Rapid modular CAR-T generation with CRISPR/Cpf1 and AAV systems

Chimeric antigen receptor \(CAR) T cells have recently become powerful players in the arsenal of immune-based cancer therapy. More recently, gene-editing technologies have enabled more direct engineering of immune cells. Targeting CAR to the TRAC locus has been shown to enhance CAR-T stability and function. However, current lentiviral, retroviral, or CRISPR/Cas9 based methods have various limitations in CAR targeting e�ciency and modularity, especially for generation of multi-component CAR-T cells. Here we describe a new method, the AAV-Cpf1 KIKO system, using a combination of viral and non-viral approaches to generate a stable CAR-T with homology-directed repair \(HDR) knock-in and immune checkpoint knockout at high e�ciency in one step. This protocol accompanies Dai et al. Nature Methods “One-step generation of modular CAR-T with AAV-Cpf1”.


Introduction
Genetically modi ed T cells expressing anti-CD19 CARs have showcased their e cacy in various liquid cancers 1-5, and have been approved for clinical use in B-cell lymphomas and leukemias 6-8.Currently used delivery platforms are primarily based on lentivirus or retrovirus.Recently, CRISPR-Cas9 systems have been used for targeted knockin of anti-CD19 CAR into the T cell receptor α constant \(TRAC) region, which have shown higher e cacy of tumor eradiation in mice 9.The generally used method to modify human T cells is currently based on the CRISPR/Cas9 system 10.Cas12a/Cpf1 hold advantages over Cas9 as it can process crRNA arrays by itself and mediate multi-target DNA cleavage in a single customized array.In the associated publication, we describe a new method for generating CAR-T cells using a combination of mRNA electroporation for LbCpf1 and AAV6 for crRNA and homology-directed repair \(HDR) template.We achieved simple yet highly e cient targeting of both HDR-mediated CAR-T knockin and immune checkpoint gene knockout \(KIKO).Notably, we were also able to deliver anti-CD19 and anti-CD22 CARs for bispeci c CARs at high e ciency, demonstrating the modularity and extensibility of this system.Other advantages of this method include design simplicity, higher delivery e ciency, lower toxicity, reduced exhaustion, increased effector function, and long term CAR enrichment.The e ciency of this approach makes it readily feasible to produce single knockin or double knockin CAR-T cells on the order of 1e8 to 1e9 from a regular source of blood in two to three weeks, which is the scale and timeline typically needed in the clinical setting.Figure 1 illustrates a simple work ow for the generation and functional testing of CAR-T cells using the AAV-Cpf1 KIKO system.We anticipate this system to be of use to the scienti c community for CAR-T research and production.
Quantify and sort labeled cells on BD FACSAria II. 4. The staining patterns were analyzed using FlowJo software 9.9.4 \(Treestar, Ashland, OR).T7E1 assay 1.Five days after electroporation, harvest the bulk transduced T cells and sorted T cells.The genomic DNA was collected using the QuickExtract DNA Extraction Solution \(Epicentre).2. PCR amplify target loci from genomic DNA around cutting site.
TRAC_suvF: CTGAGTCCCAGTCCATCACG TRAC_suvR: AGGGTTTTGGTGGCAATGG PDCD1_suvF: GTAGGTGCCGCTGTCATTGC PDCD1_suvR: GAGCAGTGCAGACAGGACCA 3. Run PCR amplicons on 2% Egel EX and purify \(with known band size) using QIAquick Gel Extraction Kit. 4. After puri cation, denature 200 ng of puri ed PCR product, anneal, and digest with T7E1, 37℃ 45min \(New England BioLabs).5. Load digested PCR products into 2% E-gel EX and quantify DNA fragment abundance using E-Gel™ Low Range Quantitative DNA Ladder \(ThermoFisher).HDR quanti cation and NGS sequencing analysis Semi-quantitative In-Out PCR 1. Use three primers for In-Out PCR: TRAC 1st: binds to a sequence from the left TRAC homology arm TRAC 2st: binds to genomic sequence outside of this AAV donor CD22CAR 3rd: recognizes a sequence contained in the m971-BBz cassette TRAC 1st: CCCTTGTCCATCACTGGCAT TRAC 2st: GCACACCCCTCATCTGACTT CD22CAR 3rd: GAAATCAAAGCGGCCGCAG 2. Normalize amplicon \ (labeled TRAC-HDR) concentration by comparison to the product resulting from the uninfected control with genomic DNA isolated from human CD4+ T cells.3. PCR products were used for Nextera library preparation following the manufacturer's protocols \(Illumina).4. Prepped libraries were sequenced using 100-bp paired-end reads on an Illumina HiSeq 4000 instrument or equivalent.Indel quanti cation 1.Some PCR products from ampli cation around cutting site of genomic DNA \(same samples as T7E1 assay) were used for Nextera library preparation following the manufacturer's protocols \(Illumina).2. Prepped libraries were sequenced using 100-bp paired-end reads on an Illumina HiSeq 4000 instrument or equivalent \(generating 29 to 74 million reads per library).3. Map paired reads to amplicon sequences \ (expected sequences provided in FASTA form to generate indices) using BWA-MEM with the -M option.4. Discard 100bp reads in SAM le that fall outside a +/-75bp window of expected cut site within the amplicon.5. Discard soft-clipped reads \(identi ed with "S" character in CIGAR string).6. Identify indel reads by the presence of "I" or "D" characters within the CIGAR string.7. Quantify cutting e ciency as percentage of indels over total \(indel plus wild-type reads) within the de ned window.HDR quanti cation 1. Map reads to possible amplicons based on primer combinations and HDR status. 2. De ne "informative" amplicons as truncated so that 100bp reads would have at least 20bp homology with the CAR sequence \(or with the other TRAC arm, in the case of wild-type sequences).Informative reads can be used to distinguish wild-type, NHEJ and HDR reads with higher con dence.3. Map paired reads to amplicon sequences using BWA-MEM with -M ag to generate SAM les. 4. Use SAMtools to convert SAM les to BAM, sort, index, and generate summary statistics of read counts with the idxstats option.5. To quantify wild-type vs NHEJ reads, take reads that mapped to "info_nonHDR" sequence \(described below), and call reads with indels \(I" or "D" characters within the CIGAR string) as NHEJ.Otherwise call reads as wild-type.6. Pool read counts for downstream analysis.7. Schema for our amplicon sequences and quanti cations provided below: amplicon_nonHDR: refers to full amplicon from F1 and R1 of genomic, wild-type DNA.amplicon_CAR_F1: refers to full amplicon from F1 and R1 of expected, integrated CAR.amplicon_CAR_F2: refers to full amplicon from F2 \(primer site within the CAR as opposed to outside) and R1 of expected, integrated CAR.info_nonHDR same as amplicon_nonHDR, except truncated to 80bp of the TRAC arms.info_CAR_F1: same as amplicon_CAR_F1, except truncated to 80bp of the TRAC arms anking the TRAC-CAR interface.info_CAR_F2: same as amplicon_CAR_F2, except truncated to 80bp of the TRAC arms anking the TRAC-CAR interface \(relevant to the right arm only, since F2 is within the CAR sequence).HDR, NHEJ, and WT scores were calculated as follows: info_nonHDR = info_WT + info_NHEJ hdr_score = info_CAR_F2/\(info_CAR_F2+info_nonHDR) wt_score = info_WT/\(info_CAR_F2+info_nonHDR) nhej_score = info_NHEJ/\(info_CAR_F2+info_nonHDR) Co-culture functional assays Stable cell line generation 1. Generate lentivirus including GFP-Luciferase reporter genes.2. Infect NALM6 cells \(ATCC) with 2x concentrated lentivirus by spinoculation in retronectincoated \(Takara) plates at 800g for 45 mins at 32°C. 3.After infection for 2 days, sort GFP positive cells \ (NALM6-GL) by ow cytometry.4. Perform a second round of sorting after culturing for an additional two days.5. Incubate cells with 150g/ml D-Luciferin \(PerkinElmer) and measure bioluminescence signal intensity by an IVIS system to assess luciferase expression.Cancer cell cytolytic assay \(kill assay) 1. Seed 2104 NALM6-GL cells in a 96 well plate.2. Co-culture modi ed T cells with NALM6-GL at indicated E:T ratios for 24 hours.3. Add 150g/ml D-Luciferin \(PerkinElmer) into each well and measure luciferase assay intensity by a plate reader \(PerkinElmer) to assess cell proliferation.T cell exhaustion assay 1. Coculture T cells modi ed by AAV with NALM6-GL cells at 0.5:1 E:T ratio for 24 hours.2. Collect cells and wash once by DPBS.Incubate cells with 0.2 μg CD22-Fc \(R&D Systems) in 100 μL DPBS for 30 mins. 3.
Stain cells with PE-IgG-Fc, PD-1-FITC, TIGIT-APC and LAG3-Percp/cy5.5 \(Biolegend) for 30 mins. 4. Measure stained cells by ow cytometry.Intracellular staining of IFNγ and TNF-1.After infection for 5 days, co-culture AAV transduced CD22BBz CAR-T cells with NALM6 at 1:1 E:T ratio in fresh media supplemented with brefeldin A and 2 ng/mL IL-2. 2. After 5 hours of incubation, collect and stain for surface CAR.HEK293FT with high passage number.Use early passage of HEK293FT cells.3. Didn't concentrate virus enough.AAV puri ed from one 150mm*25mm petri dish of cells should be concentrated to less than 200µl.Low survival rate of cell after electroporation 1. T cell cultured for too long.Freshly isolated T cells are best, do not culture T cell for too long.2. Over-activation.Over-activation of T cells by CD3/CD28 Dynabeads could induce cell apoptosis.If using CD8 T cell, decrease activation time.3. Electroporation program didn't set well.The condition for Neon system is 1600V, 10ms, and three pulses.4. Multiple use of electroporation tips.Electroporation tips are better not to be used more than twice.5. Too much virus.Virus volume should not exceed 20%.Low cutting e ciency 1. Didn't do guide optimization.Design several guides for each gene and choose the best ones.2. mRNA degraded.mRNA should be stored at -80 °C, limit thaw and freeze cycles.3. Low dose of Cpf1 mRNA used.For 10µl reaction, should add 1µg of Cpf1 mRNA.4. T cell not active.If not active, stimulate T cells by CD3/CD28 antibody / beads before electroporation.5. Wrong buffer used.R buffer has higher electroporation e ciency than T buffer.6. Virus titer is too low.Concentrate AAV to smaller volume.Low HDR e ciency 1. Virus titer is too low.Concentrate AAV to smaller volume.2. HDR arm is too short.HDR arms should ideally not be shorter than 300bp.3. AAV vector design problem.If insertion site not at the rst exon or near a promoter, add the promoter in front of CAR.

Anticipated Results
See Dai et al.Methods, "One-step generation of modular CAR-T with AAV-Cpf1".

3 .
Fix and permeabilize cells by xation/permeabilization solution \(BD) and add anti-IFNγ-APC or anti-TNF--FITC for intracellular staining.4.After 30 mins, wash stained cells by BD Perm/Wash™ buffer and measure cells by ow cytometry.Timing Time Taken AAV vector construction: 1~2 weeks AAV production and puri cation: 4~5 days T cell electroporation and infection: 5 hours CAR-T expression and detection: 5+ days depending on experiment Cytolytic assay: 24 hours T cell surface marker or exhaustion ow assay: 24 hours T cell intracellular staining: 8 hours Troubleshooting Troubleshooting Problems Possible reasons Solution Low titer of AAV 1. Plasmid purity is too low.Plasmids possibly have RNA or protein contamination.Try to remove RNA and protein completely.2.

Figures
Figures