Targeting PRC2 enhances the antitumor cytotoxic capacity of anti-CD19 CAR-T cells against hematological malignancies

Chimeric Antigen Receptor (CAR) T cell therapy was adopted as a clinical modality for patients with relapsed/refractory hematological malignancies. Despite the clinical efficacy of CAR-T cell therapy, a considerable fraction of patients still relapses during the first months following CAR-T cell infusion. The limited CAR-T cell efficiency is thought to relate to epigenetic mechanisms involved in T cell suppression and dysfunction. Here, screening of multiple epigenetic inhibitors revealed that targeting PRC2 consistently enhanced the cytotoxic/effector phenotype of CD8 T cells. Notably, PRC2 inhibition promoted the differentiation of GZMB+ effector memory 19BBζ CAR-T cells and enhanced their antitumor activity both in vitro and in vivo. Consistent with their long-lasting antitumor activity, PRC2-inhibited 19BBζ CAR-T cells did not exhibit any signs of dysfunctionality/exhaustion. Furthermore, TCR restimulation along with PRC2 inhibition of patient-derived anti-CD19 CAR-T cells also induced the development of GZMB+ effector memory cells and elicited potent antitumor responses against CD19+ Daudi cells. In line with this, the gene signature derived from in-house PRC2-inhibited 19BBζ CAR-T cells was enriched in tisagenlecleucel (tisa-cel) BBζ CAR-T cell therapy responders with large B-cell lymphoma. Collectively, our results demonstrated that targeting PRC2 may be a promising approach to enhance a functional effector program in CAR-T cells against hematological malignancies. Leukocyte removal filters from healthy donors were used to isolate PBMCs through Ficoll-Paque PLUS density gradient medium (Cytiva), followed by T cell isolation using CD3 MicroBeads (Miltenyi Biotec). CAR-T cells were generated using the "Sleeping Beauty" Transposon System. CD3 T cells were resuspended in 100μL of 1SM buffer, 10μg PT3-19BBz and 1μg SB100X plasmids, and then transferred into a 0.2 cm cuvette (Mirus Biotech®, Madison, WI). T cells were then electroporated (U-014 program) using the Lonza® Nucleofector® II electroporation system. After 24h resting, T cells were in vitro expanded with TransactTM CD3/CD28 T-cell activator [1:200] (Miltenyi Biotec) in media containing 50U recombinant human IL-2 (Thermo Fisher) together with PRC2 inhibitors or their corresponding negative controls for 7 days. Total RNA from CAR-T cells was extracted using ReliaPrep RNA Tissue Miniprep System (Promega) according to the manufacturer’s instructions.The quality of total RNA extracted from in house-produced CAR-T cells was assessed by RNA Screen Tape on TapeStation (Agilent). cDNA libraries were constructed from 400 ng of RNA using the Stranded mRNA Prep Ligation kit (Illumina). Sequencing was performed using the Illumina NextSeq 2000 system (P2, 200 cycles kit), and at least 50 million paired-end reads were obtained for each sample. Reads were aligned to the GHCh38/hg38 human genome (GTF v109) using STAR v2.7.10b (--twopassMode None --outSAMunmapped Within --chimSegmentMin 20 --limitIObufferSize 30000000 62500000) and quantified using htseq v2.0.3 (-m union -i gene_id -r pos -s no), with default parameters unless otherwise specified. Expression levels were calculated as log2(TPM+1).