Reshaping CAR-T cells through overexpression of T cell factor 1

Abstract

Introduction: Although chimeric antigen receptor (CAR) T cell therapy has revolutionized treatment for hematologic malignancies, insufficient CAR-T cell persistence remains a major limitation. T cell factor 1 (TCF-1) is a transcription factor crucial for T cell development, self-renewal, and memory formation. However, CAR-T cells typically exhibit low TCF-1 expression. This study investigated whether restoring TCF-1 expression could enhance CAR-T cell persistence and functionality.

Methods: Human peripheral blood T cells were transduced with third-generation CD19 or CD33 CAR retroviral vectors, with or without a TCF-1 (Tcf7.NGFR) construct. Phenotypic, functional, and transcriptional analyses were performed using flow cytometry, cytokine profiling, long-term killing assays, and RNA sequencing. Data mining and machine learning were applied for high-dimensional immunophenotyping.

Results: TCF-1 overexpression generated CAR-T cells with reduced apoptosis, lower activation marker expression, and an increased proportion of naïve and stem cell-like subsets. These modified cells displayed a higher CD4⁺/CD8⁺ ratio, preserved proliferative capacity, and maintained cytotoxicity with attenuated cytokine release. Long-term co-culture assays demonstrated superior persistence and sustained tumor-killing activity in TCF-1-overexpressing CAR-T cells. Transcriptomic profiling revealed downregulation of apoptotic and cytokine release pathways, and enrichment of cell cycle and metabolic pathways supporting T cell longevity.

Discussion: Overexpression of TCF-1 confers resistance to apoptosis, limits excessive activation, and promotes a less differentiated phenotype, collectively enhancing CAR-T cell persistence and long-term efficacy. These findings suggest that TCF-1 modulation represents a promising strategy to improve durability and safety of CAR-T cell therapy in relapsed or refractory hematologic malignancies.

Keywords: CAR (chimeric antigen receptor) T cells; CRS - cytokine release syndrome; T cell persistence; TCF (T-cell factor); immunotherapy.

Figure 1

Effect of TCF-1 overexpression on CAR expression, cell expansion, apoptosis, cell component of CD19.CAR-T cells. (A) Representative figure of CD19.CAR-T cells and DT19.CAR-T cell generation (n=3). (B) Representative dolt plots of CD19.CAR and NGFR expression on CD19.CAR-T cells and DT19.CAR-T cells, and statistical analysis of CD19.CAR and NGFR expression on non-transduced T cells, CD19.CAR-T cells and DT19.CAR-T cells (n=3). (C) Statistical analysis protein level of TCF-1 on non-transduced T cells, CD19.CAR- T cells, and DT19.CAR-T cells, as detected by Western Blot (n=3). (D) Proliferation dynamics of CD19.CAR-T cells and DT19.CAR-T cells from Day 3 to Day 11 (n=6). (E) The expression of protein level of caspase 3, cleaved caspase 3 and PARP in non-transduced T cells, CD19.CAR-T cells and DT19.CAR-T cells (n=3). (F) Statistical analysis of the protein level of caspase 3, cleaved caspase3, and PARP in non-transduced T cells, CD19.CAR-T cells and DT19.CAR-T cells (n=3). (G) Expression of Ki67 on non-transduced T cells, CD19.CAR-T cells and DT19.CAR-T cells (n=6). (H) CD4/CD8 composition in CD19.CAR-T cells and DT19.CAR-T cells (n=6). (I) T cell subsets in CD19.CAR-T cells and DT19.CAR-T cells (n=6). A paired t-test was used for statistical analysis. (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns= no significant difference).

Figure 2

Effect of TCF-1 overexpression on cell surface markers and cytokine release on CD19.CAR-T cells during generation. (A–G) Dynamic expression of Apotracker, CD95, CD253, CD57, CD69, CD27, and CD40L on CD19.CAR-T cells and DT19.CAR-T cells at day5, day8, day11, and day14 by flow cytometry (n=6). (H–N) Dynamic detection of IL17A, sFasL, TNF-α, Granzyme A, Granulysin, and Perforin on CD19.CAR-T cells and DT19.CAR-T cells from day5, day8, and day 14 by flow cytometry (n=3). A paired t-test was used for statistical analysis. (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns= no significant difference).

Figure 3

Effect of TCF-1 overexpression on frequency of CD19.CAR-T cell populations. (A) t-SNE plot of phonograph identified 23 cell clusters in CD19.CAR-T cells and DT19.CAR-T cells with FACS panel 1 (n=6). (B) Evaluation of distinguishing capability of cell cluster by PCA with FACS Panel1. (C) Statistic analysis of percentage of CD4⁺Apotracker⁺CD95⁺CD253⁺ and CD8⁺Apotracker⁺CD95⁺CD253⁺ in CD19.CAR-T cells and DT19.CAR-T cells. (D) t-SNE plot of phonograph identified 18 cell clusters in CD19.CAR-T cells and DT19.CAR-T cells with FACS panel 2 (n=6). (E) Evaluation of distinguishing capability of cell cluster by PCA with FACS Panel 2. (F) Statistic analysis of percentage of CD4⁺CD62L⁺ in CD19.CAR-T cells and DT19.CAR-T cells. (G) t-SNE plot of phonograph identified 27 cell clusters in CD19.CAR-T cells and DT19.CAR-T cells with FACS panel 3 (n=6). (H) Evaluation of distinguishing capability of cell cluster by PCA with FACS Panel 3. (I) Statistic analysis of percentage of CD4⁺CD27^-CD57⁺CD69^-, CD4⁺CD27^-CD57^-CD69⁺, CD4⁺CD27^-CD57⁺CD69⁺, CD4⁺CD27⁺, CD3⁺CD27⁺CD57^-CD69^- in CD19.CAR-T cells and DT19.CAR-T cells. (J) t-SNE plot of phonograph identified 18 cell clusters in CD19.CAR-T cells and DT19.CAR-T cells with FACS panel 4 (n=6). (K) Evaluation of distinguishing capability of cell cluster by PCA with FACS Panel 4. (M) Statistic analysis of percentage of CD4⁺CD40L⁺ in CD19.CAR-T cells and DT19.CAR-T cells. PCA was performed to identify the major principal components (PCs) that contributed at least 70% of the total variance for day11 and day14. The contribution of each variable (cell subset frequency) in accounting for the variability in these principal components was further analyzed. FACS panel 1: CD3, CD4, Apotracker, CD95, CD253, 7AAD; FACS panel 2: CD3, CD4, CD45RA, CCR7, CD62L, CXCR3, 7AAD; FACS panel 3: CD3, CD4, CD8, CD27, CD57, CD69, 7AAD; FACS panel 4: CD3, CD4, CD40L, CTLA4, ICOS, 7AAD. A paired t-test was used for statistical analysis. (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns= no significant difference).

Figure 4

Effect of TCF-1 overexpression on short-term killing efficiency of CD19.CAR-T cells. (A) Representative dot plots of percentage of CD107a, TNF-α, IFN-γ on CD19.CAT-T cells and DT19.CAR-T cells. Statistical analysis of percentage and mean fluorescence intensity (MFI) of CD107a, TNF-α, IFN-γ. CD19.CAR-T cells and DT19.CAR-T cells were stimulated by Nalm6 cells for 4 hours. Intracellular cytokine staining was applied to detect the cytokine release. (B) Characterization of functional CD19.CAR-T and DT19.CAR-T cell subsets. Based on the expression of CD107a, TNF-α, IFN-γ, T cells can be defined as six functional subsets: CD107a⁺TNF-α⁺IFN-γ⁺, CD107a⁺TNF-α⁺IFN-γ^-, CD107a⁺TNF-α^-IFN-γ⁺, CD107a⁺TNF-α^-IFN-γ^-, CD107a^-TNF-α⁺IFN-γ⁺, CD107a^-TNF-α⁺IFN-γ^-, CD107a^-TNF-α^-IFN-γ⁺, CD107a^-TNF-α^-IFN-γ^- (n=6). (C) Statistic analysis of six functional subsets (n=6). (D) Representative dot plots (left) and statistical analysis of killing efficiency of CD19.CAR-T cells and DT19.CAR-T cells after 24 hours in E:T ratio of 1:1 and 1:2 (right). CD19.CAR-T cells and DT19.CAR-T cells were co-cultured with Nalm6 cells respectively for 24 hours, residual tumor cells were detected by flow cytometry after 24 hours.

Figure 5

Effect of TCF-1 overexpression on proliferation and long-term cytotoxicity of CD19.CAR-T cells in the co-culture assay. (A) Representative dot plots of CD19.CAR-T cell and DT.CAR-T cell proliferation and tumor lysis stimulated by Nalm6 cells after 9 days and 15 days with E: T 1:1, respectively. (B) Statistical analysis of Nalm6 cell lysis from day1 to day 21 with E: T 1:1. Darker color (blue) indicates a better tumor lysis, while brighter color (white) indicates a worse tumor lysis. (C) Statistical analysis of CD19.CAR-T cell and DT19.CAR-T cell proliferation from day1 to day 21 with E: T 1:1. Darker color (orange) indicates a better cell expansion, while brighter color (white) indicates a worse cell expansion. (D) Dynamic changes of cell subsets during co-culture assay. Dynamic change of t-SNE plot of CAR⁺CD4⁺PD1^-Tim3^-Lag3^-, CAR^dimCD8⁺PD1^-Tim3^-Lag3^-, and CAR⁺CD8⁺PD1^-Tim3^-Lag3^- cell subsets at day1, day5, day9, and day15 (left panel). Statistical analysis of CAR⁺CD4⁺PD1^-Tim3^-Lag3^-, CAR^dimCD8⁺PD1^-Tim3^-Lag3^-, and CAR⁺CD8⁺PD1^-Tim3^-Lag3^- cell subsets from day1 to day12.

Figure 6

Effect of TCF-1 overexpression on CD19.CAR-T cells at the transcriptional level. (A) Volcano map of differentially expressed genes (DEGs) between CD19.CAR-T cells and DT19.CAR-T cells. DEGs were defined as |Log2FC| ≥ 1 and Q value ≤ 0.05. (B) KEGG pathway analysis of DEGs, showing the top 15 enriched pathways. The color indicates the P adjust value. (C) Gene ontology (GO) enrichment analysis of DEGs, showing the top 15 enriched GO terms in biological process, cellular component, and molecular function. The size of nodes is scaled by the gene number. (D) Network analysis of enriched cellular component of GO. The size of nodes is scaled by the gene number. (E) Protein-protein interaction (PPI) analysis of DEGs by Cytohubba. The top 154 DEGs were colored by orange. (F) PPI network analysis of DEGs by MCODE. (G) Prioritization of DEGs by random forest analysis. The top 46 DEGs were ranked by mean decrease in accuracy and mean decrease in Gini. (H) Identification of hub genes by combination of three different algorithms. The Venn plot illustrates the number of overlap genes in all different algorithms, including Cytohubba, MCODE, and random forest analysis. (I) Gene set enrichment analysis of CD19.CAR-T cells and DT19.CAR-T cells. On the x-axis, gene sets represent by vertical black lines, while the enrichment score (ES) is plotted on the y-axis. Points representing genes and their corresponding ES are connected by a green line. The top portion of the plot shows the running ES for the gene set as the analysis walks down the ranked list. The score at the peak of the plot (the score furthest from 0.0) is the ES for the gene set. Gene sets with a distinct peak at the beginning or end of the ranked list are generally the most interesting. The middle portion of the plot shows where the members of the gene set appear in the ranked list of genes. The leading-edge subset of a gene set is the subset of members that contribute most to the ES. For a positive ES, the leading-edge subset is the set of members that appear in the ranked list prior to the peak score. For a negative ES, it is the set of members that appear subsequent to the peak score. The significance threshold is set at false discovery rate (FDR) < 0.05. The color code represents the signal to noise. Samples of three individual donors were tested.