Specimens and cell cultures
Nine bone specimens (ilium: three cases, maxilla: three cases, mandible: three cases; Supplemental Table S1), harvested during conventional surgery conducted by the Department of Plastic and Aesthetic Surgery at Kitasato University, were used in the present study. Table S1 shows patient profiles, including sex, age, and type of bone. Small pieces of bone were prepared and subjected to primary culture in complete medium containing 10% fetal bovine serum (MP Biomedicals LLC, Santa Ana, CA) at 37°C with 95% humidity and 5% CO2. The complete medium consisted of alpha-minimum essential medium (Thermo Fisher Scientific, Waltham, MA), 100 U/mL penicillin and 100 µg/mL streptomycin (Thermo Fisher Scientific), and 1 ng/mL basic fibroblast growth factor (R&D Systems, Minneapolis, MN). The medium was changed twice a week. At confluency, MSCs were separated and harvested using 0.05% trypsin-EDTA (Thermo Fisher Scientific). Second and third passage MSCs were used in this study.
Differentiation for osteogenesis and adipogenesis
MSCs were cultured to confluence in complete medium, then seeded in 6-well plates at a density of 1 × 105 cells/well. Osteogenic differentiation was induced by culturing in osteogenic differentiation medium (ODM), consisting of complete medium supplemented with 100 nM dexamethasone (Sigma-Aldrich, St Louis, MO), 0.05 mM ascorbic acid (Wako Pure Chemical Industries, Osaka, Japan), and 10 mM β-glycerophosphate (Calbiochem, San Diego, CA). Cells were grown at 37°C with 5% CO2. The medium was replaced twice a week.
Adipogenic differentiation occurred in adipogenic differentiation medium (ADM), consisting of complete medium supplemented with 1 μM dexamethasone, 0.01 mg/ml insulin (Wako Pure Chemical Industries), 0.2 mM indomethacin (Wako Pure Chemical Industries), and 0.5 mM isobutylmethylxanthine (Sigma-Aldrich). The medium was replaced twice a week.
Chondrogenic differentiation was performed according to the pellet culture method [51]. Briefly, 2 × 105 cells were transferred to a 15-mL polypropylene tube and centrifuged at 1,000 rpm for 5 minutes at 20℃ to obtain a pellet. After at least 24 hours, the cells were subjected to differentiation induction using chondrogenic differentiation medium (CDM), consisting of High-glucose DMEM (Gibco) supplemented with 2mM L-glutamine (MP Biomedicals), 10 ng/mL human recombinant TGF-β1 (Sigma-Aldrich), 100 nM dexamethasone, 50 μg/mL ascorbic acid-2-phosphate, 100 μg/mL sodium pyruvate, and 50 mg/mL ITS+Premix (Corning). The medium was replaced twice a week.
Examination and evaluation of MSC characteristics
For osteogenic differentiation, MSCs were cultured in ODM at 37°C and 5% CO2 for 6 weeks. The ODM was replaced twice a week. For adipogenic differentiation, MSCs were cultured in ODM for 3 weeks. After culturing for 3 weeks, the medium was changed to ADM, and MSCs were cultured for an additional 3 weeks. As a control, MSCs were continuously cultured in ODM for 6 weeks.
For the calcium (Ca) production assay, MSCs were seeded at 1 × 105 cells/well in 6-well plates and cultivated in ODM or complete medium (negative control) at 37°C under 5% CO2. ODM and complete medium were replaced twice a week. Samples were collected at 4 and 6 weeks after osteogenic differentiation. ESPA Ca (Nipro, Osaka, Japan) was used to assay Ca levels.
For Ca staining with the alizarin red S staining assay, MSCs were seeded at 1 × 105 cells/well in 6-well plates and cultured in ODM or complete medium at 37°C under 5% CO2. ODM and complete medium were replaced twice a week. Alizarin red S staining was conducted 6 weeks after osteogenic differentiation by staining MSCs for 2 min with 1.3% alizarin red S solution at room temperature, washing twice with phosphate-buffered saline (PBS), fixing with 100% ethanol, and washing twice with distilled water. The excess staining solution was removed by washing three times with distilled water, then the cells were allowed to dry.
For lipid staining with oil red O solution, MSCs were cultured in ODM at 37°C under 5% CO2. ODM was replaced twice a week. After culturing for 3 weeks, the medium was changed to ADM, and MSCs were cultured for 3 weeks further. As a control, MSCs were continuously cultured for 6 weeks in ODM. Oil red O staining was performed after adipogenic differentiation by washing cells twice with PBS, fixing with 10% formalin, washing once with distilled water then once with 60% isopropanol, and staining for 20 min in oil red O solution. Cells were then washed once each with 60% isopropanol and distilled water.
The chondrogenic differentiation potency was evaluated after 3 weeks culture in CDM. The chondrocyte mass was washed twice with PBS and fixed with 4% PFA overnight at 4°C. After paraffin embedding, the specimen block was cut into 5µm sections and stained with toluidine blue (Wako Pure Chemical Industries).
Isolation of total RNA and cDNA library preparation for RNA-sequencing (RNA-seq)
For RNA-seq, ilium-derived MSCs (I-MSCs), maxilla-derived MSCs (Mx-MSCs), and mandible-derived MSCs (Md-MSCs) were seeded in 100 mm dishes at a density of 5 × 105 cells/dish and cultured in complete medium for 3 days. Total RNA was isolated using the RiboPure RNA Purification Kit (Thermo Fisher Scientific). The RNA integrity number score of every sample used for RNA-seq was >9. Strand-specific RNA libraries were constructed using the Dynabeads mRNA DIRECT Micro Purification Kit and the Ion Total RNA-Seq Kit v2.0 (Thermo Fisher Scientific). RNA-seq templates were prepared with the Ion PI Hi-Q OT2 200 Kit using the Ion OneTouch 2 and Ion OneTouch ES systems (Thermo Fisher Scientific). These templates were loaded onto Ion PI Chips and sequenced on the Ion Proton Sequencer using the Ion PI Hi-Q Sequencing 200 Kit (Thermo Fisher Scientific).
Statistical and bioinformatics analysis
Adaptor sequences on raw reads for each sample were removed by Cutadapt (v. 1.10), and low-quality bases were trimmed by Trimmomatic (v. 0.35). Sequencing reads were aligned to the human reference genome (hg19) using Bowtie2 (v. 2.2.6)/Tophat2 (v. 2.1.0) software [52]. Uniquely mapped reads were normalized to gene expression as fragments per kilobase of exon per million mapped fragments (FPKM) using Cufflinks (v. 2.2.1) packages (Tophat2-Cufflinks pipeline [53]). To detect differentially expressed genes (DEGs) between two samples, we used Cuffdiff that generates a fold change for every gene, a p-value, and the false discovery rate with the Benjamini–Hochberg correction (q-value) in the Cufflinks packages. DEGs were defined when the q-value <0.05 and the fold change of FPKM was ≥2.0. The human reference genome and the Refseq annotation were downloaded from the UCSC Genome Browser (https://genome.ucsc.edu/). Enriched Gene Ontology (GO) terms were analyzed using the Database for Annotation, Visualization and Integrated Discovery (DAVID; https://david.ncifcrf.gov/) with the annotation dataset GO biological process. Specific GO terms were obtained from the NCBI database (ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/). Most of the data were corrected and calculated by custom Perl scripts. Plots and graphs obtained from RNA-seq analysis were visualized using ggplot2 and other R packages. Clustering analysis was conducted by the function hclust of R. RNA-seq datasets generated in this study have been deposited in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive under accession numbers DRA006607–006615. In the statistical analyses for Ca content assay, all of the values are reported as the mean ± standard deviation (SD). Tukey's HSD test was performed for comparing differences between multiple groups. Differences were considered significant at p < 0.05.