Promyelocytic leukemia zinc-finger induction signs mesenchymal stem cell commitment: identification of a key marker for stemness maintenance?

Abstract

Introduction: Mesenchymal stem cells (MSCs) are an attractive cell source for cartilage and bone tissue engineering given their ability to differentiate into chondrocytes and osteoblasts. However, the common origin of these two specialized cell types raised the question about the identification of regulatory pathways determining the differentiation fate of MSCs into chondrocyte or osteoblast.

Methods: Chondrogenesis, osteoblastogenesis, and adipogenesis of human and mouse MSC were induced by using specific inductive culture conditions. Expression of promyelocytic leukemia zinc-finger (PLZF) or differentiation markers in MSCs was determined by RT-qPCR. PLZF-expressing MSC were implanted in a mouse osteochondral defect model and the neotissue was analyzed by routine histology and microcomputed tomography.

Results: We found out that PLZF is not expressed in MSCs and its expression at early stages of MSC differentiation is the mark of their commitment toward the three main lineages. PLZF acts as an upstream regulator of both Sox9 and Runx2, and its overexpression in MSC enhances chondrogenesis and osteogenesis while it inhibits adipogenesis. In vivo, implantation of PLZF-expressing MSC in mice with full-thickness osteochondral defects resulted in the formation of a reparative tissue resembling cartilage and bone.

Conclusions: Our findings demonstrate that absence of PLZF is required for stemness maintenance and its expression is an early event at the onset of MSC commitment during the differentiation processes of the three main lineages.

Figure 1

Absence of PLZF expression in cells signs their stemness state. (A) Expression profile of PLZF during the chondrogenic differentiation of MSCs as assessed by microarray analysis. At day 0, absence of PLZF expression was defined as 1, and this was used to calculate the relative expression at the remaining time points. (B) PLZF expression profile in human MSCs, ESCs, iPSs, adipose tissue, skeletal muscle, and trachea as reported in the Amazonia website. The signal intensity is shown on the y axis as arbitrary units determined by the GCOS 1.2 software (Affymetrix). When indicated, the mean expression is significantly different from that in MSCs, hESCs, and hiPS. (C) PLZF-related network was reconstructed by using the Ingenuity Pathways Analysis Software (IPA) software where PLZF acts upstream of Runx2 and regulates downstream genes. Blue-shaded genes are genes identified to be involved in chondrogenesis, grey-shaded genes either involved in osteogenesis or adipogenesis, and unshaded genes are those associated with the identified genes based on pathway analysis.

Figure 2

PLZF expression profile in the course of MSC differentiation into the three main lineages. Expression profile of PLZF during the differentiation of human (A) bone marrow- (hBM-MSC), (B) synovium- (hSYNO), and (C) adipose-derived (hASC) MSCs toward the chondroblastic, osteoblastic, and adipocytic lineages by using RT-qPCR. RT-qPCR data represent the mean ± SEM of three independent experiments.

Figure 3

Expression level of PLZF in primary mMSC, C3, and C3-PLZF cells. (A) PLZF expression in mMSCs during chondrogenic, adipogenic, and osteogenic differentiation as assessed by RT-qPCR. (B) Expression profile of PLZF in undifferentiated (D0) and differentiated C3 cells into three lineages; adipogenic (Adipo), osteogenic (Osteo), and chondrogenic cells (Chondro). (C) Expression level of PLZF in undifferentiated C3 cells and in a clone of C3 cells with high PLZF expression (C3-PLZF), as quantified by using RT-qPCR. (D) Immunofluorescence analysis of C3 and C3-PLZF by using an anti-PLZF antibody. (E) Expression profile of the master regulators of the chondrogenic, adipogenic, or osteogenic programs, Sox9, PPARγ, or Runx2, respectively, in the C3 or C3-PLZF cells.

Figure 4

Role of PLZF on MSC differentiation. (A) Effect of PLZF overexpression on MSC chondrogenic differentiation after 21 days in micropellet culture. Expression of Sox9 and the cartilage marker Col2A1 in C3 and C3-PLZF, cells assessed by RT-qPCR and immunohistochemical analysis for collagen II expression in pellets. (B) Effect of PLZF overexpression on MSC adipogenic potential. Expression level of PPAR-γ and the adipocyte marker (FABP4) determined by RT-qPCR and staining of lipid droplets with oil red O. (C) Effect of PLZF overexpression on MSC osteogenic potential. Expression level of Runx2 and the osteoblast marker osteocalcin (OC), determined by RT-qPCR, and alizarin red S staining for mineralization. RT-qPCR data represent the mean ± SEM of three independent experiments.

Figure 5

C3-PLZF cell injection leads to bone formation in an osteoarticular defect model in immunocompetent mice. (A) Quantification of the bone density (g/cm³). (B) Histologic analysis of tibial sections stained with Safranin O and Fast green. On the top, a representative control tibia. (C) Goldner trichrome histologic staining. Tibias were recovered 28 days later.

Figure 6

C3-PLZF cell injection leads to cartilage and bone formation in an osteoarticular defect model in SCID mice. (a) A representative histologic section of the tibia of a SCID mouse that has received naïve C3 cells stained with Safranin O and Fast green. (b) Distribution of CM-DiI-labeled C3 shown by fluorescence microscopy as red fluorescent cells. (c) Collagen II-positive staining on tibia sections. (d) Histologic sections of tibias stained with Safranin O and Fast green. C3-PLZF cell injection leads to cartilage formation in an osteoarticular defect model in SCID mouse. (e) Distribution of CM-DiI-labeled C3-PLZF shown by fluorescence microscopy as red fluorescent cells. (f) Collagen II-positive staining on tibia sections.