Cyclin-Dependent Kinase 1 Inhibition Potentiates the Proliferation of Tonsil-Derived Mesenchymal Stem Cells by Delaying Cellular Senescence

Abstract

Mesenchymal stem cells (MSCs) have been widely used in tissue regeneration and stem cell therapy and are currently being tested in numerous clinical trials. Senescence-related changes in MSC properties have attracted considerable attention. Senescent MSCs exhibit a compromised potential for proliferation; senescence acts as a stress response that prevents the proliferation of dysfunctional cells by inducing an irreversible cell cycle arrest. Here, we established a senescent MSC model using senescence-associated β-galactosidase, proliferation, and cell cycle assays. We further identified novel biomarker candidates for old, senescent tonsil-derived MSCs (TMSCs) using transcriptomics. A plot of the cellular senescence pathway showed cyclin-dependent kinase 1 (CDK1; +8-fold) and CDK2 (+2-fold), and transforming growth factor beta 2 (TGFB2; +2-fold) showed significantly higher expression in old TMSCs than in young TMSCs. The CDK family was shown to be related to cell cycle and proliferation, as confirmed by quantitative RT-PCR. As replicative senescence of TMSCs, the gene and protein expression of CDK1 was significantly increased, which was further validated by inhibiting CDK1 using an inhibitor and siRNA. Taken together, we suggest that the CDK1 can be used as a selective senescence biomarker of MSCs and broaden the research criteria for senescent mechanisms.

Figure 1

Changes in senescent markers with serial passaging of TMSCs. (a) Morphological changes of young and old TMSC groups were determined by measuring cell length and width. Young TMSCs exhibited a polygonal morphology, whereas old TMSCs formed dispersed shapes. (b) Senescent cells were identified as blue-stained cells under an optical microscope. Magnification: ×200; scale bar, 50 μm. The orange arrowhead indicates stained cells. Senescent cells were identified as blue-stained cells and were calculated by the average absorbance intensity of SA-β-gal staining (405 nm) from 5 randomly selected fields between young and old TMSCs. (c) Gene expression of hTERT and TRF-1 which are telomere length markers in TMSCs. (d) Cell proliferation profile of young and old TMSCs during 64 h. The blue triangle line indicates young TMSCs, and the orange rectangle line indicates old TMSCs. (e, f) The TMSCs from each experimental group were stained with PI staining solution and analyzed their cell cycle stage. The p values were considered statistically significant at ^∗∗p < 0.01 and ^∗∗∗p < 0.001.

Figure 2

Changes in the expression of stem cell surface markers in TMSCs with serial passaging. Surface markers of TMSCs were characterized using FACS analysis. Both hematopoietic (CD14, CD34, and CD45) and primitive (CD73, CD90, and CD105) cell surface markers were examined. Young and old TMSCs were marked as blue and orange peaks, respectively.

Figure 3

Transcriptomic profiles of young and old TMSCs. Whole-genome sequence data from young and old TMSCs are presented. (a) A cell cycle-related partial heatmap of hierarchical clustering analysis results indicates differentially expressing genes (rows) between the young and old TMSCs. Blue and orange circles indicate up- and downregulated genes, respectively, in old TMSCs compared with young TMSCs. For all comparisons, changes in gene expression are depicted as a heatmap. (b) 3D MDS plots for microarray gene expression data compared to young and old TMSCs were presented. (c) 3D dendrogram depicts the results of hierarchical clustering analysis of the interclass correlation between young and old TMSCs, confirming the classification of an interclass between the two groups. (d) The molecular function distribution of GO terms for DEGs between young and old TMSCs was annotated according to the ontology categories: molecular function (MF). The x- and y-axes indicate the number of DEGs and GO term gene classification, respectively (^∗∗∗p < 0.001). (e) All KEGG pathways were first classified into 10 categories. (f) These categories were the top 20 terms of the KEGG pathway according to the level of significance.

Figure 4

Changes of CDK family gene expression with replicative senescence in TMSCs. Changes in the expression of cellular senescence and cell cycle gene induced by replicative subculture were confirmed by quantitative RT-PCR (a) CDK1, (b) CDK2 (c) TGFB2, (d) CCNA2, (e) CCNB1, and (f) CCNE2. The bar graph represents relative fold changes in gene expression, determined by cT value of each sample normalized to that of respective β-actin. The p values were considered statistically significant (^∗∗∗p < 0.001, ^∗p < 0.05; n.s.: not significant).

Figure 5

Changes of CDK family protein expression with replicative senescence in TMSCs. (a) Western blot analysis of cell cycle markers which are (b) CDK1, (c) CDK2, (d) cyclin B1, and (e) cyclin E2. The bar graph represents relative fold changes in gene expression, normalized by respective GAPDH. The p values were considered statistically significant (^∗∗∗p < 0.001, ^∗p < 0.05; n.s.: not significant).

Figure 6

Confirmation of CDK1 function in TMSCs. The relative fold change bar graph of CDK1 gene expression with (a) CDK1/2 inhibitor III and (d) CDK1 siRNA. (b, e) Comparison of cell confluency for each group. Cell proliferation profile of (c) old TMSCs vs. CDK1-inhibited TMSCs and (f) old TMSCs vs. old TMSCs transfected with CDK1 siRNA. The p values were considered statistically significant (^∗∗∗p < 0.001); scale bar, 200 μm.