Twist1 regulates cellular senescence in mesenchymal stem cells

Abstract

Purpose: Mesenchymal stem cells (MSCs) are promising cells to treat skeletal related diseases such as osteoarthritis (OA). One of the limitations that reduces the therapeutic use of MSCs for OA cell-based treatments is cellular senescence. Cellular senescence is a state of the cell in which cells stop proliferating, while still be metabolically active. By acquiring a specific senescence-associated secretory phenotype (SASP), senescent MSCs can contribute to cartilage degeneration and chronic inflammation. The molecular mechanism behind cellular senescence in MSCs remains largely unknown. The transcription factor TWIST1 is suggested to play a role in cellular senescence, since its expression is reduced in slow proliferating MSCs and we previously demonstrated that high TWIST1-expressing MSCs have an enhanced expansion capacity. The aim of this research is therefore to determine the role of TWIST1 in regulating cellular senescence of MSCs. Methods: Human MSCs were obtained from bone chips from leftover iliac crest of patients undergoing cleft palate reconstructive surgery (N=4, 9-13 years). MSCs were expanded in alpha-MEM containing 10% serum and 1 ng/ml FGF2. To determine the effect of TWIST1 silencing on senescence, MSCs were treated with 15 nM TWIST1 siRNA and Lipofectamine RNAMAX Transfection Reagent (siTWIST1). Scramble siRNA and untreated cells were used as controls. To overexpress TWIST1, MSCs were transduced with a third generation TWIST1 overexpression lentiviral construct. Untreated MSCs and MSCs which overexpress GFP or an empty vector were used as control. The transduction efficiency, of GFP-positive MSCs, was determined using fluorescent microscopy. To determine cellular senescence, we determined senescence-associated beta-galactosidase (SA-β-gal) activity and quantified negative, low and high positive cells. In addition gene expression for IL8, IL6, IL1B, P21 and P16 were analyzed to assess cellular senescence in MSCs. Gene expression was normalized using the best housekeeper index of HPRT, RSP27a and GAPDH. The expansion capacity of the MSCs was determined by cell count. To analyze statistical significance, linear mixed model was applied, the different conditions were considered as fixed parameter and the donors as random parameter. Bonferroni post hoc test was performed to correct for multiple comparisons. Results: TWIST1 transduced MSCs had an average of 151-fold increase in TWIST1 expression detected by RT-PCR, a reduced percentage of SA-B-gal positive cells (71% reduction) and a reduced expression of the SASP related genes IL6 (1.8-fold reduction), IL8 (2.4-fold reduction) and IL1B (1.3-fold reduction) compared to transduction control (Figure 1). In addition, the expansion capacity of MSCs overexpressing TWIST1 was increased, resulting in a 1.5-fold increase in cell number after 4 days. These data indicate that TWIST1 delays cellular senescence in MSCs. Next, we determined the effect of TWIST1 silencing on cellular senescence associated markers. A high percentage of siTWIST1 MSCs were high positive for SA-B-gal (9.5-fold compared to controls, Figure 2A) and cells had a reduced expansion capacity compared to control conditions (20-fold reduction in cell number after 14 days). These data suggest that silencing TWIST1 in MSCs increases cellular senescence. Surprisingly, the expression of SASP related genes IL6 and IL8 was reduced in siTWIST cells compared to untreated control (IL6, 1.6-fold; IL8, 7.2-fold reduction, Figure 2B) and no induction of IL1B was observed in the siTWIST1 MSCs. Low IL6, IL8 and IL1B expression in senescent cells was recently described as mitochondrial dysfunction-associated senescence (MiDAS). We then further analyzed siTWIST1 MSCs for the expression of senescent markers not associated with a SASP phenotype and found that P16 and P21, associated to the cell cycle regulation, were increased after TWIST1 silencing (5.9-fold and 1.6-fold, respectively) in accordance with a senescent phenotype. Overall, these data indicate that TWIST1 silencing results in cellular senescence of MSCs with an uncommon SASP expression. Conclusions: Our Results indicate that TWIST1 plays a role in a special type of cellular senescence, previously described as MiDAS, with a unique SASP expression profile, lacking IL6, IL8 and IL1B upregulation. In addition these data show that different type of senescent cells can have a different SASP expression and therefore potentially have different effects on the microenvironment. Here, we show how TWIST1 may be a target to modulate cellular senescence to improve MSC-based OA therapies.View Large Image Figure ViewerDownload Hi-res image Download (PPT)