Amnion-Derived Mesenchymal Stromal/Stem Cell Paracrine Signals Potentiate Human Liver Organoid Differentiation: Translational Implications for Liver Regeneration

Abstract

The prevalence of end-stage liver diseases has reached very high levels globally. The election treatment for affected patients is orthotopic liver transplantation, which is a very complex procedure, and due to the limited number of suitable organ donors, considerable research is being done on alternative therapeutic options. For instance, the use of cell therapy, such as the transplantation of hepatocytes to promote liver repair/regeneration, has been explored, but standardized protocols to produce suitable human hepatocytes are still limited. On the other hand, liver progenitor and multipotent stem cells offer potential cell sources that could be used clinically. Different studies have reported regarding the therapeutic effects of transplanted mesenchymal stromal/stem cells (MSCs) on end-stage liver diseases. Moreover, it has been shown that delivery of MSC-derived conditioned medium (MSC-CM) can reduce cell death and enhance liver proliferation in fulminant hepatic failure. Therefore, it is believed that MSC-CM contains many factors that probably support liver regeneration. In our work, we used an in vitro model of human liver organoids to study if the paracrine components secreted by human amnion-derived MSCs (hAMSCs) affected liver stem/progenitor cell differentiation. In particular, we differentiated liver organoids derived from bipotent EpCAM⁺ human liver cells and tested the effects of hAMSC secretome, derived from both two-dimensional (2D) and three-dimensional (3D) hAMSC cultures, on that model. Our analysis showed that conditioned medium (CM) produced by 3D hAMSCs was able to induce an over-expression of mature hepatocyte markers, such as ALB, NTCP, and CYP3A4, compared with both 2D hAMSC cultures and the conventional differentiation medium (DM). These data were confirmed by the over-production of ALB protein and over-activity of CYP3A4 observed in organoids grown in 3D hAMSC-CM. Liver repair dysfunction plays a role in the development of liver diseases, and effective repair likely requires the normal functioning of liver stem/progenitor cells. Herein, we showed that hAMSC-CM produced mainly by 3D cultures had the potential to increase hepatic stem/progenitor cell differentiation, demonstrating that soluble factors secreted by those cells are potentially responsible for the reaction. This work shows a potential approach to improve liver repair/regeneration also in a transplantation setting.

Keywords: 3D liver organoid culture; hepatic progenitor cell differentiation; hepatocyte culture; human amnion-derived mesenchymal stem cells; liver regeneration.

Figure 1

Generation of liver organoids. (A) Human liver cell suspensions were analyzed by FACS analysis and separated into EpCAM⁺ and EpCAM⁻ cells. (B) Representative DIC image of organoid cultures. (C) Immunofluorescent confocal microscopy imaging of liver organoid showing epithelial cell adhesion molecule (EpCAM) (green). (D) Number of cells counted per well at each passage from P1 to P5. (E,F) Gene expression of stem/progenitor, ductal/epithelial markers, and hepatocyte markers. Transcript levels were normalized to those of GAPDH and expressed as relative expression (2^−ΔCt). Data are means ± SD. *p < 0.05 vs. HDFa. DIC, differential interference contrast.

Figure 2

Human amnion-derived mesenchymal stem cells (hAMSCs) cultured as both monolayer and spheroids. (A) Representative DIC images of hAMSCs grown in monolayer (2D cultures). (B) Representative DIC images of hAMSCs grown as spheroids (3D cultures). (C) Immunofluorescence staining localization of fatty acid binding protein 4 (FABP4), Osteocalcin (OC), and Aggrecan (ACAN) in hAMSCs grown as both monolayer and spheroids. (D) Graphic depicts FABP4, OC, and ACAN fluorescence intensity in hAMSCs grown as both monolayer and spheroids. (E) Cytofluorimetric analysis of the surface marker in hAMSCs grown in 2D cultures and 3D cultures at passage 2. DIC, differential interference contrast. *p < 0.05 vs. 2D hAMSCs.

Figure 3

Differentiation of organoids. (A) Experimental design (with DIC images of undifferentiated and differentiated liver organoids) to study the potential effects of MSC-derived conditioned medium (CM). (B–E) Expression analysis of markers of hepatic differentiation (ALB, NTCP, and CYP3A4) and of the stem marker (LGR5) in liver organoids differentiated for 15 days in each condition. DM, differentiation medium; EM, expansion medium; 2D hAMSC-CM, DM conditioned for 3 days by hAMSCs grown in monolayer; 3D hAMSC-CM, DM conditioned for 3 days by hAMSCs grown as spheroids. Transcript levels were normalized to those of GAPDH and expressed as fold change (2^−ΔΔCt) vs. EM. Data are means ± SD.; *p < 0.05 vs. EM; ^#p < 0.05 vs. 2D hAMSC-CM. DIC, differential interference contrast.

Figure 4

Analysis of (A) albumin secretion and (B) CYP3A4 activity in human liver organoids differentiated for 15 days in each condition. EM, expansion medium; DM, differentiation medium; 2D hAMSC-CM, DM conditioned for 3 days by hAMSCs grown in monolayer, 3D hAMSC-CM, DM conditioned for 3 days by hAMSCs grown as spheroids. Data are means ± SD. *p < 0.05 vs. EM. ^#p < 0.05 vs. 2D hAMSC-CM.

Figure 5

Viability and secretion of cytokines/chemokines and growth factors in monolayer (2D) and spheroids cultures (3D) of human amnion-derived mesenchymal stem cells (hAMSCs). (A) Quantification of both alive and necrotic/apoptotic cells in 2D and 3D cultures of hAMSCs. (B) The conditioned medium (CM) was collected for both 2D and 3D hAMSCs cultured in differentiation medium (DM) for 72 h and selected cytokines/chemokines and growth factors were analyzed. Data are means ± SD. *p < 0.05 vs. DM. ^#p < 0.05 vs. 2D hAMSC-CM.