90%) regardless of the presence of Cas9-GFP plasmid. In Opti-MEM medium, however, electroporation induced high rates of cell death (12.1-16.0% viability). Furthermore, the frequency of GFP+ cells after electroporation in Opti-MEM was 1.6-fold (TRAC1) and 1.9-fold (TRAC2) lower compared to Solution V. To evaluate the efficiency of TRAC gRNAs in depleting the TCR, electroporated cells in Solution V were analyzed after 1, 3 and 8 days for CD3 expression. Notably, TRAC gRNA 1 led to a remarkable reduction in the frequency of CD3+ cells (from 91.6 to 7.41%). In contrast, this reduction was less pronounced for TRAC gRNA 2 (decrease of 21.8%). For the B2M gRNAs analysis, we electroporated the corresponding Jurkat cells in Solution V and observed similar transfection efficiency compared to TRAC/CAR+ cells in the same media (60% of GFP+ cells). In addition, the B2M gRNAs efficiency was evaluated by HLA-I immunolabeling, and we observed a change in the frequency of HLA-I+ cells only for the gRNA 2, with a reduction of 51.9% 8 days post electroporation. Finally, we confirmed the superior editing performance of TRAC1 and B2M2 gRNAs through the T7 endonuclease 1 assay. Conclusion: These data demonstrate the feasibility of TCR and HLA-I depletion by knocking out the TRAC and B2M genes using CRISPR/Cas9 and establish the best gRNA sequences for this purpose, paving the groundwork for the generation of a platform to produce off-the-shelf allogeneic CAR-T cells. Funding: FAPESP (2013/08135-2, 2019/18672-1, 2019/18702-8 and 2020/02043-2) and CNPq (442484/2020-8).