Activation of Notch1 signalling promotes multi-lineage differentiation of c-Kit(POS)/NKX2.5(POS) bone marrow stem cells: implication in stem cell translational medicine

Abstract

Introduction: Transplantation of bone marrow mesenchymal stem cells (BMSCs) can repair injured hearts. However, whether BMSC populations contain cells with cardiac stem cell characteristics is ill-defined. We report here that Notch signalling can promote differentiation of c-Kit(POS)/NKX2.5(POS) BMSCs into cardiomyocyte-like cells.

Methods: Total BMSCs were isolated from Sprague-Dawley rat femurs and c-Kit(POS) cells were purified. c-Kit(POS)/NKX2.5(POS) cells were isolated by single-cell cloning, and the presence of cardiomyocyte, smooth muscle cell (SMC), and endothelial cell differentiation markers assessed by immunofluorescence staining and semi-quantitative reverse-transcription polymerase chain reaction (RT-PCR) analysis. Levels of c-Kit and Notch1-4 in total BMSCs and c-Kit(POS)/NKX2.5(POS) BMSCs were quantitated by flow cytometry. Following infection with an adenovirus over-expressing Notch1 intracellular domain (NICD), total BMSCs and c-Kit(POS)/NKX2.5(POS) cells were assessed for differentiation to cardiomyocyte, SMC, and endothelial cell lineages by immunofluorescence staining and real-time quantitative RT-PCR. Total BMSCs and c-Kit(POS)/NKX2.5(POS) cells were treated with the Notch1 ligand Jagged1 and markers of cardiomyocyte, SMC, and endothelial cell differentiation were examined by immunofluorescence staining and real-time quantitative RT-PCR analysis.

Results: c-Kit(POS)/NKX2.5(POS) cells were present among total BMSC populations, and these cells did not express markers of adult cardiomyocyte, SMC, or endothelial cell lineages. c-Kit(POS)/NKX2.5(POS) BMSCs exhibited a multi-lineage differentiation potential similar to total BMSCs. Following sorting, the c-Kit level in c-Kit(POS)/NKX2.5(POS) BMSCs was 84.4%. Flow cytometry revealed that Notch1 was the predominant Notch receptor present in total BMSCs and c-Kit(POS)/NKX2.5(POS) BMSCs. Total BMSCs and c-Kit(POS)/NKX2.5(POS) BMSCs overexpressing NICD had active Notch1 signalling accompanied by differentiation into cardiomyocyte, SMC, and endothelial cell lineages. Treatment of total BMSCs and c-Kit(POS)/NKX2.5(POS) BMSCs with exogenous Jagged1 activated Notch1 signalling and drove multi-lineage differentiation, with a tendency towards cardiac lineage differentiation in c-Kit(POS)/NKX2.5(POS) BMSCs.

Conclusions: c-Kit(POS)/NKX2.5(POS) cells exist in total BMSC pools. Activation of Notch1 signalling contributed to multi-lineage differentiation of c-Kit(POS)/NKX2.5(POS) BMSCs, favouring differentiation into cardiomyocytes. These findings suggest that modulation of Notch1 signalling may have potential utility in stem cell translational medicine.

Figure 1

Representative morphology of total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. (A) Bone marrow mesenchymal stem cells (BMSCs) gradually showed fibroblast-like shapes by day 3 after plating. (B) Cells reached confluency at day 14. (C) High magnification of confluent BMSCs. (D) A single c-Kit^POS/NKX2.5^POS BMSC showing a polygonal shape at 3 days after plating. (E) The cell clone at confluency after 21 days. (F) High magnification of confluent c-Kit^POS/NKX2.5^POS BMSCs. Scale bar: (A, B, D, E) 30 μm, and (C, F) 60 μm. Original magnifications: (A, B, D, E) ×100, and (C, F) ×200.

Figure 2

Phenotypic identification of c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. (A) Confluent, c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells (BMSCs) were passaged and third-passage cells were subjected to immunofluorescence staining for c-Kit, NKX2.5, and α-sarcomeric actin (α-SA). Unsorted total BMSCs served as control. (B) Assessment of c-Kit expression in total BMSCs before and after magnetic activated cell sorting by flow cytometric analysis. (C) Semi-quantitative RT-PCR analysis of markers for stem cells (c-Kit), cardiomyocytes (NKX2.5, GATA-4, and cardiac troponin T (cTnT)), smooth muscle cells ( SM22α) and endothelial cells (von Willebrand factor (vWF)). A sample from a neonatal Sprague–Dawley rat heart was used as positive control. (D) Flow cytometry was performed to detect the presence of Notch1 to Notch4 receptors. For immunofluorescence staining, target proteins were detected with fluorescein isothiocyanate (FITC)-conjugated or phycoerythrin (PE)-conjugated IgG. Images were captured by fluorescence microscopy. Scale bar: 50 μm. For semi-quantitative RT-PCR analysis, β-actin mRNA was used as an internal control. For flow cytometric analysis, isotype control IgG was used to set the threshold.

Figure 3

Functional identification of total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. (A) Uninduced and induced bone marrow mesenchymal stem cells (BMSCs) and c-Kit^POS/NKX2.5^POS BMSCs were stained after 21 days of osteogenic or adipogenic differentiation with Alizarin Red and Oil Red O, respectively. (B) For smooth muscle cell differentiation, BMSCs were treated with transforming growth factor-β1-supplemented medium for 8 days, and then subjected to immunofluorescence (IF) staining for smooth muscle myosin heavy chain (SM-MHC). Target proteins were detected with fluorescein isothiocyanate-conjugated IgG. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Original magnification: (A) ×200. Scale bar: 30 μm.

Figure 4

Forced NICD expression activation of Notch1 signalling in total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Total bone marrow mesenchymal stem cells (BMSCs) and c-Kit^POS/NKX2.5^POS BMSCs were infected with NICD-Ad or NC-Ad, with uninfected cells serving as an additional control (MOCK). After 8 days, the cells were harvested for (A) immunofluorescence staining for notch receptor intracellular domain (NICD) and Hes1, and (B) real-time quantitative RT-PCR analysis of Hes1 mRNA expression. For immunofluorescence staining, target proteins were detected with fluorescein isothiocyanate-conjugated IgG. (C) Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Images were captured by confocal microscopy and merged. Scale bar: 50 μm. Triplicate experiments were performed for real-time quantitative RT-PCR analysis (n = 3). ***P <0.001 versus other groups. NC-Ad, negative control-expressing adenovirus; NICD-Ad, NICD-expressing adenovirus.

Figure 5

Forced NICD expression induces multilineage differentiation of total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Total bone marrow mesenchymal stem cells (BMSCs) and c-Kit^POS/NKX2.5^POS BMSCs were infected with NICD-Ad or NC-Ad, with uninfected cells serving as an additional control (MOCK). After 8 days, the cells were harvested for (A) real-time quantitative RT-PCR analysis and (B) immunofluorescence staining for markers for cardiomyocytes (NKX2.5 and cardiac troponin T (cTnT)), smooth muscle cells (SM22α), and endothelial cells (von Willebrand factor (vWF)). Triplicate experiments were performed for real-time quantitative RT-PCR analysis (n = 3). *P <0.05, ***P <0.001, ^# P <0.05, ^## P <0.01, ^### P <0.001 versus other associated groups. For immunofluorescence staining, images were captured by confocal microscopy. Target proteins were detected with phycoerythrin-conjugated or fluorescein isothiocyanate-conjugated IgG. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Scale bar: 50 μm. EC, endothelial cell; NC-Ad, negative control-expressing adenovirus; NICD, notch receptor intracellular domain; NICD-Ad, NICD-expressing adenovirus; SMC, smooth muscle cell.

Figure 6

Exogenous Jagged1 activation of Notch1 signalling in total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Total bone marrow mesenchymal stem cells (BMSCs) and c-Kit^POS/NKX2.5^POS BMSCs were classified into Jagged1, IgG, N-(2S-(3,5-difluorophenyl)acetyl)-l-alanyl-2-phenyl-1,1-dimethylethyl ester-glycine (DAPT), dimethyl sulfoxide (DMSO), and MOCK treatment groups (see Methods). After 8 days, activation of Notch1 signalling was assessed by immunofluorescence staining of notch receptor intracellular domain (NICD) and Hes1 protein expression (A) and quantitative RT-PCR analysis of Hes1 mRNA expression (B). For immunofluorescence staining, target proteins were detected with fluorescein isothiocyanate-conjugated IgG. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Scale bar: 50 μm. Representative images were captured by confocal microscopy. Triplicate experiments were performed for real-time quantitative RT-PCR analysis (n = 3). ***P <0.001 versus other groups.

Figure 7

Jagged1 activation of Notch1 signalling induces multilineage differentiation of total and c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Total bone marrow mesenchymal stem cells (BMSCs) and c-Kit^POS/NKX2.5^POS BMSCs were classified into Jagged1, IgG, N-(2S-(3,5-difluorophenyl)acetyl)-l-alanyl-2-phenyl-1,1-dimethylethyl ester-glycine (DAPT), dimethyl sulfoxide (DMSO), and MOCK treatment groups (see Methods). After 8 days, the expression of differentiation markers for cardiomyocytes (NKX2.5 and cardiac troponin T (cTnT)), smooth muscle cells (SM22α), and endothelial cells (von Willebrand factor (vWF)) were examined by (A) real-time quantitative RT-PCR and (B) immunofluorescence staining. Triplicate experiments were performed for real-time quantitative RT-PCR analysis (n = 3). *P <0.05, **P <0.01, ^## P <0.01, ^### P <0.001 versus other associated groups. For immunofluorescence staining, target proteins were detected with phycoerythrin-conjugated or fluorescein isothiocyanate-conjugated IgG. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Representative images were captured by confocal microscopy. Scale bar: 50 μm.

Figure 8

Activation of Notch1 signalling favours cardiac differentiation of c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Positive immunofluorescence (IF) staining was assessed for cardiac troponin T (cTnT), smooth muscle cells (SM22α), and von Willebrand factor (vWF) in 10 representative fields of Jagged1-treated c-Kit^POS/NKX2.5^POS bone marrow mesenchymal stem cells. Ratios to 4′,6-diamidino-2-phenylindole-positive cells represent differentiation potentials for cardiomyocytes, smooth muscle cells, and endothelial cells.