5-azacytidine improves the osteogenic differentiation potential of aged human adipose-derived mesenchymal stem cells by DNA demethylation

Abstract

The therapeutic value of adipose-derived mesenchymal stem cells (Ad-MSCs) for bone regeneration is critically discussed. A possible reason for reduced osteogenic potential may be an age-related deterioration of the Ad-MSCs. In long term in vitro culture, epigenomic changes in DNA methylation are known to cause gene silencing, affecting stem cell growth as well as the differentiation potential. In this study, we observed an age-related decline in proliferation of primary human Ad-MSCs. Decreased Nanog, Oct4 and Lin28A and increased Sox2 gene-expression was accompanied by an impaired osteogenic differentiation potential of Ad-MSCs isolated from old donors (>60 a) as compared to Ad-MSCs isolated from younger donors (<45 a). 5-hydroxymethylcytosine (5 hmC) and 5-methylcytonsine (5 mC) distribution as well as TET gene expression were evaluated to assess the evidence of active DNA demethylation. We observed a decrease of 5 hmC in Ad-MSCs from older donors. Incubation of these cells with 5-Azacytidine induced proliferation and improved the osteogenic differentiation potential in these cells. The increase in AP activity and matrix mineralization was associated with an increased presence of 5 hmC as well as with an increased TET2 and TET3 gene expression. Our data show, for the first time, a decrease of DNA hydroxymethylation in Ad-MSCs which correlates with donor-age and that treatment with 5-Azacytidine provides an approach which could be used to rejuvenate Ad-MSCs from aged donors.

Figure 1. Declined Proliferation and Osteogenic Differentiation in Ad-MSCs from old donors: Ki67 staining was performed as proliferation assay.

(A) Representative image of a merged immunofluorescent staining for Ki67 (red) and nuclei (blue) from young and old Ad-MSCs. (B) Quantification of Ki67-positive Ad-MSCs (N = 3, n = 5/age group) points to a reduced proliferation in aged Ad-MSCs. After 14 days of osteogenic differentiation of young and old Ad-MSCs osteoblasts function was assessed by measuring (C) AP activity (each group: N = 7, n = 3), (D–G) matrix mineralization and (H–J) gene expression changes. Matrix mineralization was determined by (D & F) von Kossa (each group: N = 5, n = 3) and (E & G) Alizarin red staining (each group: N = 6, n = 3). mRNA levels of the osteogenic transcription factors (H) Runx2 and (I) Osterix, as well as the osteogenic marker (J) Osteocalcin were determined by semi-quantitative RT-PCR (each group: N = 5, n = 3). GAPDH was used for normalization. *p<0.05, **p<0.01, ***p<0.001 (student’s t-test).

Figure 2. Altered expression of pluripotency genes in Ad-MSCs from old donors.

Basal expression levels of pluripotency genes were determined in Ad-MSCs from young and old donors (each group: N = 6, n = 3). mRNA levels of (A) Lin28A (2 donors from the young Ad-MSCs had to be excluded from the data set, due to artefacts in the melting curve), (B) Oct4, (C) Nanog and (D) Sox2 were determined by qRT-PCR. (E) Expression changes of Lin28A, Oct4, Nanog and Sox2 were confirmed on the protein level by Western blot analysis (N = 3/group). For both techniques GAPDH was used for normalization. *p<0.05, **p<0.01 (student’s t-test).

Figure 3. Distribution of 5-hydroxymethylcytosine (5 hmC) and 5-methylcytosine (5 mC) in Ad-MSCs.

Basal mRNA levels of (A) TET2 and (B) TET3 were determined by qRT-PCR in Ad-MSCs from young and old donors (each group: N = 6, n = 3). (C) Expression changes of TET2 and TET3 were confirmed on the protein level by Western blot analysis (N = 3/group). For both techniques GAPDH was used for normalization. (D) Young and old Ad-MSCs (each group: N = 3, n = 5) were stained for 5-hydroxymethylcytosine (5 hmC/red) and 5-methylcytosine (5 mC/green). Cell nuclei were counterstained with Hoechst 33342 (blue). (E&F) For quantification 5 hmC or 5 mC positive nuclei were counted. *p<0.05, **p<0.01 (student’s t-test).

Figure 4. 5-Azacytidine treatment increased TET2 and TET3 gene expression.

Young and old Ad-MSCs were treated with 5 µM 5-Azacytidine for 48 h. mRNA levels of (A) TET2, (B) TET3, (E) Sox2, (F) Lin28A (2 donors from the young Ad-MSCs had to be excluded from the data set, due to artefacts in the melting curve), (G) Oct4 and (H) Nanog were determined by qRT-PCR (N = 6, n = 3/age group). (I) Expression changes of TET2, TET3, Lin28A, Oct4, Nanog and Sox2 were confirmed on the protein level by Western blot analysis (Representative figure for N = 2/group). GAPDH was used for normalization. Data are presented as relative expression changes (log₁₀) of 5-Azacytidine treated Ad-MSCs compared to the corresponding untreated Ad-MSCs. Distribution of 5-hydroxymethylcytosine (5 hmC) and 5-methylcytosine (5 mC) in the 5-Azacytidine treated and untreated Ad-MSCs was determined by immunofluorescent staining (N = 3, n = 5/age group). Cell nuclei were counterstained with Hoechst 33342. For quantification (D) 5 hmC and (E) 5 mC positive nuclei were counted. Data are represented as log₁₀ (fold of control). *p<0.05, **p<0.01, ***p<0.001 as compared to untreated cells (student’s t-test).

Figure 5. 5-Azacytidine treatment improves osteogenic differentiation of Ad-MSCs from old donors.

Ki67 staining was performed to assess the proliferation capacity of young and old Ad-MSCs after treatment with 5 µM 5-Azacytidine for 48 h. (A) Representative image of a merged immunofluorescent staining for Ki67 (red) and nuclei (blue). (B) Quantification of Ki67-positive Ad-MSCs (N = 3, n = 5/age group) points to an induced proliferation in aged Ad-MSCs after treatment with 5-Azacytidine. Young and old Ad-MSCs were pre-treated with 5 µM 5-Azacytidine for 48 h before osteogenic differentiation was induced. In the differentiated cells osteoblasts function was assessed by measuring (C) AP activity (N = 7, n = 3/age group), (D–G) matrix mineralization and (H–J) gene expression changes. Matrix mineralization was determined by (D & F) von Kossa (N = 4, n = 3/age group) and (E & G) Alizarin red staining (N = 6, n = 3/age group). mRNA levels of the osteogenic transcription factors (H) Runx2 and (I) Osterix, as well as the osteogenic marker (J) Osteocalcin were determined by semi-quantitative RT-PCR (N = 5, n = 3/age group). GAPDH was used for normalization. All data are presents as relative changes (log₁₀) of 5-Azacytidine pre-treated differentiated Ad-MSCs compared to the corresponding differentiated Ad-MSCs without pre-treatment.*p<0.05, **p<0.01, ***p<0.001 as compared to differentiated Ad-MSCs without pre-treatment (student’s t-test).