Adult stem cells hold great promise for use in tissue repair and regeneration. Recently, adipose tissue-derived stem cells (ADSCs) were found to be an appealing alternative to bone marrow stem cells (BMSCs) for bone tissue engineering. The main benefit of ADSCs is that they can be easily and abundantly available from adipose tissue. However, our prior study discovered an important phenomenon that BMSCs have greater osteogenic potential than ADSCs in vitro and epigenetic regulation plays a critical role in runt-related transcription factor 2 (Runx2) expression and thus osteogenesis. In this study, we aimed to improve the osteogenic potential of ADSCs by histone deacetylase inhibitor sodium butyrate (NaBu). We found that NaBu promoted rat ADSC osteogenic differentiation by altering the epigenetic modifications on the Runx2 promoter.
Extensive injury, congenital malformations or diseases, which cause large bone defects, create a major challenge to orthopedic surgeons, and sometimes bone tissue regeneration is needed. Although the use of embryonic stem cells is appealing because of their pluripotentiality, their practical use is limited due to ethical concerns and regulation. Mesenchymal stem cells (MSCs) are multipotent adult stem cells with limited self-renewal capacity that can be isolated from bone marrow, adipose tissue and cord blood, and so on (Guan et al. 2012). Due to their ability to differentiate into osteoblastic, chondrogenic, myogenic and adipogenic cells in vitro and in vivo, MSCs are considered an important source for regenerative medicine and provide a unique model to better understand early differentiation events (Mundra et al. 2013). Adipose-derived stem cells (ADSCs) are easily and abundantly available and capable of differentiating into multiple mesenchymal cell types, such as adipocytes, chondrocytes, osteoblasts and myoblasts (Al et al. 2011; Musumeci et al. 2011; Vishnubalaji et al. 2012; Wan et al. 2012). Therefore, ADSCs represent a kind of promising stem cells for bone tissue engineering.
Osteogenic differentiation of MSCs is a complex process associated with numerous signaling pathways and transcription factors. Runt-related transcription factor 2 (Runx2) is essential for osteoblast differentiation and bone formation. Humans with heterozygous mutations or deletions of Runx2 develop cleidocranial dysplasia (Lee et al. 1997; Gersbach et al. 2004). Overexpression of Runx2 can induce and upregulate the expression of multiple osteoblast-specific genes in non-osteogenic cells (Zhang et al. 2006; Phillips et al. 2007). Thus, the expression level of Runx2 is responsible for the MSC osteogenesis capacity; however, the precise molecular pathways by which osteogenesis regulators act on Runx2 and coordinate osteogenesis is still largely unknown.
In recent years, epigenetic regulation has been identified as an important mechanism of stem cell differentiation. Scientists have also elucidated that unique patterns of DNA methylation and histone modifications play an important role in the induction of MSC differentiation toward specific lineages (Li et al. 2011). Histone deacetylase (HDAC) inhibitors have multiple roles, including inhibition of proliferation and induction of differentiation in various in vitro and in vivo models. It has been reported that several HDAC inhibitors promote osteoblast maturation and the expression of osteoblast-specific genes through upregulation of Runx2 activity in MC3T3-E1 pre-osteoblast cells (Schroeder & Westendorf 2005). However, the precise mechanism has not been fully elucidated.
Sodium butyrate (NaBu) is the sodium salt of the short-chain fatty acid butyric acid (Ghosh et al. 2012). Butyrate is produced by anaerobic bacterial fermentation of dietary fibers and is known to play a key role in the homeostasis of the gastrointestinal tract. NaBu was shown to inhibit histone deacetylases (HDACs) at therapeutic concentrations. In this study, we aim to examine the effect of NaBu on osteogenic differentiation of ADSCs and elucidate the precise mechanisms that direct ADSC osteogenic differentiation.
Materials and methods
All animals and experimental procedures were approved by the local Institutional Animal Care and Use Committee complying with the “Guide for the Care and Use of Laboratory Animals” published by the National Academy Press (NIH Publication No. 85e23, revised 1996).
Isolation of adipose-derived stem cells
Adipose-derived stem cells were harvested from 4-week-old male Sprague Dawley (SD) rats. The rats were killed by cervical dislocation and adipose tissue in the inguinal groove was isolated and washed extensively with equal volumes of phosphate-buffered saline (PBS) to remove blood cells. The isolated adipose tissue was then digested with 0.1% collagenase type I (Sigma Diagnostics Inc., St Louis, MO, USA) with intermittent shaking at 37°C for 30 min. The floating adipocytes were separated from the stromal cell fraction by centrifugation (150 × g) for 5 min. The pellets were filtered through a 200 μm nylon mesh to remove cellular debris and incubated overnight at 37°C with 5% CO2 in culture medium (Dulbecco's modified Eagle's medium [DMEM], 10% fetal bovine serum and 100 U/mL penicillin/streptomycin). The primary cells were cultured for 4–5 days until they reached confluence and were defined as passage “0”. The cells were passaged at a ratio of 1:3. The medium was changed every 1 to 2 days.
Identification of ADSC characteristics
The specific cell surface antigen markers of ADSCs were examined with flow cytometry (FCM). The antibodies for positive markers included CD44 (ab25579) and CD90 (ab25672), whereas the negative markers included IgG1 (ab18404, ab37368), CD34 (ab23830) and CD45 (ab25603) (Abcam, HKSTP, N.T. Hong Kong).
A trilineage-induced differentiation experiment was also performed to identify the multiple differentiation potential of ADSCs, which included adipogenesis, osteogenesis and chondrogenesis. Briefly, ADSCs at passage 2 were incubated in a six-well plate at a density of 105 cells per well with SD rat ADSCs adipogenic or osteogenic differentiation medium (Cyagen Biosciences Inc., Sunnyvale, CA, USA) for adipogenesis or osteogenesis induction, respectively. The cells were examined for adipogenesis by oil red O staining after 2 weeks of culture, or for osteogenesis by alkaline phosphatase (ALP) and alizarin red staining after 2 weeks of culture. For chondrogenesis, pellet culture was performed. Briefly, ADSCs were digested with trypsin, and resuspended in DMEM in a 15 mL polypropylene centrifuge tube. A total of 3 × 105 cells/tube were washed with DMEM twice at 150 g for 5 min, resuspended in 0.5 mL of SD rat ADSCs chondrogenic differentiation medium (Cyagen Biosciences Inc.), and centrifuged at 150 g for 5 min. The pellet was incubated at the bottom of the tube with the supernatant at 37°C in 5% CO2 for 24 h, and the tube was gently flicked to ensure the pellet was free-floating. The medium was changed every 2–3 days. After 3 weeks of incubation, the pellet was fixed in 4% paraformaldehyde and embedded in paraffin. Toluidine blue staining was then performed to assess the glycosaminoglycan formation in the ECM of the pellet.
Western blot analysis
Western blot analysis was performed as previously described (Liu et al. 2009). Antibodies against acetyl-histone H3K9, trimethyl-histone H3K9, trimethyl-histone H3K4 and histone H3 were purchased from Upstate Biotechnology (Lake Placid, NY, USA). Bound primary antibodies were detected with secondary anibodies conjugated with horseradish peroxidase (HRP) (Upstate) and visualized by enhanced chemiluminescence (ECL).
Flow cytometry analysis for cell apoptosis
Cells were cultured in a six-well plate at 1 × 106 cells per well with complete DMEM in the presence or absence of a HDAC inhibitor NaBu (Sigma) that was dissolved in PBS, at indicated concentrations. After 72 h of culture, cells were harvested and washed three times in PBS. Cells from each sample were processed for Annexin V fluorescein isothiocyanate (FITC)/ propidium iodide (PI) apoptosis detection (Becton Dickinson, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. All experiments were carried out in triplicate and repeated three times.
Cells were cultured in a 96-well plate at 1 × 103 cells per well with 100 μL complete DMEM in presence or absence of a HDAC inhibitor NaBu for 72 h. Cytotoxicity was measured by performing WST-8 assay using a Cell Counting Kit-8 (CCK-8) (Dojindo Laboratories, Kamimashiki Gun, Kumamoto, Japan), according to the manufacturer's instructions. The 450 nm absorbance was measured with a microplate reader (Bio-Rad, La Jolla, CA, USA). All experiments were carried out in triplicate and three independent experiments were performed.
Cells were seeded into a six-well plate at 1 × 105 cells per well and harvested at assigned time points post-induction. Total RNA extraction was performed using Trizol Reagent (Invitrogen, Carlsbad, CA, USA) based on the manufacturer's instructions. Amplifications were performed with the ABI 7300 Real-Time PCR System (ABI, Carlsbad, CA, USA) with different primers. The primers used were as follows: Runx2 forward TCCAGACCAGCAGCACTCC and reverse TCAGCGTCAACACCATCATTC; osteopontin (Opn) forward AATGAAGGGCCCTGAGC and reverse GCCAGTTCTGCAAGGAAGC; osteocalcin (Ocn) forward AACGGTGGTGCCATAGATGC and reverse AGGACCCTCTCTCTGCTCAC; ALP forward CCAGCAGGCTTACCAAGAA and reverse TTTATCGCACAAAGGGAACA; 18S rRNA forward GTAACCCGTTGAACCCCATT and reverse CCATCCAATCGGTAGTAGCG. All annealing temperatures were set at 60°C. Transcription levels were normalized to 18S rRNA level. Each value represents the average of at least three independent experiments.
Chromatin immunoprecipitation assay
Formaldehyde cross-linked chromatin preparation and chromatin immunoprecipitation (ChIP) were carried out using the ChIP assay kit (Upstate). Briefly, cross-linking of proteins to DNA was performed by adding formaldehyde drop-wise directly to the media for a final concentration of 1% and then was rotated gently at room temperature for 10 min. Cells were washed twice in ice-cold PBS, resuspended in sodium dodecyl sulfate (SDS)-lysis buffer and sonicated until crosslinked chromatin was sheared to an average DNA fragment length of 100–800 bp. Normal rabbit IgG was used as a negative control, and antibodies against acetyl-histone H3K9 and trimethyl-histone H3K9 were used for each immunoprecipitation. The immunoprecipitated DNAs were then analyzed using real-time PCR. For the rat Runx2 promoter region, the forward primer was 5′-GGACCGCCTCCTTCCAACT-3′, the reverse primer was 5′-TCACTCGCCTCCGTCTACC-3′. The annealing temperature was 60°C. The amount of immunoprecipitated DNA was normalized to the input DNA. Each value represents the average of at least three independent experiments.
The results are expressed as means ± standard deviation (SD). Comparisons between groups were analyzed using Student's t-test or analysis of variance (ANOVA), and the Student-Newman–Kleuss method was used to estimate the level of significance. Differences were considered to be statistically significant at P < 0.05.
ADSC primary cultures and characterization
Adipose-derived stem cells isolated from the inguinal groove adipose tissue of 4-week-old rats were initially plated in cell culture dishes (diameter 100 mm). Adipose-derived stem cells at the third passage became more uniform and grew in a monolayer with typical fibroblast morphology (Fig. 1B). All characterizations were performed on the cells at passage 3. Flow cytometry results showed that the ADSCs consisted of a single phenotypic population positive for CD44 (99.51%) and CD90 (99.73%). By contrast, the cells were negative for other markers of the hematopoietic lineage, including the lipopolysaccharide receptor CD34 (3.22%) and the leukocyte common antigen CD45 (0.74%), indicating their non-hematopoietic origin (Fig. 1A).
To further characterize ADSCs, we performed a multilineage differentiation experiment to identify their multiple-differentiation potentials. The results showed that the ADSCs were successfully differentiated into adipocytes (Fig. 1C a), osteocytes (Fig. 1C b) and chondrocytes (Fig. 1C c).
Osteogenic differentiation capacity of ADSCs
The expression of osteogenic genes was assessed by real-time RT-PCR at 0, 3, 7 and 10 days after induction. These genes include Runx2 and ALP (osteogenesis marker genes), Ocn (an early marker of osteogenesis) and Opn (an extracellular matrix protein and a marker of mature osteoblasts). As shown in Fig. 2A-D, the messenger RNA (mRNA) expression of all four osteogenic genes was gradually increased with time.
The effect of NaBu on the survival of ADSCs
Epigenetic modulation plays a critical role in stem cells differentiation (Berdasco & Esteller 2011; Han & Yoon 2012). In order to evaluate whether there was cytotoxicity effect of histone deacetylase inhibitor NaBu, we first examined NaBu's effect on ADSC survival. The cells were treated with NaBu (0–4 mmol/L) for 72 h. Then we performed flow cytometry analysis to quantify the apoptosis rate of ADSCs with Annexin V FITC/PI staining. As shown in Fig. 3A and 3B, NaBu had no effect on the percentage of Annexin V positive ADSCs at concentrations of 0.1 mmol/L, 0.2 mmol/L and 0.5 mmol/L. However, the percentage of Annexin V positive cells increased along with 1 mmol/L and 4 mmol/L NaBu treatment, indicating an enhanced apoptosis at higher doses of NaBu.
It has been reported that NaBu could inhibit proliferation of human mesenchymal stem cells (Lee et al. 2009). In order to determine whether NaBu could affect the proliferation of ADSCs, we performed CCK-8 analysis. As shown in Fig. 3C, although ADSC viability was decreased in a dose-dependent manner, there was no significant effect between the concentration of 0.1 and 1 mmol/L. Considering the apoptosis analysis results together, NaBu had no effect on cell proliferation and cell apoptosis at lower doses. Therefore, 0.1 mmol/L of NaBu was used in the following experiments.
The effect of NaBu on the osteogenic differentiation capacity of ADSCs
To examine the osteogenic effect of NaBu, ADSCs were treated with 0.1 mmol/L NaBu for 3 days and then induced by osteogenic differentiation medium. As shown in Fig. 4A, NaBu could significantly enhance the ALP and alizarin red staining of ADSCs, suggesting that NaBu treatment significantly increased osteogenic differentiation capacity of ADSCs. Furthermore, we found that the expression of Runx2, Opn, Ocn and ALP (Fig. 4B and 4C) increased after NaBu treatment. Histone deacetylase inhibition is an unspecific process that may cause broader responses. In order to determine whether the epigenetic change is specific to Runx2, we also detected the expression of Sox9 (a chondrogenic marker gene) and PPAR (an adipogenic marker gene) at 0, 3, 7 and 10 days after osteogenic induction by real-time RT-PCR. Their expression was not affected by NaBu treatment (data not shown), indicating that the epigenetic change is specific to Runx2 during the process of osteogenic induction. The above results showed that the suppression of HDAC enzymatic activity enhanced osteogenic differentiation concomitantly with the stimulation of osteogenesis-specific genes expression.
In our previous study, we examined the epigenetic characteristics of ADSCs during its osteogenic differentiation process and found that the degree of transcriptionally permissive histone modifications of H3K9 acetylation was increased after osteogenic induction; while the level of transcriptionally repressive histone modification of H3K9 trimethylation was simultaneously decreased (Hu et al. 2013). Runx2 is essential for osteoblast differentiation and bone formation. To determine whether epigenetic changes of histone H3 occurred on the Runx2 promoter during osteogenesis, we performed ChIP assays. As shown in Fig. 4D, the recruitment of acetylated H3K9 on the Runx2 promoter was increased after 3 days osteogenic induction in ADSCs. In contrast, the recruitment of trimethylated H3K9 was decreased. Meanwhile, ChIP assay demonstrated that NaBu significantly increased acetylated histone H3K9 recruitment onto the Runx2 promoter, while it decreased methylated histone H3K9 recruitment. These epigenetic modification changes were in accordance with the expression profile of Runx2 in Fig. 4C. The results indicate that the upregulation of Runx2 during osteogenesis is tightly associated with the selective induction of histone hyperacetylation at its promoter regions.
Stem cell-based regenerative medicine holds great promise for clinical applications in repair of diseased tissues. MSCs can be isolated from most adult tissues. Adipose-derived stem cells and BMSCs are the most thoroughly investigated populations of adult stem cells for clinical applications. Compared with BMSCs, ADSCs can be easily and abundantly available from adipose tissue and may be a promising candidate cell source for tissue engineering. Our previous study indicates the differences in the osteogenic differentiation capacity between BMSCs and ADSCs (Hu et al. 2013). The present study aimed to explore its mechanism and improve the osteogenic potential of ADSCs for further tissue engineering.
Increasing evidence indicates that the multilineage differentiation ability of stem cells is regulated by transcription factors and lineage specification genes (Collas et al. 2008). Runx2 is an osteogenic-specific transcription factor and essential for osteoblast differentiation and bone formation. Overexpression of Runx2 can induce and upregulate expression of multiple osteoblast-specific genes in non-osteogenic cells. Our previous studies have suggested that the Runx2 expression level is highly correlated with the osteogenic differentiation ability of ADSCs.
Recent studies have suggested that the lineage-specific differentiation of adult stem cells could be influenced by epigenetic states that include different modifications of chromatin and such epigenetic regulation could alter the genes' accessibility to transcription factors and other regulators (Gabory et al. 2011). DNA methylation and histone modifications are the major epigenetic modifications, which is tightly linked to mammalian gene regulation and some cellular processes. Our previous results demonstrated that the degree of H3K9 acetylation and H3K4 trimethylation in ADSCs increased, while the level of H3K9 trimethylation decreased. Meanwhile, the epigenetic modifications exhibited a dynamic pattern that is well correlated to the expression profile of Runx2 after osteogenic induction, especially H3K9 acetylation and trimethylation. From Figure 2A we can see the expression of Runx2 was increased in ADSCs in a time-dependent manner. Consistently, the recruitment of acetylated H3K9 on the Runx2 promoter was increased, while the recruitment of trimethylated H3K9 was reduced in differentiating ADSCs. These results indicate that histone modification may play an important role in the osteogenesis process.
Indeed, the manipulation of epigenetic modifications on the genes associated with lineage-specific differentiation will enhance the osteogenic differentiation of ADSCs. NaBu, a potent histone deacetylase inhibitor, can increase gene transcription commonly through promoting the acetylation level of histones and chromatin relaxation, and the degree of this process is dependent on cell type, drug dose and treatment interval.
In this study, the effects of NaBu on osteogenic differentiation of ADSCs are clearly demonstrated. The treatment of NaBu significantly increased osteogenic differentiation of ADSCs. The increased level of acetylated histone H3K9 binding to the Runx2 promoter region further indicates that the upregulation of Runx2 during osteogenesis is tightly associated with the selective induction of histone hyperacetylation at the promoter region of this gene.
The findings in this study indicate that the epigenetic states are responsible for the osteogenic differentiation capacity of ADSCs, and pretreatment with NaBu could enhance the osteogenic potential of ADSCs in vitro, which provides useful information for further clinical application in ADSC-based bone tissue regeneration.
This work was supported by Specialized Research Fund for the Doctoral Program of Higher Education (20110001130001), the National Natural Sciences Foundation of China (90919022, 81101350, 81070112, 81071675), the 111 Project of China (B07001), China Postdoctoral Science Foundation (20110490249).
Xiaoqing Hu, Yingfang Ao and Chunyan Zhou designed research; Xiaoqing Hu, Yutuo Fu, Xin Zhang, Linghui Dai, Jingxian Zhu performed research; Xiaoqing Hu and Yutuo Fu analyzed data; Xiaoqing Hu, Yingfang Ao and Chunyan Zhou wrote the paper.
The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. The authors declare no conflicts of interest.