A clinically applicable and scalable method to regenerate T-cells from iPSCs for off-the-shelf T-cell immunotherapy

Abstract

Clinical successes demonstrated by chimeric antigen receptor T-cell immunotherapy have facilitated further development of T-cell immunotherapy against wide variety of diseases. One approach is the development of "off-the-shelf" T-cell sources. Technologies to generate T-cells from pluripotent stem cells (PSCs) may offer platforms to produce "off-the-shelf" and synthetic allogeneic T-cells. However, low differentiation efficiency and poor scalability of current methods may compromise their utilities. Here we show improved differentiation efficiency of T-cells from induced PSCs (iPSCs) derived from an antigen-specific cytotoxic T-cell clone, or from T-cell receptor (TCR)-transduced iPSCs, as starting materials. We additionally describe feeder-free differentiation culture systems that span from iPSC maintenance to T-cell proliferation phases, enabling large-scale regenerated T-cell production. Moreover, simultaneous addition of SDF1α and a p38 inhibitor during T-cell differentiation enhances T-cell commitment. The regenerated T-cells show TCR-dependent functions in vitro and are capable of in vivo anti-tumor activity. This system provides a platform to generate a large number of regenerated T-cells for clinical application and investigate human T-cell differentiation and biology.

Fig. 1. Generation of CD4⁺CD8αβ⁺ DP T-cells from T-iPSCs in feeder-free (Ff) culture conditions.

a Scheme of hematopoietic induction from T-iPSC clones. EBs were generated from a single-cell dissociated human T-iPSCs proliferated in a feeder- and serum-free condition and induced to differentiate into mesoderm and later into hematopoietic cells in the presence of CH, BMP-4, bFGF, and VEGF. The emerging hematopoietic cells were proliferated in the presence of hematopoietic cytokines. SB was added on day 2 to induce definitive hematopoiesis. CH, CHIR99021; SB, SB431542. b Representative kinetic analysis of hematopoietic differentiation from a T-iPSC clone, TkT3V1-7, assessed based on the expression levels of CD34 and CD43 at the indicated days after induction (n = 3). c Schedule of T-cell differentiation of T-iPSC-HPCs in immobilized-DL4 and retronectin culture plates. d Representative flow cytometry plots of T-iPSC-HPCs differentiated on DL4 and RN cultures in the presence of SCF, TPO, FLT3L, and IL-7 for 21 days (n = 4). e Frequencies of CD5⁺CD7⁺ progenitor T-cells (blue) and CD4⁺CD8αβ⁺ DP-cells (purple) generated from T-iPSC-HPCs 21 days after differentiation on DL4 and RN cultures (n = 4). f DL4 and RN cell yields after 21 days of culture normalized to the input T-iPSC-HPC numbers (n = 3). Data represent mean ± SD of n independent experiments.

Fig. 2. Optimization of T-cell differentiation in DL4 and RN cultures.

a Representative kinetic flow cytometric analysis of T-cell differentiation on DL4 and RN cultures assessed at the indicated days. b Frequencies of Annexin V(-)/7-AAD (-) live cells in differentiating cultures over 3 weeks in the presence or absence of SDF1α and/or SB203580 (n = 4). c Frequencies of the indicated cell types defined as in (a) in the differentiation culture over 21 days in the presence or absence of SDF1α and/or SB203580 (n = 4). d Kinetics of differentiating T-iPSC-HPC counts over 21 days on DL4 and RN in the presence or absence of SDF1α and/or SB203580 (n = 4). e Gene expression related to T-cell differentiation, Notch signaling pathway, and a myeloid lineage gene (SPI1 (PU.1)) detected by qPCR after T-iPSC-HPCs were cultured for 48 or 96 h on DL4 and RN + / − SS (n = 3). Data represent mean ± SEM of n independent experiments.^*P < 0.05; ^**P < 0.01; ^***P < 0.001 by one-way ANOVA with Tukey’s multiple comparison test.

Fig. 3. Generation of CD8αβ⁺ T-cells from antigen-specific T-iPSCs in the optimized condition.

a Representative flow cytometry plots of three T-iPSC-derived differentiating cells after 21 days in SS + condition showing expression levels of CD5 vs CD7 and CD8αβ vs CD4. b Frequencies of CD5⁺CD7⁺ progenitor T-cells (blue) and CD4⁺CD8αβ⁺ DP-cells (purple) generated from the indicated T-iPSC-HPCs after 21 days on DL4/RN + SS condition (n = 3). c Schedule of maturation of DL4-cells to induce CD4⁻CD8αβ⁺ T-cells. d CD4⁻CD8αβ⁺ T-cell yield as determined by input T-iPSC count during differentiation processes. e Representative flow cytometry plots obtained 42 days after differentiation (after 7 days in maturation culture as shown in (c) showing the expression levels of tetramer and CD3 (left) and CD8β and CD4 (right) of regenerated T-cells (n = 3). Data represent mean ± SEM of n independent experiments. f–h In vitro cytotoxicity assay of proliferated CD8αβ⁺ SP T-cells measured by ⁵¹Cr release assay using LCL cells loaded with specific peptides (Nef38 for H25-4 (f) and Gag28-8 for H25-31 (g)) or SK-Hep cells transduced with GPC3 or empty vectors (GPC3#16-1 (h)) as target cells. Data are representative of two independent experiments.

Fig. 4. Generation of CD8αβ⁺ SP T-cells from TCR-transduced HLA-homo iPSCs.

a Scheme of production starting from HLA-homo iPSC to WT1-TCR-transduced iPSC CD8αβ⁺ T-cells. Wk, week; HPCs, hematopoietic progenitor cells; DP, CD4⁺CD8⁺ double-positive; CTLs, CD4⁻CD8αβ⁺ cytotoxic T-cells. b, c Flow cytometry plot of differentiating untransduced (top) or WT1-TCR transduced (bottom) HLA-homo iPSCs (b) 14 days after differentiation (HPC stage) gated on the CD14⁻CD235α⁻ population and (c) 35 days after differentiation (DP stage) showing the expression levels of αβTCR and CD3 (left), CD5 and CD7 (middle), and CD8α and CD4 (right) (n = 3). d Kinetics of total differentiating cell count during T-cell differentiation in a large-scale culture generated from 1 × 10⁵ WT1-iPSC-HPCs seeded on a DL4/RN-coated 10 cm dish (n = 3 independent experiments). Data represent mean ± SD. e Flow cytometry plots of mature CD4⁻CD8αβ⁺αβTCR⁺ T-cells generated from WT1-transduced HLA-homo iPSCs at day 42 after differentiation. f VDJ frequency of WT1-TCR-transduced iPSC CD8 SP T-cells determined by next-generation sequencing of CDR3 regions of TCRα (left) and TCRβ (right) (n = 3 independent experiments).

Fig. 5. In vitro and in vivo functions of WT1-TCR-transduced HLA-homo iCD8αβ T-cells.

a Fold proliferation of WT1-TCR iCD8αβ T-cells in an optimized protocol as shown (top) for four sequential rounds of stimulations. Data are representative of two independent experiments. b Flow cytometry plots of proliferated WT1-iPSC CD8 SP T-cells representing the expression levels of tetramer and naive/memory T-cell markers (n = 3 independent experiments). c In vitro cytotoxicity assay of proliferated WT1-TCR iCD8αβ T-cells measured by DELFIA assay using LCL cells loaded with specific peptides as target cells (left) and NCI-H226 (right). Data are representative of two independent experiments. d Intracellular cytokine production of WT1-TCR iCD8αβ T-cells after 6 h stimulation with PMA/Ionomycin. Data are representative of two independent experiments. e, f In vivo anti-tumor activity of WT1-TCR iCD8αβ T-cells. NSG mice were intraperitoneally inoculated with NCI-H226-expressing luciferase, treated 4 times weekly with HBSS or WT1-TCR iCD8αβ T-cells, and monitored for (e) tumor volume and (f) survival rate (n = 5 animals each). *P = 0.0016. Data represent mean ± SEM. ^*P < 0.05; ^**P < 0.01 (Welch’s two samples t-test (two-tailed) (e); log-rank test (two-tailed) (f)).

Fig. 6. In vitro and in vivo functions of iCART cells.

a A schematic showing generation and assessment of iCART cells from iT cells. b Designs of RV-19BBz CAR (top) and RV-mbIL-15 (bottom). This design allows the co-expression of the CAR and truncated EGFR and mbIL-15 and LNGFR from the same LTR promoter by using a self-cleaving P2A sequence. LTR, long terminal repeat, Porcine teschovirus self-cleaving 2A sequence. c Representative flow cytometry plots of untransduced and 19BBz-transduced iT cells (iCART cells) showing expression levels of CD19 CAR, IL-15Rα, CD3, αβTCR, CD8α, CD8β (left), naive/memory T-cell markers (middle), and exhaustion markers (right). d Cytotoxic activity of iT cells and iCART cells using CD19⁺ NALM-6 and CD19^- CCRF-CEM as target cells. Data represent two independent experiments. e Representative histograms showing cell divisions of CellTrace Violet-labeled iT cells (left) and iCART cells (right) cocultured with CD19 + NALM-6 or CD19- CCRF-CEM (red) or without target cells (blue) for 6 days. Cell divisions of iT cells and iCART cells activated by retronectin/CD3 conditions are shown (top panels). Data represent two independent experiments. f, g In vivo anti-tumor activity of iT cells and iCART cells in a systemic tumor model. NOG mice were intravenously inoculated with NALM-6-expressing luciferase 4 days before treatment, treated once intravenously with PBS, 1 × 10⁷ iT-cells, iCART-cells, or primary CART-cells, and monitored for (f) tumor volume and (g) survival rate (n = 5 mice each). Values in parentheses represent the fraction of mice without tumor relapse. h NALM-6-bearing mice were treated with 1 × 10⁷ iT cells or iCART cells. At 10 and 15 days after treatment, mice were euthanized and bone marrow cells were collected. Presence of iT cell or iCART cells (GFP^-CD19^-humanCD45⁺) and NALM-6 (GFP⁺CD19⁺) cells were analyzed by flow cytometry. ^*P < 0.05 (log-rank Mantel–Cox test with Bonferroni corrections, two-tailed).