Resource Optimizing CRISPR/Cas-Mediated CAR Knockin in Primary Human T Cells Using MaxCyte Electroporation

Poster

Optimizing CRISPR/Cas-Mediated CAR Knockin in Primary Human T Cells Using MaxCyte Electroporation

Abstract

Non-viral gene knockin in T cells

Improving knockin with HDR enhancers, small template DNA and CTS

Engineering CAR T cells

Engineering functional and effective CAR T cells

Optimizing electroporation protocols

Summary

Download as a PDF

Abstract

In recent years, chimeric antigen receptor (CAR) T cells have emerged as a leading treatment for various hematological cancers. Despite demonstrating great promise, there are several challenges that limit CAR T cell therapies, including concerns regarding efficacy, safety and manufacturability.

To address these, groups have turned to non-viral engineering of CAR T cells using CRISPR gene editing. In addition to reduced immunogenicity and manufacturing costs, this technology enables precise knockin of tumor-targeting receptors and knockout of genes responsible for rejection, toxicity and immunosuppression. To CRISPR engineer CAR T cells, Cas nucleases, guide RNA (gRNA) and homology-directed repair (HDR) template, DNA must be efficiently delivered into T cells without jeopardizing their viability or functionality.

With this in mind, we sought to generate optimized workflows that enable CRISPR-mediated gene editing in primary human T cells using MaxCyte® electroporation.

To this end, we first identified optimal electroporation protocols and concentrations of sgRNA, Cas9 and HDR template required to knock in GFP at the TRAC locus. Next, we compared different repair template designs, including the addition of Cas9 targeting sequences (CTS). We also investigated multiple HDR template DNA formats, including conventional plasmids and Nanoplasmids™. In addition, we explored the use of small molecule enhancers and identified several that further improved editing efficiencies.

These findings were then used to engineer CD19-targeting CAR T cells. With activated T cells from healthy donors, we could achieve CAR expression levels of greater than 70%. In addition to efficient and reproducible CAR knockin in T cells from multiple donors, cells engineered using this workflow were viable and retained the ability to expand and eradicate CD19-expressing target cells.

Together, these demonstrate the capability of MaxCyte’s clinically validated ExPERT™ electroporation technology to enable non-viral engineering of CAR T cells using the CRISPR/Cas system.

MaxCyte enables non-viral gene knockin in activated T cells

A) Viable T cells at 24 & 48 hours

Bar graph illustrates > 40% of viable T cells at 24 hours post electroporation, across EP protocols and plasmid amounts and >60% at 48 hours.

B) Efficiencies at 24 & 48 hours

Bar graph illustrates >60% of transfection efficiencies at 24 hours post EP, across protocols and plasmid amounts and slightly higher at 48 hours.

C) GFP expression at 24 & 48 hours

Bar graph illustrates ​>between 0.E+00 and 5.E+04 M​FI at 24 hours post electroporation, across EP protocols and plasmid amounts​, and ​s​lightly lower at 48 hours.​ Also shows steady increase of MFI with greater plasmid amounts.

D) GFP expression at 24 & 48 hours

Flow cytometry histograms show​ same ​G​FP expression for three plasmid amounts at the two time points as Figure 1C.

E) Viabilities at days two, four & seven

The bar graph ​illustrates ​c​ell viabilities (expressed as a percentage via AO/PI) ​at three time points under various conditions of cargo concentration and different ratios of Cas9 and sgRNA, generally increasing over time from​ across most conditions​.​

F) TRAC KO efficiencies at days two, four & seven

​B​ar graph​ shows TRAC knockout efficiencies in viable T cells ​a​t three time points across varying cargo concentrations and ​E​P conditions, with efficiencies generally trending higher with higher concentrations and later time points.

G) TRAC expression at days two, four & seven

Bar graph shows TRAC expression, measured by MFI, ​tends to decrease over time, with varying effects of cargo concentration compared to the ​control condition (no EP) at each time point.

H) TRAC KO & GFP KI at days four & seven

Eight heat maps illustrate the percentage of GFP-positive cells expressing TCRα/β under varying concentrations of a plasmid HDRT at two time points, with percentages increasing with higher concentrations of HDRT and thereby suggesting a dose-dependent effect on expression. 

I) Viabilities at days four & seven

​B​ar graph​ shows relatively stable cell viabilities, measured via AO/PI (%), ​a​t two time points across different experimental conditions, including a control group ("No EP") and​ three concentrations of​ plasmid HDRT.

J) TRAC KO efficiencies at days four & seven

​B​ar graph demonstrates that increasing concentrations ​of p​lasmid HDRT significantly improve TRAC knockout efficiencies in viable T cells compared to no​ EP, with high efficiencies maintained​ at two time points.

K) GFP KI efficiencies at days four & seven

​Bar graph illustrates increasing GFP ​k​nockin efficiencies, measured as ​% TCR-GFP+ out of viable T cells, with higher concentrations of ​p​lasmid HDRT compared to ​n​o EP, with slightly higher efficiencies ​a​t later time point for each concentration condition.

L) GFP expression at days four & seven

​Bar graph illustrates​ increasing GFP expression, measured by MFI, ​a​s​ concentrations of​ plasmid HDRT increase, with slightly higher ​expression at earlier time point across all conditions.

Figure 1: CD3+ T cells were isolated from healthy donors and activated for 48 hours. Activated T cells were transfected with 50, 100 or 200 µg/mL of a GFP-encoding plasmid using the ExPERT GTx® and Expanded T cell 3 (ETC3) or Expanded T cell 4 (ETC4) to determine an appropriate electroporation (EP) protocol for template DNA delivery (A-D). Viable T cells (A), transfection efficiencies (B) and mean fluorescent intensity (MFI) (C, D) were determined by flow cytometry at 24 and 48 hours post EP. To determine conditions for TRAC knockout (KO), varying molar ratios of Cas9 and sgRNA (1:0.5, 1:1, 1:2 and 1:3) were complexed to form ribonucleoproteins (RNP). Activated T cells were then transfected with different final RNP concentrations (0.5, 1.0, 2.0 and 4.0 µm) using ETC4 (E-G). Viabilities (E), TRAC KO efficiencies (F) and TRAC expression (G) were determined using flow cytometry. For GFP knockin (KI), optimal concentrations of HDR template (HDRT) were determined by co-transfecting activated T cells with 0.5 µm of RNP (1 Cas9:2 sgRNA) and either 50, 100 or 200 µg/mL of a plasmid HDRT for GFP KI and viabilities, (I) TRAC KO (H, J), GFP KI (H, K) and GFP expression (L) were measured at day four and day seven post EP.

HDR enhancers, small template DNA and addition of CTS improve knockin

A) GFP KI workflow

​Illustrated workflow fo​r engineering T cells, involving initial activation, preparation of a master mix with RNP and DNA, electroporation using the ​MaxCyte ExPERT GTX ​p​latform for gene editing, subsequent T cell expansion, and downstream analysis for viability and gene editing outcomes like knockout and knockin.

B) Viable T cells at days two & five

The bar graph illustrates ​r​elatively stable percentage​ (~50%) of viable T cells under various ​c​o-transfection conditions​ by the second time point recorded.

C) TRAC KO efficiencies at days two & five

The ​bar graph illustrates TRAC knockout​ efficiency in​ viable T cells under various experimental conditions ​a​t two time points, as measured by the percentage of TCR-negative ​T cells, with efficiencies >95% by the later time point.

D) GFP KI efficiencies at days two & five

​Bar graph ​s​hows the percentage of GFP+TCR- out of viable T cells, which generally increase​s ​over time across different ​experimental conditions, with npGFP + CTS consistently showing the highest ​GFP KI efficiency.

E) GFP expression at days two & five

​Bar graph illustrates increasing GFP expression, measured by MFI, ​across two time points, with the addition of CTS generally enhancing expression for both pGFP and npGFP constructs, and npGFP constructs consistently yielding higher expression than pGFP constructs.

F) Viable T cells at day four

​B​ar graph illustrates ​close ​to 50 percent of viable T cells ​a​t one time point with enhancers ​at different concentrations compared to without enhancers and without EP, both of which yielded a higher percentage of viable cells (>55%).

G) TRAC KO efficiencies at day four

Bar graph illustrates nearly 100% efficiency of TCR- viable T cells at one time point with and without enhancers at all concentrations tested, by contrast to only 5% efficency for control condition (no EP).

H) GFP KI efficiencies at day four

​Bar graph illustrates ​increased GFP ​knock​in efficiencies, represented as the percentage of TCR-GFP+ ​T cells​, ​at higher concentrations of both enhancers compared to no enhancer and no EP for one time point.

I) GFP expression at day four

​B​ar chart displays GFP expression ​v​ia MFI ​a​t one time point ​w​ith and without enhancers at different concentrations​; ​increasing enhancer concentration generally led to increased expansion compared to controls.

J) Viable T cells at days four & seven

B​ar graph illustrates viable T cells (% 7-AAD- T cells out of total events) under different experimental conditions​ at ​two time points over week period, with higher viability for "npGFP" and "npGFP + CTS" conditions​ and, for all conditions, at day seven.

K) TRAC KO efficiencies at days four & seven

​B​ar graph illustrates consistent >90% TRAC gene knockout​ efficiency in T cells at​ two time points over one week​ for all experimental conditions compared to control.

L) GFP KI efficiencies at days four & seven

​This bar graph indicates that the highest GFP KI efficiencies were consistently observed with the npGFP + CTS condition at both recorded time points.

M) GFP expression at days four & seven

​Bar graph​ shows npGFP + CTS consistently yielding the highest GFP expression at both time points, and higher expression levels in general on day seven.

N) GFP expression at days four & seven

​T​he flow cytometry histogram ​depicts changes in GFP expression across different cell populations ​b​etween the two time points, as indicated by shifts in the fluorescence intensity distributions.

Figure 2: The workflow for GFP KI established in Figure 1 (A) was further improved. Activated T cells were co-transfected with 0.5 μM or RNP and 200 µg/mL of plasmid (pGFP), plasmid with a Cas9-targeting sequence (CTS) (pGFP + CTS), Nanoplasmid™ (npGFP) or Nanoplasmid with a CTS (npGFP + CTS) HDRT (B-E). Viable T cells (B), TRAC KO (C), GFP KI (D) and MFI (E) were measured at days two and five post EP and demonstrated that smaller HDRTs as well as a CTS improved KI. Similarly, activated T cells were co-transfected with 0.5 μM of RNP and 200 µg/mL of pGFP and samples were split into 7 wells and received either no enhancer of 0.5, 1 or 2 μM of M3814 or AZ7648 (F-I). Viable T cells (F), TRAC KO (G), GFP KI (H) and MFI (I) were measured at day four. Both enhancers improved KI in a dose-dependent manner, but 2 μM M3814 improved KI the most. These two strategies were then combined and found to further enhance GFP KI in activated T cells (J-N).

MaxCyte reproducibly engineers CAR T cells without impacting expansions

A) Viable T cells at days four & seven

​The bar graph suggests that T cell viability, as measured by the percentage of 7-AAD-T cells out of total events, is generally maintained or slightly improved in the presence of 2 uM M3814 in all three donors at both time points.

B) TRAC KO efficiencies at days four & seven

This bar graph shows a significant increase of TRAC knockout efficiency with pCAR and npCAR co-transfections compared to the control, with the addition of M3814 further enhancing efficiency, at both recorded time points for all three donors.

C) CAR KI efficiencies at days four & seven

Bar graph compares CAR knockin efficiencies across three conditions, with and without the addition of M3814, for three donors at two time points, ultimately showing that pCAR and npCAR methods significantly increase efficiency.

D) CAR expression at days four & seven

Bar chart illustrates the addition of M3814 significantly and consistently increases CAR expression (MFI), particularly in the pCAR and npCAR conditions, across donors and time points.

E) Viabilities over 14-day expansion

This line graph illustrates increasing viabilities (via AO/PO %) of different cell treatment groups over time, with and without M3814; the pCAR group consistently shows the highest viability among the treated groups in the presence of M3814.

F) Viable cell counts over 14-day expansion

Line graph compares how M3814 influences viable cell counts (x1x10^6) across different transfection conditions over two-week period, with cell counts generally rising over time with and without the addition of M3814.

G) TRAC KO over 14-day expansion

This line graph shows about >90% TCR out of viable T cells for all experimental conditions over time, regardless of the presence of M3814, indicating its minimal effect on TRAC KO efficiency.

H) CAR KI over 14-day expansion

This line graph demonstrates generally higher and increasing CAR knockin efficiencies in the presence of CTS, especially in the presence of M3814, across time points.

I) CAR+ T cell counts over 14-day expansion

In this line graph, all conditions, minus the control, show an increase in CAR+ T cell counts over time, with npCAR + CTS consistently demonstrating the highest cell counts in both the absence and presence of M3814.

J) CAR expression over 14-day expansion

optimizing-crispr-cas-CAR_fig3-J

Figure 3: Optimized conditions for GFP KI were used to engineer CD19 CAR T cells using the ExPERT GTx. Activated T cells from three healthy donors were transfected with 0.5 μM of RNP and 200 μg/mL of plasmid (pCAR) or Nanoplasmid (npCAR) HDRT. Samples were split in half, and one half received 2 μM of M3814 post EP (A-D). Viable T cells (A), TRAC KO efficiencies (B), CAR KI efficiencies (C) and CAR expression (D) were measured at days four and seven post EP and demonstrate that MaxCyte can be used to reproducibly engineer CAR T cells in multiple donors. Activated T cells from the donor achieving the highest CAR KI (Donor 3) were then transfected with 0.5 μM of RNP and 200 μg/mL of different HDRTs for CD19 CAR KI including pCAR, pCAR with a CTS (pCAR + CTS). Samples were split in half, and one half received 2 μM of M3814 post EP. Cells were then expanded for 14 days (E-J). Viabilities (E), viable cell counts (F), TRAC KO efficiencies (G), CAR KI efficiencies (H), CAR+ T cell counts (I) and CAR expression (J) were measured at day four, seven, 11 and 14, and CAR T cells engineered using MaxCyte retained their ability to expand.

CAR T cells engineered using MaxCyte are functional and effective

A) Viabilities at days four, seven, 11 & 14

Bar graph shows cell viability (via AO/PI%) under three different conditions over 14 days, with npCAR maintaining highest, above 90%, at four time points.

B) TRAC KO efficiencies at days four, seven, 11 & 14

Bar chart indicates that both npCAR and npCAR + M3814 conditions consistently achieved near 100% TRAC KO efficiency over two-week period compared to control, which was under 10%.

C) CAR KI efficiencies at days four, seven, 11 & 14

Bar graph shows CAR knockin efficiency, measured as % TCR-CAR+ cells, generally increases over time for all conditions, with npCAR + M3814 consistently demonstrating the highest efficiency across all measured time points in two weeks.

D) CAR expression at days four, seven, 11 & 14

optimizing-crispr-cas-CAR_fig4-d

E) Target cell killing

This graph shows that npCAR consistently yields the highest relative killing percentage, significantly outperforming the M3814 condition, especially at lower E/T ratios, and maintaining high killing efficiency across all tested ratios.

F) Granzyme B

Bar graph illustrates how Granzyme B expression, measured via % GzB+ in CD8+ T cells, changes with increasing E/T ratios under three experimental conditions, with npCAR +M3814 remaining the highest.

G) IFN-γ

Three flow cytometry plots compare cells based on Side Scatter (SSC) and Interferon-gamma (IFN-γ) levels under three conditions. A consistent gate is applied across the plots to highlight a specific cell population, showing changes in the IFN-γ positive cell population between the npCAR and npCAR + M3814 conditions.

Figure 4: To assess the functionality of CAR T cells engineered using MaxCyte, 30 million activated T cells were co-transfected with 0.5 μM of RNP and 200 μg/mL of npCAR HDRT. CAR KI samples were split in half, and one half received 2 μM of M3814 post EP. Cells were then expanded for 14 days. Viabilities (A) TRAC KO efficiencies (B), CAR KI efficiencies (C) and CAR expression (D) were examined at days four, seven, 11 and 14. At day 14 post EP, CAR T cells were co-cultured with CD19+ target cells (Ramos) at varying effector-to-target (E/T) cell ratios (0.25:1, 0.5:1, 1:1. 2:1) (E-G). Target cell killing was determined via flow cytometry at 24 hours and was found to be antigen-specific and increased with higher E/T ratios (E). CD8+ CAR T cells also produced Granzyme B (GzB) in response to target cells, with those cells co-cultured at lower E/T ratios having greater intracellular GzB at 24 hours (F). CD8+ CAR T co-cultured with Ramos cells at a 1:1 E/T ratio were also shown to produced IFN-γ after 24 hours of co-culture (G), demonstrating that these cells retain their cytotoxic functions and are effective at eradicating target tumor cells.

Optimizing electroporation protocols to further improve gene knockin

A) Viable T cells

Bar graph illustrates ​higher percentage of 7-AAD- T cells out of total events​, or greater viability, for ETC4 and EP protocols one and two compared to the other two protocols, which show notable decrease.

B) TRAC KO efficiencies

Bar graph demonstrates significantly higher TRAC KO efficiencies for protocols 1-4 demonstrate, consistently achieving over 90% TCR-negative viable T cells, compared to no EP.

C) GFP KI efficiencies

​Bar graph demonstrates a progressive increase in ​GFP knockin efficiency​, measured via percentage of TCR-GFP+ cells out of viable T cells​, from "No EP" and "ETC4" through "Protocol 1" to "Protocol 4."

D) GFP expression

In this bar graph, GFP expression, measured in MFI, consistently rises with each subsequent protocol after ETC4, indicating enhanced expression with optimized protocols.

Figure 5: To further improve gene KI in activated T cells without the use of small molecule enhancers, EP protocols were further optimized. Activated T cells were co-transfected with 0.5 μM of RNP and 200 μg/mL of pGFP HDRT using ETC4 or one of four modified protocols: Protocol 1, Protocol 2, Protocol 3 or Protocol 4 (A-D). Viable T cells (A), TRAC KO efficiencies (B), GFP KI efficiencies (C) and GFP expression (D) were measured at day four post EP via flow cytometry. GFP KI efficiencies as well as GFP expression could be improved with modified EP protocols without the use of M3814 but improved gene editing occasionally came at the cost of viable T cells. Efforts are ongoing to design optimized EP protocols that ensure high gene editing without compromising cell viability or functionality using the ExPERT GTx.

Related resources

Thumbnail for Optimizing CRISPR/Cas-mediated CAR Knockin in Primary Human T Cells Using MaxCyte presentation
Presentation
MaxCyte Electroporation Enables CRISPR/Cas-Mediated CAR Knockin in Primary Human T cells
The left of the illustration shows an image of a human figure and IV bag to represent harvest cells from healthy donor and the right shows an image of the ExPERT GTx®. Between them is a depiction of CRISPR RNP + ssCTS template being delivered to the TRAC locus of activated T cells.
Poster
Highly Efficient Engineering of Difficult-to-Transfect Immune Cells Using MaxCyte Electroporation
View all resources