Leveraging CRISPR gene editing technology to optimize the efficacy, safety and accessibility of CAR T-cell therapy

Abstract

Chimeric Antigen Receptor (CAR)-T-cell therapy has revolutionized cancer immune therapy. However, challenges remain including increasing efficacy, reducing adverse events and increasing accessibility. Use of Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) technology can effectively perform various functions such as precise integration, multi-gene editing, and genome-wide functional regulation. Additionally, CRISPR screening using large-scale guide RNA (gRNA) genetic perturbation provides an unbiased approach to understanding mechanisms underlying anti-cancer efficacy of CAR T-cells. Several emerging CRISPR tools with high specificity, controllability and efficiency are useful to modify CAR T-cells and identify new targets. In this review we summarize potential uses of the CRISPR system to improve results of CAR T-cells therapy including optimizing efficacy and safety and, developing universal CAR T-cells. We discuss challenges facing CRISPR gene editing and propose solutions highlighting future research directions in CAR T-cell therapy.

Fig. 1. The mechanism and superiorities of CRISPR-Cas system applied in CAR T-cell therapy.

A The mechanism and advantage of CRISPR-Cas9. In CRISPR-Cas9 system, the Cas9 nuclease is guided by a sgRNA and directed to the desired target sites. When Cas9 hybridizes to the target site, it generates DSBs, then the DSB was repaired through endogenous pathways including non-homologous end-joining (NHEJ) and homology-directed repair (HDR). ZFNs and TALENs rely on protein domains to recognize target sequences, which are built by assembling amino acid modules. B CRISPR tools within CAR T-therapy. CRISPR interference (CRISPRi) or CRISPR activation (CRISPRa) are consist of dCas9 fused with transcriptional repressors or activators, respectively. The epigenome-modifying enzyme enables the epigenetic regulation. Base editors are typically a fusion protein combining a nCas nickase with a deaminase domain, which catalyses the substitution of a single nucleotide at the PAM strand in the R-loop. Prime editors consist of a nCas fused with a reverse transcriptase domain. A prime editing guide RNA (pegRNA) contains a template of the desired sequence at the 3′ end as well as a target-specific spacer sequence. Cas12a is guided to the target DNA sequence by self-processed mature crRNA. Cas12a recognizes a PAM motif matching the target sequence, 5’-TTTN-3’, and forms a more reversible binding with the DNA target sequence than Cas9. Subsequently, Cas12a cleaves the non-target strand (NTS) followed by the target strand (TS), creating a DSB. Cas13d is capable of self-processing the crRNA precursor into mature crRNA, which then guides the cleavage of target RNA, enabling the regulation of gene expression at the transcriptional level. Therefore, by constructing multi-crRNA expression arrays, Cas13d also possesses the functional characteristic of regulating the expression of multiple genes. Cas13d has no limitations from the protospacer flanking sequence (PFS), and can almost target any RNA sequence, possessing a broad editing range. Created with BioRender.com.

Fig. 2. The genome editing feature of CRISPR systems.

The application of the CRISPR system in CAR T-cell therapy mainly includes site-specific integration and multi-gene editing. By transferring CAR HDR templates encoded in the form of electroporation or AAV into T cells, the CAR sequence can be specifically integrated into the TRAC site, reducing graft-versus-host reactions and being used to produce universal CAR T-cells. This has achieved the “one stone two birds” gene editing effect of knocking in and knocking out, avoiding the oncogenic risk of random integration of viral vectors. Since the Cas protein responsible for DNA cutting is always the same, only the corresponding gRNA needs to be designed according to the target site, enabling the Cas enzyme to be expressed simultaneously with various gRNAs in the cell, thus achieving the purpose of multi-gene targeting and editing. By using electroporation or lipid particle encapsulation, Cas mRNA and gRNA or pre-formed ribonucleoprotein complexes (RNP) can be directly delivered into cells. This will transiently express the CRISPR editing system, with many advantages such as low cost, low off-target probability, and no oncogenic mutagenesis risk of viral vectors. Created with BioRender.com.

Fig. 3. CRISPR screening applied for CAR T-cell therapy.

A CRISPR screening can be classified into loss-of-function (LOF) and gain-of-function (GOF) screening based on different strategies for gene perturbation. The commonly used method for CAR T-cell functional targets identification involves introducing Cas9 protein via electroporation and transducing a lentivirus sgRNA library into the T cell pools, while GOF screening uses lentivirus packaging dgRNA library into Cas9-expressing T cells. The newly identified targets are validated in CAR T-cells with gene knockout or activation. It is also possible to conduct CRISPR screening directly in CAR T-cell pools, but performing so after lentivirus-mediated CAR gene introduction into T cells could lead to interference. Therefore, the CAR gene and the gene-edited crRNA library can be inserted into the same TRAC locus using the CLASH system (AAV library, Cas12a), achieving effective CRISPR screening in CAR T-cells. Marker expression can be employed for sorting by comparing the sgRNA sequencing between population with higher and lower expression levels, or sorting by comparing the sequencing results with a control group through co-culturing with tumor cells under immune challenge. B In Cas9-expressing tumor cells, introducing an sgRNA lentivirus library and screening through marker expression or immune challenge of CAR T-cells can identify tumor immune escape mechanisms and drug-resistant genes targeting CAR T-cell therapy. Created with BioRender.com.

Fig. 4. CRISPR system applied for advancing CAR T-cell therapy.

(i) Optimize the efficacy of CAR T-cells. The inhibitory pathways, the immunosuppressive factors caused by relative cells and hypoxia in TME, as well as the epigenetics of T cells, can all contribute to T cell exhaustion and decrease the effector function of CAR T-cells. Therefore, it is possible to maximize the therapeutic efficacy of CAR T-cells in the treatment of malignancies by employing CRISPR system to knock out immune checkpoints, TME-responsive receptors, and other molecules (such as Fas for promoting apoptosis, and ATG5 for increasing autophagy). Additionally, eliminating negative regulatory factors of cytokines, inflammatory factors, and CAR molecule expression using CRISPR system can also enhance the efficacy. (ii) Control adverse events of CAR T-cell therapy. GM-CSF knockout CAR T-cells can effectively alleviate CRS and ICANS, while depletion of shared antigens on T cell surface (CD7, CD5), and CD33 on HPSC can tackle OTOT toxicity. Created with BioRender.com.

Fig. 5. The application of CRISPR system in Manufacturing UCAR T-cells.

A UCAR T-cells can be generated from healthy allogeneic donors to benefit multiple recipients. The manufacturing process for UCAR T-cell products starts with a source of third-party healthy T lymphocytes collected by leukapheresis. CRISPR/Cas-mediated precision editing of T cells or induced pluripotent stem cells (IPSCs) has the potential to eliminate expression of endogenous αβ-TCR, HLA and PD-1 etc, and insert a recombinant DNA coding for a CAR gene simultaneously. T cells or differentiated IPSC-T cells are then expanded using anti-CD3/anti-CD28 beads and cytokines. The remaining αβ TCR-positive cells are magnetically removed using anti-αβ TCR antibodies. The product is then packed and shipped to hospitals for using. B Combining CAR transfer with CRISPR-Cas-mediated genome editing offers the strategies to enhance CAR T-cells. Endogenous αβ TCR is removed to prevent GVHD, and HLA class I and class II molecules are also removed to prevent HVGD (through deletion of B2M/HLA-A, B and CIITA respectively). Persistence can also be achieved by deleting CD52 (allow cells to persist in the presence of alemtuzumab for lymphodepletion), or by deleting the NK cell activator Poliovirus receptor (PVR) and addition of a natural killer (NK) cell inhibitor (such as HLA-E, HLA-G, CD47 and CD300a TASR) for resistance to NK cells attack. Disrupting programmed cell death protein 1 (PD-1) enables UCAR T-cell to counteract some mechanisms of immunosuppression from tumor. Created with BioRender.com.

Fig. 6. Challenges and future direction of CRISPR gene editing applied in CAR T-cell therapy.

A Adverse events and feasible solutions. Off-target mutations are caused by sgRNA recognizing the off-target locus in the genome and making unnecessary cleavage at the wrong locus. Cas-CLOVER, fuses the catalytically inactive dCas9 with the Clo51 (CLOVER) nuclease domain and uses two gRNAs to only cleave when the Clo51 nuclease dimerizes. Another approach “spacer-nick” combines the nCas9 with a pair of PAM-out sgRNAs at a long spacer distance. Hence Cas9n is guided by the spacer-nick sgRNAs to two target sequences on opposite strands and nicks both DNA strands at an optimal distance to preserve efficient HDR while minimizing NHEJ events. Chromosomal aberrations, such as large deletions and translocations, are indeed frequent phenomena. Here shows possible instances of chromosomal loss and translocations. The key factor leading to chromosomal aberrations is the generation of DSBs. BE develops precise single-base substitutions at multiple sites enabling DSB-free gene editing, thereby reducing the risk of gene rearrangements. B Traditional delivery methods include viral vectors and electroporation. Viral vectors mainly include LV, AV, and AAV. CRISPR components delivered by viral vectors in the form of DNA may increase the risk of off-target, and viral vectors also have other disadvantages including immunogenicity, cargo size limitations, and lower transduction efficiency. The delivery of electroporation in CAR T-cells may cause cell damage. Peptide delivery is an emerging delivery system for CRISPR components in CAR T-cells. LNP and VLP have the potential for in vivo delivery, but two basic issues need to be considered, that is, whether the relevant cell types can be targeted and avoid cell damage and immune rejection in vivo. Created with BioRender.com.