Generation of "Off-the-Shelf" Natural Killer Cells from Peripheral Blood Cell-Derived Induced Pluripotent Stem Cells

Abstract

Current donor cell-dependent strategies can only produce limited "made-to-order" therapeutic natural killer (NK) cells for limited patients. To provide unlimited "off-the-shelf" NK cells that serve many recipients, we designed and demonstrated a holistic manufacturing scheme to mass-produce NK cells from induced pluripotent stem cells (iPSCs). Starting with a highly accessible human cell source, peripheral blood cells (PBCs), we derived a good manufacturing practice-compatible iPSC source, PBC-derived iPSCs (PBC-iPSCs) for this purpose. Through our original protocol that excludes CD34+ cell enrichment and spin embryoid body formation, high-purity functional and expandable NK cells were generated from PBC-iPSCs. Above all, most of these NK cells expressed no killer cell immunoglobulin-like receptors (KIRs), which renders them unrestricted by recipients' human leukocyte antigen genotypes. Hence, we have established a practical "from blood cell to stem cells and back with less (less KIRs)" strategy to generate abundant "universal" NK cells from PBC-iPSCs for a wide range of patients.

Keywords: cancer; cell therapy; cytotoxicity; immunotherapy; induced pluripotent stem cells; killer cell immunoglobulin-like receptors; natural killer cells; peripheral blood cells.

Figure 1

Generation of NK Cells from hPSCs (A) A schematic of a two-stage protocol for production of NK cells from hPSCs. (B–I) Morphological changes during differentiation of H1 cells into NK cells. Phase contrast images show (B) undifferentiated H1 cells; (C) H1 and OP9 co-culture, day 12; (D) first differentiated cells and OP9-DLL1 co-culture, day 7 (day 19); (E) second differentiated cells and OP9-DLL1 co-culture, day 7 (day 26); (F and G) third differentiated cells and OP9-DLL1 co-culture, day 7 (day 33); (H and I) fourth differentiated cells and OP9-DLL1 co-culture, day 7 (day 40). (J–L) Phenotypic changes during differentiation of H1 cells into NK cells. Flow cytometric analysis shows (J) CD34+ cells from H1 and OP9 co-culture, day 12; (K and L) CD56+ CD45+ cells from third (day 33) and fourth (day 40) co-culture on OP9-DLL1.

Figure 2

Generation of PBC-iPSCs for Production of NK Cells PBCs from different donors were used to generate various PBC-iPSC lines. Results were obtained using PBC-iPSC (donor A) lines (A–K) and PBC1-iPSC (donor B) lines (L). (A) Morphology of two PBC-iPSC (donor A) lines, PBC-iPSC#9 and PBC-iPSC#8.3. (B and C) TCRB gene (B) and TCRG gene (C) clonality assays to detect rearranged TCRβ and TCRγ chain genes in PBC-iPSC (donor A) lines. Positive amplified product is indicated by the yellow arrow. (D–K) Production of NK cells from two PBC-iPSC (donor A) lines, PBC-iPSC#9 and PBC-iPSC#8.3. (D–G) Phenotypic changes during differentiation of non-T cell-derived PBC-iPSC#9 line into NK cells. Flow cytometric analysis shows no TCRαβ expression during differentiation. (H–K) Phenotypic changes during differentiation of T cell-derived PBC-iPSC#8.3 line into NK cells. Flow cytometric analysis shows re-expression of TCRαβ during differentiation. (L) TCRB gene and TCRG gene clonality assays to detect rearranged TCRβ and TCRγ chain genes in PBC1-iPSC (donor B) lines.

Figure 3

Morphology, Purity, and Phenotype of PBC-iPSC-NK Cells (A) Morphology of fifth differentiated cells and OP9-DLL1 co-culture, day 7 (day 47) starting with PBC-iPSC#9 (donor A). (B) Morphology of PBC-iPSC#9-NK cells (donor A) after harvesting and purification by density gradient centrifugation using Ficoll-Paque followed by overnight culture. (C) Purity of PBC-iPSC#9-NK cells (donor A) as evaluated by flow cytometry. (D) Phenotype of PBC-iPSC#9-NK cells (donor A). (E) Phenotype of PBC1-iPSC#4-NK cells (donor B).

Figure 4

Functions of PBC-iPSC-NK Cells (A and B) IFN-γ secretion by PBC-iPSC#9-NK cells (donor A) upon stimulation with K562 and Raji cells as detected by ELISPOT assay. ELISPOT images (A) and spot counting (mean ± SD, n = 3) (B) are shown. (C and D) GrB secretion by PBC-iPSC#9-NK cells upon stimulation with K562 cells as detected by ELISPOT assay. ELISPOT images (C) and spot counting (mean ± SD, n = 3) (D) are shown. (E and F) Cytotoxicity of PBC-iPSC#9-NK cells against K562 and Raji cells as measured by flow cytometry. A representative flow cytometric analysis (E) and a result summary (F) are shown. (G and H) ADCC of PBC-iPSC#9-NK cells against Raji cells in the presence of anti-CD20 humanized antibody as measured by flow cytometry. A representative flow cytometric analysis (G) and a result summary (H) are shown. These data are representative of three independent experiments.

Figure 5

Expansion of Fresh and Cryopreserved PBC-iPSC-NK Cells (A and B) Expansion of fresh PBC-iPSC#9-NK cells by K562-mbIL15-41BBL in G-Rex10 starting with different NK cell numbers. Absolute numbers of NK cells during a 14-day expansion (A) and fold changes after expansion (B) are shown. (C and D) Phenotype (C) and cytotoxicity against K562 (D) of fresh PBC-iPSC#9-NK cells after expansion as measured by flow cytometry. (E) Viability of expanded PBC-iPSC#9-NK cells after freeze/thaw procedure. (F) Expansion of cryopreserved PBC-iPSC#9-NK cells. (G) Cytotoxicity of cryopreserved PBC-iPSC#9-NK cells against K562 after expansion. These data are representative of three independent experiments.

Figure 6

Cytotoxicity of PBC-iPSC-NK Cells against Cancer Cells Three different sources of PBC-iPSC-NK cells, including PBC-iPSC#9-NK cells (donor A), PBC1-iPSC#4-NK cells (donor B), and PBC2-iPSC#12-NK cells (donor C), were used for cytotoxicity assay against a wide variety of cancer cell lines: K562 (A), SK-OV-3 (B), SW480 (C), HCT-8 (D), MCF7 (E), and SCC-25 (F). PB-NK cells expanded from three different donors (donor 1, donor 2, and donor 3) were used as controls. (G–I) A short-term cultured primary tumor cell line CRC7.4 was derived from a colorectal cancer sample and characterized by immunostaining (G–J) and flow cytometry (K). These primary tumor cells were then use as target cells to evaluate the cytotoxicity of PBC-iPSC#9-NK cells (L). PB-NK cells expanded from three donors (donor 4, donor 5, and donor 6) were used as controls. Student's paired two-tailed t test was used to analyze the difference between the specific lysis of target cells by PBC-iPSC-NK cells and that by PB-NK cells. The p values were calculated for each type of target cell and a p value less than 0.05 was considered to be statistically significant. These data are representative of three independent experiments.

Figure 7

KIR Typing of PBC-iPSC-NK Cells (A–J) KIR genotyping and mRNA expression profiling. Electrophoresis of PCR products of genomic DNA shows KIR gene content and endogenous β-actin gene of a PB-NK donor (donor 4) (A) and three PBC donors for PBC-iPSC#9 (donor A) (C), PBC1-iPSC#4 (donor B) (G), and PBC2-iPSC#12 (donor C) (I). Electrophoresis of PCR products of cDNA shows mRNA expression of KIR genes and endogenous β-actin gene in PB-NK cells (B), PBC-iPSC#9 cells (D), PBC-iPSC#9-NK cells (E), post-expansion PBC-iPSC#9-NK cells (F), PBC1-iPSC#4-NK cells (H), and PBC2-iPSC#12-NK cells (J). (K–P) KIR phenotyping of different NK cells by flow cytometry.