Human pluripotent stem cell-derived hepatic progenitors exhibit a partially hypoimmunogenic phenotype and actively inhibit immune responses

Abstract

Introduction: GStemHep cells are human cryopreserved hepatic progenitors derived from pluripotent of stem cells (GStem cells) using a cGMP-compliant protocol. They were highly effective in rescuing mice from acute liver failure.

Methods: The objective of this study was to analyze the immunogenicity and immunoregulatory properties of GStemHep cells.

Results: As compared to GStem cells, GStemHep cells showed complete loss of HLA-I (ABC) and they lacked of expression of HLA-II, HLA-G, HLA-E and PD-L1. GStemHep cells also showed increased expression of CD47, maintained high expression of indoleamine 2,3-dioxygenase (IDO) and heme oxygenase-1 (HO-1) and reduced expression of CD200. In comparison with GStem cells, GStemHep cultured in inflammatory conditions increased the expression of PD-L1, CD200, HO-1, HLA-E, CD47 and HLA-I (ABC) as well as maintained expression of IDO and were negative for HLA-II and HLA-G. GStemHep culture in basal or inflammatory conditions has a low or absent immunogenic activity on T cells, associated to a suppressive effect on proliferation partially mediated by IDO. We observed phagocytosis of GStemHep by macrophages that was partially inhibited by CD47 expression. NK cells were activated by resting GStemHep cells. Upon culture in inflammatory conditions that induced expression of HLA-I molecules in GStemHep cells NK cell activation was reduced. Thus, GStemHep cells are partially hypoimmune cells due to the expression of several immune checkpoint inhibitors and the absence of HLA-I molecules. In inflammatory conditions, the expression of several of these molecules was increased but also of HLA-I that could be immunogenic for T cells but it was inhibitory for NK cells.

Discussion: GStemHep cells show a favorable immunological profile for their use as allogeneic off-the shelf treatment of liver diseases with loss of hepatocyte function.

Keywords: hepatic progenitors; liver; pluripotent stem cells; tolerance; transplantation.

Figure 1

Expression profile of immune markers in GStemHep and GStem cells. The expression of various molecules (CD47, PDL1, CD200, IDO, HO-1 and HLA- E/G/ABC/DR) involved in the cellular immune response was analyzed in thawed cells by flow cytometry. (A) Representative FACS histogram of the pluripotent stem cell line GStem (grey) and representative GStemHep cells from one production batch (blue). Cells incubated with isotype control are in red. (B) Percentage expression of different markers in several productions of GStemHep (n=17) after thawing compared with the pluripotent stem cell line GStem (n=3, mean ± SEM, two-way ANOVA test, ****p < 0.0001).

Figure 2

Expression profile of immune markers in GStemHep after inflammatory treatment. GStemHep were thawed and plated in culture medium with (inflammatory GStemHep) or without (cultured GStemHep) inflammatory cytokines (TNFα/IL1β/IFNγ). The expression of various molecules (CD47, PDL1, CD200, IDO, HO-1 and HLA-E/G/ABC/DR) involved in the cellular immune response was analyzed by flow cytometry in both culture conditions and compared with thawed cells. (A) Representative FACS histogram of thawed (dark blue), cultured (light blue) and inflammatory (orange) GStemHep. Cells incubated with isotype control are in red. (B) Percentage expression of different markers in several productions of GStemHep (n=4) after culture compared with thawed cells (n=3) and cultured cells in basal conditions (n=4). (two-way ANOVA test, *p < 0.05; ***p < 0.001; ****p < 0.0001).

Figure 3

Activation and proliferation responses of T cells in contact with GStemHep in vitro. GStemHep were seeded and cocultured with PMBCs from fresh blood for 3 days. PBMC alone were used as a negative control. PBMC incubated with phorbol 12-myristate 13-acetate (PMA)/ionomycine or CD3/CD28 beads were used as a positive control. Activation (CD69, CD25 and HLA-DR) and proliferation (Cell Proliferation Dye) of gated CD3+ T lymphocytes was analyzed by flow cytometry. (A) Flow cytometry analysis strategy and representative histogram of negative (PBMC only, blue) and positive (activated PBMC, red) control. (B) Expression of activation and proliferation markers on CD3+ T cells at PBMC: GStemHep ratios 1:1 (n=6) and 1:5 (n=7) after coculture with GStemHep. The addition of PMA/ionomycin or CD3/CD28 beads is performed in the respective positive control (activated PBMC) of activation (n=4) and proliferation (n=6). (C) Expression of activation and proliferation markers on CD3+ T cells at PBMC: GStemHep ratios 1:1 and 1:5 after coculture with GStemHep in inflammatory condition (n=3) (D) Proliferation of CD3+ T cells in coculture with GStemHep (1:1 ratio) in the presence or absence of CD3/CD28 activation beads and IDO inhibitor (n=4). (two-way ANOVA test, P Value: *p < 0.05; ***p < 0.001; ****p < 0.0001; ns, not significant).

Figure 4

Phagocytosis response of macrophages in contact with GStemHep in vitro. Monocyte-derived macrophages labelled with CPD670 (CPD670+) were cultured for 2 hours with GStemHep CPD450+. Then, phagocytosis is rated by the percentage of M0 macrophages (CPD670+) that have phagocytosed GStemHep (CPD670+ CPD450+). (A) Flow cytometry analysis strategy and representative histogram of negative (no coculture) and positive (coculture) conditions. (B) Percentage phagocytosis of macrophages in contact with thawed (n=7) or freshly cultured (n=8) GStemHep (Mann-Whitney test, P Value: **p=0.0012). (C) Impact of CD47 blockade: Percentage of phagocytosis of macrophages in contact with GStemHep in antibody-free condition (w/o Ab), with isotype control antibody (CTL) or anti-CD47 antibody (n=8). (D) Impact of CD47 overexpression: Phagocytosis of M0 in coculture with GStemHep transduced with CD47 lentiviral vector (CD47^high) relative to the condition with untreated GStemHep (NT) (n=11). (Wilcoxon test: **p<0.01).

Figure 5

Cytotoxic response of NK cells in contact with GStemHep in vitro. Purified NK cells were cultured for 5 hours with GStemHep and an anti-CD107a antibody. Then, CD56+ NK cells were analyzed by flow cytometry for surface expression of the degranulation marker CD107a as an indicator of NK cell activation. Cultured NK cells alone were used as a negative control. (A) Flow cytometry analysis strategy and representative histogram of results on two donors for NK only, NK + GStemHep and NK + inflammatory GStemHep conditions. (B) Percentage of membrane CD107a expression on CD56+ NK cells in culture alone or cocultured with GStemHep at NK: GStemHep ratios 1:1 (n=8) and 1:2 (n=3) (Kruskal-Wallis test: **p < 0.01; ns: not significant). (C) Fold change in CD107a percentage of NKs in the presence of GStemHep (1:1 ratio) cultivated or not in inflammatory medium versus NK control alone (n=5) (Wilcoxon test: *p < 0.05).