Cryopreserved cGMP-compliant human pluripotent stem cell-derived hepatic progenitors rescue mice from acute liver failure through rapid paracrine effects on liver cells

Abstract

Background: Liver transplantation remains the only curative treatment for end-stage liver diseases. Unfortunately, there is a drastic organ donor shortage. Hepatocyte transplantation emerged as a viable alternative to liver transplantation. Considering their unique expansion capabilities and their potency to be driven toward a chosen cell fate, pluripotent stem cells are extensively studied as an unlimited cell source of hepatocytes for cell therapy. It has been previously shown that freshly prepared hepatocyte-like cells can cure mice from acute and chronic liver failure and restore liver function.

Methods: Human PSC-derived immature hepatic progenitors (GStemHep) were generated using a new protocol with current good manufacturing practice compliant conditions from PSC amplification and hepatic differentiation to cell cryopreservation. The therapeutic potential of these cryopreserved cells was assessed in two clinically relevant models of acute liver failure, and the mode of action was studied by several analytical methods, including unbiased proteomic analyses.

Results: GStemHep cells present an immature hepatic phenotype (alpha-fetoprotein positive, albumin negative), secrete hepatocyte growth factor and do not express major histocompatibility complex. A single dose of thawed GStemHep rescue mice from sudden death caused by acetaminophen and thioacetamide-induced acute liver failure, both in immunodeficient and immunocompetent animals in the absence of immunosuppression. Therapeutic biological effects were observed as soon as 3 h post-cell transplantation with a reduction in serum transaminases and in liver necrosis. The swiftness of the therapeutic effect suggests a paracrine mechanism of action of GStemHep leading to a rapid reduction of inflammation as well as a rapid cytoprotective effect with as a result a proteome reprograming of the host hepatocytes. The mode of action of GStemHep relie on the alleviation of inhibitory factors of liver regeneration, an increase in proliferation-promoting factors and a decrease in liver inflammation.

Conclusions: We generated cryopreserved and current good manufacturing practice-compliant human pluripotent stem cell-derived immature hepatic progenitors that were highly effective in treating acute liver failure through rapid paracrine effects reprogramming endogenous hepatocytes. This is also the first report highlighting that human allogeneic cells could be used as cryopreserved cells and in the absence of immunosuppression for human PSC-based regenerative medicine for acute liver failure.

Keywords: Acetaminophen; Acute liver failure; Cryopreservation; Hepatic progenitor cells; Pluripotent stem cells; Regenerative medicine; Thioacetamide; cGMP.

Fig. 1

Expression profile of differentiation markers in GStemHep cells. A Schematic diagram of PSC-derived GStemHep cells. Human PSC were differentiated into GStemHep cells with a 10-day protocol. The expression profile of the indicated markers was analyzed in cells and culture supernatants. Pluripotency, endoderm and hepatic markers are shown in blue, green and orange, respectively (except for immunofluorescence staining). B RT‒qPCR analysis of the expression of key markers for differentiation stages and commitment to liver fate in GStemHep (n = 10). The results were normalized to the GAPDH housekeeping gene and expressed as the fold change relative to PSC. C Representative FACS dot plot of a PSC and a GStemHep production batch (detection threshold of OCT4 flow cytometry is 0.2%). Cells incubated with isotype control are in gray. D Average expression of pluripotency factor (OCT4) and hepatoblast markers (SOX17/HNF4A/AFP) by GStemHep (n = 9). E Immunofluorescence staining of GStemHep and human PSC for pluripotency (OCT4), proliferation (Ki67), endoderm (SOX17) and hepatic markers (HNF4A et AFP). The marker proteins and DAPI are shown in green and blue, respectively (magnification 20x). F Detection of secreted human AFP and human HGF in culture supernatant by ELISA (n = 10) (***p = 0.0007 Mann‒Whitney test)

Fig. 2

Cryopreservation of GStemHep. After 10 days of differentiation, GStemHep cells were harvested and frozen in a cryopreservation solution. Cell viability and adhesion capacity were analyzed after thawing. A Viability of freshly produced GStemHep at harvest and thawing after different months of freezing at − 150 °C; each dot represents a different batch or thawing (n = 202) (Kruskal‒Wallis test, ns: not significant). B Morphology of GStemHep (1 year frozen) 24 h after thawing and seeding onto a culture plate (Magnification 5x)

Fig. 3

Therapeutic effect of GStemHep in NOD/SCID mice with APAP-induced acute liver failure. After APAP intoxication, mice were treated (APAP + GStemHep) or not (APAP only) with intrasplenic injection of 1 × 10⁶ thawed GStemHep. Mouse survival was followed over 10 days, or mice were euthanized at 3 h, 6 h and 24 h post treatment for analysis of liver damage markers. A Survival curve of APAP-induced ALF mice treated with GStemHep injection (n = 46) compared with non-treated group (n = 60) (****p < 0.0001 log-rank (Mantel‒Cox) test). B Biochemical analysis of liver damage markers: ASAT (up) and ALAT (down) in the serum of each group (n = 18 for APAP only, n = 19 for APAP + GStemHep and n = 8 for no APAP) 24 h after APAP intoxication (*p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.0001 one-way ANOVA test). C Necrosis quantification on HES-stained liver sections in APAP only (n = 5) and APAP + GStemHep groups (n = 5) at 3 h, 6 h and 24 h after GStemHep transplantation (**p < 0.01 two-way ANOVA test). D Representative HES-stained liver sections from each group (n = 5 per group) 24 h after GStemHep transplantation (magnification 5x). Areas of necrosis are delimited by the red dotted lines

Fig. 4

GStemHep tracking in the APAP-ALF model after transplantation. After APAP intoxication and cell transplantation (APAP + GStemHep), cell homing was analyzed in the liver and the spleen, and human AFP were quantified in the serum of mice at different times post-treatment (3 h, 6 h and 24 h). Untreated APAP-ALF mice served as controls (APAP only). A Detection of human ALU DNA sequences (human specific/290 bp) in mouse liver by PCR at 24 h after transplantation; each number represents a different mouse. B Immuno-histochemical staining for specific human Ku80 in mouse liver and spleen at 24 h after transplantation (Magnification 10 × on the left and 20 × on the right), data representative of 10 analyzed mice. (C) Quantification of human AFP in mouse serum by ELISA 3 h, 6 h and 24 h after transplantation (n = 4 in APAP only, n = 5 in APAP + GStemHep at 6 h and 9 h, n = 14 in APAP + GStemHep at 24 h post-transplantation), each point represents a different mouse

Fig. 5

Mechanistic effects of GStemHep in APAP-ALF NOD/SCID mice. After APAP intoxication, mice were treated (APAP + GStemHep) or not (APAP only) with 1 × 10⁶ thawed GStemHep. Mouse livers were collected at 3 h, 6 h and 24 h post cell transplantation to analyze variations in gene expression. RT‒qPCR analysis of the expression of inflammation or regeneration markers (mIL1RN, mIL6, mCCL2, mTGFβ, mKi67 and mVEGFa) in the APAP only and APAP + GStemHep groups (n = 5 per group). The results were normalized to the mGAPDH housekeeping gene and expressed as the fold change relative to healthy control mice (*p < 0.05; **p < 0.01; ns: not significant, Mann‒Whitney test)

Fig. 6

Mechanistic effects of GStemHep in APAP-ALF NOD/SCID mice. After APAP intoxication, mice were treated (APAP + GStemHep) or not (APAP only) with 1 × 10⁶ thawed GStemHep. Mouse livers were collected at 6 h post cell transplantation to analyze variations in protein expression. A Proteome-wide data set variance described by Principal Component Analysis (PCA) on healthy (no APAP, n = 3), APAP only (n = 4) and APAP + GStemHep groups (n = 5), 6 h after cell transplantation (i.e., 9 h post-APAP intoxication). B KEGG pathway analysis of differentially regulated (up- and downregulated) proteins in the APAP + GStemHep group compared to the APAP group. (I) Upregulation pathways in pink. (II) Downregulation pathways in yellow. C Heatmap visualizing the intensities of differentially regulated proteins in three GO annotations (blood coagulation, complement activation and metabolic process) between all groups; each line represents a protein

Fig. 7

Therapeutic effect of GStemHep and cell tracking in C57Bl/6 mice with TAA-induced ALF. After TAA intoxication, mice were treated (TAA + GStemHep) or not (TAA only) with intrasplenic injection of 1 × 10⁶ GStemHep. Mouse survival was monitored for 7 days, or mice were euthanized at 24 h post treatment for PCR analysis of the liver. Sera were collected at different times. A Survival curve of TAA-induced ALF mice treated with GStemHep (n = 101) compared with non-treated group (n = 80) (****p < 0.0001 log-rank (Mantel‒Cox) test). B Quantification of human AFP in the serum of the TAA-only group (n = 5) and the TAA + GStemHep group at 24 h (n = 13) and 7 days (n = 5) after GStemHep transplantation by ELISA. C Detection of human ALU DNA sequences in mouse livers by PCR at 24 h after transplantation; each number represents a different mouse