Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Mar 4;8:81.
doi: 10.3389/fcell.2020.00081. eCollection 2020.

Significance of MEF2C and RUNX3 Regulation for Endochondral Differentiation of Human Mesenchymal Progenitor Cells

Affiliations
Free PMC article

Significance of MEF2C and RUNX3 Regulation for Endochondral Differentiation of Human Mesenchymal Progenitor Cells

Simon I Dreher et al. Front Cell Dev Biol. .
Free PMC article

Abstract

Guiding progenitor cell development between chondral versus endochondral pathways is still an unachieved task of cartilage neogenesis, and human mesenchymal progenitor cell (MPC) chondrogenesis is considered as a valuable model to better understand hypertrophic development of chondrocytes. Transcription factors Runx2, Runx3, and Mef2c play prominent roles for chondrocyte hypertrophy during mouse development, but little is known on the importance of these key fate-determining factors for endochondral development of human MPCs. The aim of this study was to unravel the regulation of RUNX2, RUNX3, and MEF2C during MPC chondrogenesis, the pathways driving their expression, and the downstream hypertrophic targets affected by their regulation. RUNX2, RUNX3, and MEF2C gene expression was differentially regulated during chondrogenesis of MPCs, but remained low and unregulated when non-hypertrophic articular chondrocytes were differentiated under the same conditions. RUNX3 and MEF2C mRNA and protein levels rose in parallel to hypertrophic marker upregulation, but surprisingly, RUNX2 gene expression changed only by trend and RUNX2 protein remained undetectable. While RUNX3 expression was driven by TGF-β and BMP signaling, MEF2C responded to WNT-, BMP-, and Hedgehog-pathway inhibition. MEF2C but not RUNX3 levels correlated significantly with COL10A1, IHH, and IBSP gene expression when hypertrophy was attenuated. IBSP was a downstream target of RUNX3 and MEF2C but not RUNX2 in SAOS-2 cells, underlining the capacity of RUNX3 and MEF2C to stimulate osteogenic marker expression in human cells. Conclusively, RUNX3 and MEF2C appeared more important than RUNX2 for human endochondral MPC chondrogenesis. Pathways altering the speed of chondrogenesis (FGF, TGF-β, BMP) affected RUNX2 or RUNX3, while pathways changing hypertrophy (WNT, PTHrP/HH) regulated mainly MEF2C. Taken together, reduction of MEF2C levels is a new goal to shift human cartilage neogenesis toward the chondral pathway.

Keywords: BMP; FGF; PTHrP; RUNX2; WNT; cartilage; chondrogenesis; hypertrophy.

Figures

FIGURE 1
FIGURE 1
Transcription factor signature of expanded MPCs. MPCs and ACs were harvested at passages 2–3 of expansion (n = 4 donors). (A) RUNX2, (B) RUNX3, (C) MEF2C, and (D) SOX9 mRNA levels were determined by qPCR and levels were referred to reference genes HNRPH1 and CPSF6. Data are shown as box plots, with each box representing the interquartile range (IQR) extending between the 25th and 75th percentiles, and lines inside the boxes representing the median. Whiskers extend to minimum and maximum values. *, p ≤ 0.05, Mann–Whitney U test. (E) RUNX2, RUNX3, MEF2C, and SOX9 protein levels were determined by Western blot analysis in cell lysates from three donors (MPC1-3, AC1-3) on the same gel, β-actin levels were determined as internal reference, and SAOS-2 cell lysate was used as positive control on the same gel (PCtrl).
FIGURE 2
FIGURE 2
Characterization of MPC and AC differentiation. MPC and AC pellets were subjected to chondrogenic induction for 6 weeks. Samples harvested in weekly intervals were subjected to gene expression analysis by qPCR for indicated genes (A–E), and levels were referred to reference genes HNRPH1 and CPSF6 (n = 4 donors). (F) Alkaline phosphatase (ALP) activity was determined in culture supernatants pooled from five pellets (n = 4 donors). Data are shown as mean ± standard deviation over the time course of 6 weeks. MPC versus AC; Mann–Whitney U test; *, p ≤ 0.05 versus d0 (d7 ALP activity) ANOVA Tukey; #, p ≤ 0.05. (G–I): Paraffin sections of AC and MPC pellets at day 42 were stained for (G) collagen type II and (I) collagen type X by immunohistochemistry or with (H) Safranin O/Fast Green to visualize proteoglycan deposition. Scale bars represent 200 μm.
FIGURE 3
FIGURE 3
Transcription factor regulation during MPC chondrogenesis versus AC redifferentiation. MPC and AC pellets were subjected to chondrogenic induction for 6 weeks. Samples were harvested in weekly intervals, and transcription factor levels were determined on mRNA and protein levels (n = 4 donors). Indicated gene expression levels were determined by qPCR (A–D) and referred to reference genes HNRPH1 and CPSF6. Data show mean ± standard deviation. MPC versus AC, Mann–Whitney U test; *, p ≤ 0.05 versus d0 ANOVA Tukey; #, p ≤ 0.05. RUNX2, RUNX3, MEF2C, and SOX9 protein levels were determined by Western blot analysis, β-actin levels were determined as internal reference, and SAOS-2 cell lysate was used as positive control on the same gel (Pos. Ctrl). Shown is one representative of three to four experiments.
FIGURE 4
FIGURE 4
Effects of TGF-β withdrawal on RUNX2, RUNX3, and MEF2C. MPC pellets were subjected to chondrogenic induction for 3 weeks before TGF-β was omitted from chondrogenic medium for half of the pellets, while other ingredients remained constant until week 6 (n = 5 donors). Expression of indicated genes was determined at day 42 by qPCR (A,B), and levels were referred to reference genes HNRPH1 and CPSF6. ALP activity was quantified in culture supernatants (n = 4 donors) (A). Expression levels of indicated genes and ALP activity are referred to control cultures (dashed lines). Box plots were generated as described in Figure 1, outliers are depicted as ∘, and extreme values as □. *, p ≤ 0.05, Mann–Whitney U test compared to control. (C) RUNX3 and (D) MEF2C protein levels were determined by Western blot analysis, β-actin levels were determined as internal reference. One representative of three experiments is shown.
FIGURE 5
FIGURE 5
Effects of BMP-receptor inhibition on RUNX2, RUNX3, and MEF2C. MPC pellets were subjected to chondrogenic induction for 2 weeks before treatment with 10 μM dorsomorphin or 0.1% DMSO until week 6 (n = 4 donors). Expression of indicated genes was determined by qPCR (A,B), and levels were referred to reference genes HNRPH1 and CPSF6. ALP activity was quantified in culture supernatants (A). Expression levels of indicated genes and ALP activity are referred to DMSO cultures (dashed lines). Box plots were generated as described in Figure 1. *, p ≤ 0.05, Mann–Whitney U test compared to control.
FIGURE 6
FIGURE 6
Effects of FGF-receptor inhibition on RUNX2, RUNX3, and MEF2C. MPC pellets were subjected to chondrogenic induction for 6 weeks and treated with 250 nM of the FGFR-Inhibitor PD173074 or with the corresponding amount of DMSO from day 7 onward. Protein levels of pERK1/2 and total ERK1/2 as reference were detected by Western blot (A). Shown is one representative of three experiments. (B) Gene expression of COL2A1 was determined during 42 days of chondrogenesis with PD173074 (black line) or DMSO control (dashed line) (n = 3 donors), (C,D) expression of indicated genes was determined by qPCR at day 21 of chondrogenesis, and levels were referred to reference genes HNRPH1 and CPSF6. ALP activity was quantified in culture supernatants (C). Expression levels of indicated genes and ALP activity are referred to DMSO cultures (dashed lines). Box plots were generated as described in Figure 1. *, p ≤ 0.05, Mann–Whitney U test compared to control. (E,F) RUNX3 and MEF2C protein levels were determined in samples with DMSO or PD173074 at indicated timepoints, β-actin levels were determined as internal reference. One representative of three experiments is shown.
FIGURE 7
FIGURE 7
Effects of WNT-signaling inhibition on RUNX2, RUNX3, and MEF2C. MPC pellets were subjected to chondrogenic induction for 5 weeks and treated with 2 μM IWP-2 or the corresponding amount of DMSO from day 14 (n = 5 donors). Expression of indicated genes was determined by qRT-PCR (A,B), and levels were referred to reference genes HNRPH1 and CPSF6. ALP activity was quantified in culture supernatants (A). Expression levels of indicated genes and ALP activity are referred to DMSO cultures (dashed lines). Box plots were generated as described in Figure 1, outliers are depicted as ∘, and extreme values as □. *, p ≤ 0.05, Mann–Whitney U test. RUNX3 (C) and MEF2C (D) protein levels were determined in samples treated with IWP-2 (+) or with DMSO (–) by Western blot analysis, β-actin levels were determined as internal reference. Three representative experiments are shown.
FIGURE 8
FIGURE 8
Effects of pulsed PTHrP treatment on RUNX2, RUNX3 and MEF2C. MPC pellets were subjected to chondrogenic induction for 1 week before half of the pellets were stimulated with 2.5 nM PTHrP(1-34) for 6 h daily while the others received daily medium exchange until week 6 (n = 5 donors). Expression of indicated genes was determined by qPCR (A–C), and levels were referred to reference genes HNRPH1 and CPSF6. (A) Gene expression of the HH downstream target GLI1 (n = 3). ALP activity was quantified in culture supernatants (B). Expression levels of indicated genes and ALP activity are referred to DMSO cultures (dashed lines). Box plots were generated as described in Figure 1, outliers are depicted as ∘ and extreme values as □. *, p ≤ 0.05, Mann–Whitney U test compared to control. (D) Overview of pathways regulating RUNX2, RUNX3, and MEF2C during MPC chondrogenesis.
FIGURE 9
FIGURE 9
Correlation of RUNX2, RUNX3, and MEF2C with hypertrophic marker gene expression after lineage shift (WNT/HH inhibition). (A–I) Pearson correlation between the expression levels (% reference gene) of RUNX2, RUNX3, MEF2C and the hypertrophic genes COL10A1, IHH, and IBSP at indicated treatments, respectively.
FIGURE 10
FIGURE 10
Effects of siRNA-mediated knockdown of MEF2C, RUNX2, and RUNX3 in SAOS-2 osteosarcoma cells. SAOS-2 cells were transfected with siMEF2C, siRUNX2, siRUNX3, or a non-targeting control siRNA (siCtrl) 48 h prior to analysis (n = 6 biological replicates from four independent experiments). (A) Target-gene expression was determined by qPCR. (B) MEF2C, RUNX2, and RUNX3 protein levels were determined by Western blot analysis; β-actin levels were determined as internal reference. One representative of four biological replicates (two to three experiments) is shown. Expression of indicated genes was determined by qPCR (C–E), and levels were referred to reference genes HNRPH1 and CPSF6. Expression levels of indicated genes are referred to siCtrl cultures (dashed lines). Box plots were generated as described in Figure 1; outliers are depicted as ∘. Significance compared to siCtrl was determined by Mann–Whitney U test *, p ≤ 0.05, with Bonferroni correction.

Similar articles

See all similar articles

References

    1. Arnold M. A., Kim Y., Czubryt M. P., Phan D., Mcanally J., Qi X., et al. (2007). MEF2C transcription factor controls chondrocyte hypertrophy and bone development. Dev. Cell 12 377–389. 10.1016/j.devcel.2007.02.004 - DOI - PubMed
    1. Bonyadi Rad E., Musumeci G., Pichler K., Heidary M., Szychlinska M. A., Castrogiovanni P., et al. (2017). Runx2 mediated induction of novel targets ST2 and Runx3 leads to cooperative regulation of hypertrophic differentiation in ATDC5 chondrocytes. Sci. Rep. 7:17947. - PMC - PubMed
    1. Brady K., Dickinson S. C., Hollander A. P. (2015). Changes in chondrogenic progenitor populations associated with aging and osteoarthritis. Cartilage 6 30S–35S. 10.1177/1947603515574838 - DOI - PMC - PubMed
    1. Buckwalter J. A., Mankin H. J. (1997). Articular cartilage.2. Degeneration and osteoarthrosis, repair, regeneration, and transplantation. J. Bone Joint Surg. Am. 79A 612–632.
    1. Caron M. M., Emans P. J., Cremers A., Surtel D. A., Coolsen M. M., Van Rhijn L. W., et al. (2013). Hypertrophic differentiation during chondrogenic differentiation of progenitor cells is stimulated by BMP-2 but suppressed by BMP-7. Osteoarthritis Cartilage 21 604–613. 10.1016/j.joca.2013.01.009 - DOI - PubMed
Feedback