Development of off-the-shelf hematopoietic stem cell-engineered invariant natural killer T cells for COVID-19 therapeutic intervention

Abstract

Background: New COVID-19 treatments are desperately needed as case numbers continue to rise and emergent strains threaten vaccine efficacy. Cell therapy has revolutionized cancer treatment and holds much promise in combatting infectious disease, including COVID-19. Invariant natural killer T (iNKT) cells are a rare subset of T cells with potent antiviral and immunoregulatory functions and an excellent safety profile. Current iNKT cell strategies are hindered by the extremely low presence of iNKT cells, and we have developed a platform to overcome this critical limitation.

Methods: We produced allogeneic HSC-engineered iNKT (^AlloHSC-iNKT) cells through TCR engineering of human cord blood CD34⁺ hematopoietic stem cells (HSCs) and differentiation of these HSCs into iNKT cells in an Ex Vivo HSC-Derived iNKT Cell Culture. We then established in vitro SARS-CoV-2 infection assays to assess ^AlloHSC-iNKT cell antiviral and anti-hyperinflammation functions. Lastly, using in vitro and in vivo preclinical models, we evaluated ^AlloHSC-iNKT cell safety and immunogenicity for off-the-shelf application.

Results: We reliably generated ^AlloHSC-iNKT cells at high-yield and of high-purity; these resulting cells closely resembled endogenous human iNKT cells in phenotypes and functionalities. In cell culture, ^AlloHSC-iNKT cells directly killed SARS-CoV-2 infected cells and also selectively eliminated SARS-CoV-2 infection-stimulated inflammatory monocytes. In an in vitro mixed lymphocyte reaction (MLR) assay and an NSG mouse xenograft model, ^AlloHSC-iNKT cells were resistant to T cell-mediated alloreaction and did not cause GvHD.

Conclusions: Here, we report a method to robustly produce therapeutic levels of ^AlloHSC-iNKT cells. Preclinical studies showed that these ^AlloHSC-iNKT cells closely resembled endogenous human iNKT cells, could reduce SARS-CoV-2 virus infection load and mitigate virus infection-induced hyperinflammation, and meanwhile were free of GvHD-risk and resistant to T cell-mediated allorejection. These results support the development of ^AlloHSC-iNKT cells as a promising off-the-shelf cell product for treating COVID-19; such a cell product has the potential to target the new emerging SARS-CoV-2 variants as well as the future new emerging viruses.

Keywords: Allogeneic adoptive cell transfer; Coronavirus disease 2019 (COVID-19); Hematopoietic stem cell (HSC); Invariant natural killer T (iNKT) cell; Off-the-shelf cellular product; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

Fig. 1

In vitro generation and characterization of allogenic HSC-engineered iNKT (^AlloHSC-iNKT) cells. a Experimental design to generate ^AlloHSC-iNKT cells in vitro. CB cord blood, HSC hematopoietic stem cell, SG suicide gene, Lenti/iNKT-SG lentiviral vector encoding an iNKT TCR gene and a suicide gene, ATO artificial thymic organoid. b Generation of iNKT cells (identified as iNKT TCR⁺CD3⁺ cells) during ATO culture. A 6B11 monoclonal antibody was used to stain iNKT TCR. c Generation of iNKT cells (identified as iNKT TCR⁺CD3⁺ cells) during Feeder-free culture. d Yields of ^AlloHSC-iNKT cells generated from ATO and Feeder-free cultures. e FACS characterization of surface marker expression and intracellular cytokine and cytotoxic molecule production of ^AlloHSC-iNKT cells. Periphery blood mononuclear cell (PBMC)-derived iNKT (PBMC-iNKT) cells and conventional αβ T (PBMC-Tc) cells were included as controls. Representative of over 5 experiments

Fig. 2

^AlloHSC-iNKT cells directly target and kill SARS-CoV-2 infected cells. a Schematics showing the engineered 293T-FG, 293T-ACE2-FG, and Calu3-FG cell lines. ACE2 angiotensin converting enzyme 2, Fluc firefly luciferase, EGFP enhanced green fluorescent protein, F2A foot-and-mouth disease virus 2A self-cleavage sequence. b FACS detection of ACE2 on 293T-FG, 293T-ACE2-FG, Calu3-FG, and ^AlloHSC-iNKT cells. c–h In vitro direct killing of SARS-CoV-2 infected cells by ATO culture-generated ^AlloHSC-iNKT cells. c Experimental design. d Target cell killing data of ^AlloHSC-iNKT cells at 24-h post co-culturing with infected cells (n = 5). e FACS detection of CD69, Perforin and Granzyme B of ^AlloHSC-iNKT cells at 24-h post co-culturing with SARS-CoV-2 infected 293T-ACE2-FG cells. f ELISA analysis of IFN-γ production (n = 3). h SARS-CoV-2 infected cell killing mechanisms of ^AlloHSC-iNKT cells. NKG2D and DNAM-1 mediated pathways were studied (n = 5). i Immunofluorescence analysis of direct targeting of SARS-CoV-2 infected cells by ^AlloHSC-iNKT cells. Representative of 3 experiments. Data are presented as the mean ± SEM. ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by Student's t test (d, f and g), or by 1-way ANOVA h. See also Additional file 1: Fig. S1

Fig. 3

^AlloHSC-iNKT cells reduce virus-infection promoted inflammatory monocytes. a Experimental design. 293T-ACE2-FG cells were infected by SARS-CoV-2 virus. After 1 day, ATO culture-generated ^AlloHSC-iNKT cells and donor-mismatched PBMCs were added and incubated for 24 h. Flow cytometry was used to detect cell populations. b FACS detection of CD14⁺ monocytes, T cells, and B cells in PBMCs. c Quantification of b (n = 5). d FACS detection of CD1d expression on CD14⁺ monocytes, T cells, and B cells. e Quantification of d (n = 5). Representative of 3 experiments. Data are presented as the mean ± SEM. ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by 1-way ANOVA

Fig. 4

Safety and immunogenecity of ^AlloHSC-iNKT cells. a–b Studying the graft-verus-host (GvH) response of ^AlloHSC-iNKT cells using an in vitro mixed lymphocyte reaction (MLR) assay. PBMC-Tc cells were included as a responder cell control. a Experimental design. PBMCs from 5 different healthy donors were used as stimulator cells. b ELISA analyses of IFN-γ production at day 4 (n = 3). N, no stimulator cells. c–d Studying the GvH response of ^AlloHSC-iNKT cells using NSG mouse model. PBMC-Tc were included as a control. A Experimental design. B Kaplan–Meier survival curves of experimental mice over time (n = 6). e–g Studying T cell-mediated alloreaction against ^AlloHSC-iNKT cells using an in vitro MLR assay. Irradiated ^AlloHSC-iNKT cells (as stimulators) were co-cultured with donor-mismatched PBMC cells (as responders). Irradiated PBMC-Tc cells were included as a stimulator cell control. e Experimental design. PBMCs from 3 different healthy donors were used as responders. f ELISA analyses of IFN-γ production at day 4 (n = 3). g FACS analyses of HLA-I and HLA-II expression on the indicated stimulator cells (n = 5). h–j Studying HLA-I and HLA-II expression on ^AlloHSC-iNKT cells under SARS-CoV-2 infection. ^AlloHSC-iNKT cells were co-cultured with SARS-CoV-2 infected target cells. PBMC-Tc cells were included as a control. h Experimental design. i FACS analyses of HLA-I and HLA-II expression on the indicated stimulator cells. j Quantification of i (n = 5). Representative of 3 experiments. Data are presented as the mean ± SEM. ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by Student's t test (j and g), by 1-way ANOVA (b and f), or by log rank (Mantel–Cox) test adjusted for multiple comparisons (d)