Cullin-5 deficiency promotes chimeric antigen receptor T cell effector functions potentially via the modulation of JAK/STAT signaling pathway

Abstract

Chimeric antigen receptor (CAR) T cell is a promising therapy for cancer, but factors that enhance the efficacy of CAR T cell remain elusive. Here we perform a genome-wide CRISPR screening to probe genes that regulate the proliferation and survival of CAR T cells following repetitive antigen stimulations. We find that genetic ablation of CUL5, encoding a core element of the multi-protein E3 ubiquitin-protein ligase complex, cullin-RING ligase 5, enhances human CD19 CAR T cell expansion potential and effector functions, potentially via the Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathway. In this regard, CUL5 knockout CD19 CAR T cells show sustained STAT3 and STAT5 phosphorylation, as well as delayed phosphorylation and degradation of JAK1 and JAK3. In vivo, shRNA-mediated knockdown of CUL5 enhances CD19 CAR T treatment outcomes in tumor-bearing mice. Our findings thus imply that targeting CUL5 in the ubiquitin system may enhance CAR T cell effector functions to enhance immunotherapy efficacy.

Fig. 1. Genome-wide clustered regularly interspaced short palindromic repeats (GW CRISPR) knockout (KO) screening of CD19 chimeric antigen receptor (CAR) T cells identified CUL5 as a candidate gene to enhance CAR T-cell survival.

a Outlines of the GW CRISPR KO screening of CD19 CAR T cells. After Day 10 of the culture, GW CRISPR KO-CD19 CAR T cells were repeatedly stimulated with γ-irradiated Raji cells. DNAs from CAR T cells at Days 10 and 40 were compared using next-generation sequencing. Stim, stimulation; EP, electroporation; tdx, transduction. b GFOLD score data and calculation. Genes with GFOLD greater than 2 were ranked. Individual ranking scores were added up from top to bottom (n = 3). c GFOLD score summary of three replicated data from three donors. The right panel depicts the expansion of the dotted area. d Secondary CRISPR KO screening of CD19 CAR T cells. Each of the six candidate genes was knocked out using two different sgRNAs. Data from experiments on three different donors are acquired and presented as the mean and standard error of the mean. e, f CUL5 KO using two different sgRNAs and its efficiency. The immunoblots show CUL5 expression in sgRNA-transduced CD19 CAR T cells (e). CUL5 to β-actin expression ratio in sgRNA-transduced CD19 CAR T cells (f). (n = 3, different donors) (g, h) Green fluorescent protein (GFP) positivity in different types of CUL5 KO-CAR T cells over time. CD19 CAR T cells (g) and CD37 CAR T cells (h) were stimulated by γ-irradiated tumor cells. Raji cells were used for Ag stimulation in both CD19 and CD37 CAR T cells. GFP-positive cells were sgRNA-transduced cells. (n = 4 for Ctrl sgRNA and CUL5 sgRNA#4690, n = 3 for CUL5 sgRNA#8377, different donors in g, n = 3, different donors in (h) Data were expressed as mean ± SEM. One-way ANOVA was used for comparing Ctrl and sgCUL5 (f). Two-way ANOVA with Tukey’s multiple comparisons test (with adjustment) was used for (g and h). Source data are provided as a Source Data file.

Fig. 2. CUL5 knockout (KO) promotes sustained proliferation and enhances effector functions later after stimulation.

a–d Changes in the percentage of CUL5 KO-CD19 CAR T cells generated by sgRNA#4690 from Days 10–20 (a) and 20–30 (b) or by sgRNA#8377 from Days 10–20 (c) and 20–30 (d). (n = 4 in a and b; n = 3 in c and d) (e) Representative flow plots of CellTrace Violet (CTV)-labeled control (Ctrl) or CUL5 KO-CD19 CAR T-cell division assays. GFP− fractions are unmodified CAR T cells, whereas GFP+ fractions are CUL5 KO-CAR T cells. f, g CTV mean fluorescence intensity (MFI) after stimulation at either earlier (f, Days 0–3, n = 3) or later time points (g, Days 4–7, n = 5) in Ctrl or CUL5 KO-CD19 CAR T cells. h, i Ki-67+ cell percentage after stimulation during earlier (h, Days 0–3, n = 4) or later (i, Days 4–7, n = 5) culture periods in Ctrl or CUL5 KO-CD19 CAR T cells. j Intracellular cytokine staining of Ctrl or CUL5 KO-CD19 CAR T cells stimulated with K562 or Raji cells. Representative flow plots of interferon-gamma (IFN-γ) and interleukin (IL)-2 are shown. k, l Intracellular cytokine staining for IFN-γ (k) and IL-2 (l). (n = 5) (m) CD25 MFI after Raji stimulation in Ctrl or CUL5 KO-CD19CAR T cells. Data were expressed as mean ± SEM. (n = 3) (n) Phenotype analysis of Ctrl and CUL5 KO-CD19 CAR T cells for CD45RA and C-C chemokine receptor 7 (CCR7) on Day 30. o, p Percentages of CD45RA( − )-CCR7(−) effector memory T cells and CD45RA( − )-CCR7(+) central memory T cells after CD3/CD28 beads stimulation. q, r Percentages of CD45RA( − )-CCR(−) effector memory T cells and CD45RA( − )-CCR7(+) central memory T cells after Raji cell stimulation. s CCR7 MFI after Raji stimulation. (n = 5 in (o–s) Data from three to five different donors are acquired and each dot represents data from each donor. The paired two-sided Student’s t-test was used for (a–d, f–i, and o–s). The two-way ANOVA with the Šidák multiple comparisons test was used for (k and l). The multiple unpaired t-test was used for (m). Source data are provided as a Source Data file.

Fig. 3. CUL5 knockout (KO) CD19 chimeric antigen receptor (CAR) T cells upregulate the Janus kinase/signal transducers and activators of the transcription (JAK/STAT) pathway.

a Numbers of differentially expressed genes (DEGs) in control (Ctrl) and two different sgRNA-modified CUL5 KO-CD19 CAR T cells (Ctrl: n = 4, sgRNA#4690: n = 4, and sgRNA#8377: n = 3). b Venn diagram of DEGs in CUL5 KO-CD19 CAR T cells prepared using two different sgRNAs. c Heatmap displaying the significantly upregulated and downregulated genes in comparisons between CUL5KO-CD19CAR T cells treated with sgRNA#4690 and Ctrl CAR T cells. Each column represents the respective donor’s result. d Over-representation analysis (hallmark gene sets) of upregulated genes in CUL5 KO-CD19 CAR T cells versus Ctrl sgRNA-transduced CD19 CAR T cells. The p-values were all adjusted by Benjamini–Hochberg false discovery rate (FDR) procedure. –log₁₀ (adjusted p-value) is shown.

Fig. 4. CUL5-knockout (KO) CD19 chimeric antigen receptor (CAR) T cells demonstrate intracellular signaling augmentation via phospho-STAT3 (p-STAT3) and p-STAT5.

a, c Representative flow plots of intracellular p-STAT3 (a) and p-STAT5 (c). Truncated epidermal growth factor receptor (tEGFR)-transduced T cells were stimulated with anti-CD3/CD28 beads, whereas control (Ctrl) and CUL5 KO-CD19 CAR T cells were stimulated with Raji cells in a 1:5 ratio. The data represent five and three independent experiments on different donors, respectively in (a and c). b The mean fluorescence intensity (MFI) of p-STAT3 at the indicated time after stimulation. (n = 5, different donors) (d) The MFI of p-STAT5 at the indicated time after stimulation. (n = 3, different donors) (e) p-STAT5+ percentage at the indicated time after stimulation with Raji cells in medium without interleukin (IL)-2 supplementation. (n = 3, different donors) Each dot represents data from each donor. Note that the two-way ANOVA with the Šidák multiple comparisons test was used for (b, d, and e). Source data are provided as a Source Data file.

Fig. 5. CUL5 is involved in Janus kinase 3 (JAK3) degradation in response to interleukin (IL)-2 signaling-mediated activation.

a A two-in-one lentiviral vector construct in which CUL5 shRNA and CD19 chimeric antigen receptor (CAR) are linked. b The efficiency of CUL5 knockout (KO) induced by shCUL5-CD19 CAR gene transfer into CD8 + T cells from healthy donors. c CUL5 to β-actin expression ratio in shGFP- and shCUL5-CD19 CAR T cells. Data are presented from three independent experiments on three different donors. d–f The expression of unphosphorylated and phosphorylated Janus kinase/signal transducers and activators of transcription (JAK-STAT) proteins was estimated in activated shGFP- and shCUL5-CD19 CAR T cells. The activated cells were harvested four days after co-culturing in a 1:5 E: T ratio with γ-irradiated Raji cells. The JAK/STAT protein to β-actin expression ratio in activated shGFP- and shCUL5-CD19 CAR T cells is shown in (e) and (f). Data are presented from three independent experiments on three different donors. g, h JAK1, phospho-JAK1 (p-JAK1), JAK3, and p-JAK3 expressions in wild type (WT)-KHYG-1 and shCUL5-KHYG-1 cells. KHYG-1 cells were stimulated with IL-2 (200 IU/ml) after overnight IL-2 deprivation (g). In addition to IL-2 administration, CHX was added at a concentration of 40 μg/ml for the indicated times (h). i Data on CUL5 and CUL5-related adapter protein KO-CD8 + T cells over time. Cells were prepared in the same manner as CUL5 KO cells from primary human CD8 + T cells and stimulated with anti-CD3/CD28 beads every 10 days. (n = 5, different donors) Fig. 5g and h are from one experiment representative of three independent experiments. Data were expressed as mean ± SEM. Statistical significance was determined using the two-sided Student’s t-test for c and the two-way ANOVA with the Šidák multiple comparisons test was used for (e, f, and i. Source data are provided as a Source Data file.

Fig. 6. In vivo efficacy of CUL5 knockout (KO)- and knockdown (KD)-CD19 CAR T cells.

a A schematic diagram of in vivo experiments. Nonobese diabetic/severe combined immunodeficiency (NOD/SCID) common-gamma chain KO (NOG) mice were injected with 5 × 10⁵ Raji/ffluc-green fluorescent protein (GFP) cells via the tail vein on Day 0, followed by 1  × 10⁶ truncated epidermal growth factor receptor (tEGFR)-, control (Ctrl)-, or CUL5 KO-CD19 CAR T cells on Day 7. Tumor burden was assessed using BLI at the indicated time points and weekly thereafter. b Representative BLI of Raji-inoculated mice treated with Ctrl- or CUL5 KO-CD19 CAR T cells over time. c Tumor burden in mice treated with Ctrl- or CUL5 KO-CD19 CAR T cells at the indicated time points. Data are presented as the mean and SEM, compared using the two-way ANOVA, and acquired from three independent experiments on three different donors (tEGFR-T cells: n = 4; Ctrl- and CUL5 KO-CD19 CAR T cells: n = 9 per group). d Kaplan-Meier survival analysis of Raji/ffluc-bearing NOG mice (log-rank test). e The number of CAR T cells derived from 1 × 10⁶ CD8 + T cells on Day 7. Data are derived from three independent experiments on three different donors. Data were expressed as mean ± SEM. f The effect of CUL5 KD on activated CD19 CAR T cell proliferation. These cells were stimulated with γ-irradiated Raji cells in a 1:5 E: T ratio every 10 days. Data were expressed as mean ± SEM. (n = 3, different donors) Note that the paired t-test was used for e, whereas the one-way ANOVA was used for (f). g, h In vivo efficacy assessment using a subcutaneous tumor model. On Day 0, Raji/ffluc-GFP cells were subcutaneously inoculated. On Day 10, 1 × 10⁶ tEGFR-, shGFP-, or shCUL5-CD19 CAR T cells were injected. g Subcutaneous tumor volume. The curve was censored when any of the mice in the group were euthanized. Data are presented as the mean and SEM, compared using the two-way ANOVA, and acquired from three independent experiments on three different donors (n = 9 per group). h Kaplan-Meier analysis of NOG mice subcutaneously injected with Raji/ffluc cells (log-rank test). (i) Kaplan-Meier analysis of NOG mice intravenously injected with Raji/ffluc cells as in (a). 1 × 10⁶ tEGFR-T, CUL5 KO-, or shCUL5-CD19 CAR T cells were injected. Data are acquired from three independent experiments on four different donors (tEGFR-T cells: n = 8; CUL5 KO- and shCUL5-CD19 CAR T cells: n = 12 per group) (log-rank test). Source data are provided as a Source Data file.