This study was approved by the Institutional Review Board of Yonsei University Wonju College of Medicine (CR 319009). BMSCs at P1 or P2 from six cirrhosis patients (Korean, mean age 45.8±8.8 years) were obtained with informed consent from Pharmicell Co., Ltd. (Sungnam, Korea). Bone marrow mononuclear cells from six healthy volunteers (Caucasian, mean age 43.2±8.7 years) were purchased from the American Type Culture Collection (Manassas, VA, USA) and plated in 100 mm dishes (2×10⁵ cells/cm²) with low-glucose Dulbecco’s modified Eagle’s medium (LG-DMEM, Gibco BRL, Rockville, MD, USA) containing 10% fetal bovine serum (FBS, Gibco BRL) and penicillin/streptomycin. After 2 days, the medium was changed to remove non-adherent cells, and then the cell culture medium was changed twice weekly. When the cells reached 90% confluence (P0), the BMSCs were trypsinized and passaged at a density of 5×10³ cells/cm². Population doubling time (PDT) was determined by dividing the total number of hours in culture by the number of doublings. To calculate the cumulative cell numbers, BMSCs were serial passaged until the PDT of each BMSC reached 150 h. BC or BP represents BMSCs from healthy volunteer or patient. The number after BC or BP represents each individual in the volunteers’ or the patients’ group.

All animal experimental protocols and procedures were approved by the Institutional Animal Care and Use Committee of Yonsei University Wonju College of Medicine (YWC-180724-1). Male Sprague-Dawley (SD) rats (7 weeks old) were purchased from Orient Bio Inc. (Seongnam, Korea). Cirrhosis was induced by intraperitoneal injection of thioacetamide (TAA, Sigma-Aldrich, St. Louis, MO, USA; 200 mg/kg body weight) twice a week for 12 weeks. After administration of TAA for 12 weeks, rats were sacrificed with isoflurane anesthesia (Ifran, Hana Pham, Hwaseong, Korea). Femora were aseptically removed and washed 3 times with PBS. Thereafter, the bone marrows were flushed out using LG-DMEM onto 24-well plates. The culture medium was changed twice weekly for two weeks. For direct visualization of the colonies, the cells were washed with PBS and fixed in 95% ethanol for 5 min, and then the cells were incubated for 30 min at room temperature in 0.5% crystal violet in 95% ethanol. Excess stain was removed by washing with distilled H₂O. The plates were dried and the number of CFU-F was counted. We defined a CFU-F unit as consisting of more than 100 cells using a microscope.

A total of 5×10⁵ BMSCs were stained with antibodies conjugated with phycoerythrin (PE) against CD73, CD90, and CD105 (BD Biosciences, San Jose, CA, USA) for 20 min at room temperature. PE-conjugated mouse IgGs were used as the control isotype. The fluorescence intensity of the cells was evaluated by flow cytometry (FACS Aria III; BD Biosciences).

Total RNA was extracted using TRIzol Reagent (Gibco BRL). The reverse transcription reaction was conducted using RT RreMix Kit (iNtRON Biotechnology, Sungnam, Korea) to detect stemness genes (Nanog, OCT4, and Sox2), differentiation genes [leptin, peroxisome proliferator-acti-vated receptor γ (PPARγ), and CCAAT-enhancer-binding protein α (C/EBPα)], and parameters of mitochondrial activities [cytochrome c oxidase subunit I (MT-CO1), transcription factor A, mitochondrial (Tfam), succinate dehydrogenase complex iron sulfur subunit B (SDHB), catalase (Cat), glutathione peroxidase 1 (GPX1), and manganese superoxide dismutase (MnSOD)] using gene-specific primers (Table 1). The reagents in 10 μl reaction mixture included cDNA, primer pairs, and the SYBR Green PCR Master Mix (Applied Biosystems, Dublin, Ireland). All qPCR reactions were performed in duplicate. Peptidylprolyl Isomerase A was used for normalization. The 2^−(ΔΔCt) method was used to calculate the relative fold change of mRNA expression.

BMSCs in P3 were stained for β-gal activity as described by Dimri et al. (21). Briefly, 4×10⁴ cells were seeded in 12-well plates and cultured for 2 days. The β-gal activity was assessed with a senescence β-gal staining kit (Cell Signaling Technology) according to the manufacturer’s instructions. The percentage of senescent cells was represented by the number of stained cells in the total population.

Proteins were extracted, separated by 10% Tris-glycine on SDS-PAGE, transferred to an Immobilon membrane (Millipore), and then incubated with primary antibodies against p53, p21, and GAPDH (1：1000, Santa Cruz Biotech, Santa Cruz, CA, USA), followed by incubation with peroxidase-conjugated secondary antibody (1：2000, Santa Cruz Biotech). The membrane was then treated with EZ-Western Lumi Pico (DOGEN, Seoul, Korea) and visualized using ChemiDoc XRS+ system (Bio-Rad, Hercules, CA, USA).

BMSCs (2×10⁴ cells/cm²) were seeded in 6-well plates and cultured for 1 week. The medium was then changed to an adipogenic medium [10% FBS, 1 μM dexame-thasone, 0.5 mM 3-isobutyl-1-methylxanthine, 10 μg/ml insulin, and 100 μM indomethacin in high glucose (HG)-DMEM] for an additional 3 weeks. Cells were fixed in 4% paraformaldehyde for 10 min, stained with fresh Oil Red O solution to stain the lipid droplets, and photographed. Oil Red O was then eluted with isopropanol, and the extracted Oil Red O was quantitated by measuring the absorbance at 540 nm.

Total genomic DNA of BMSCs were isolated using QIAamp DNA Mini Kit (Qiagen, Germantown, MD, USA), and mtDNA was analyzed using Absolute Human Mitochondrial DNA Copy Number qPCR kit (Sciencell Research Laboratories, Carlsbad, CA, USA), according to the manufacturer’s instructions.

MMP was measured using the lipophilic cationic dye 5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolylcar-bocyanine iodide (JC-1, Molecular Probes, Thermo Fisher Scientific). BMSCs were seeded and treated for 4 h with carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FC CP) or vehicle control (dimethyl sulfoxide). The fluorescence intensity for both aggregates and monomer of JC-1 was measured with a fluorescence microplate reader (Flexstation II, Molecular Devices, San Jose, CA, USA; JC-1 aggregates: excitation/emission=540/590; JC-1 monomer: excitation/emission=490/535).

BMSCs were lysed with the cell lysis reagent supplied in ATP Bioluminescence assay kit HS II (Roche Diag-nostic GmbH), and the lysates were centrifuged to recover the supernatant. After the reaction, ATP levels were measured with a microplate luminometer (Synergy 2, Bio-Tek Ins-trument, Winooski, VT, USA) and normalized to the protein concentration.

BMSCs were plated on Seahorse 96 well plate (Agilent Technologies, Cedar Creek, TX, USA). After 24 h, cells were changed to XF DMEM medium containing 1 M glucose, 100 mM pyruvate, and 200 mM L-glutamate (pH 7.4, Agilent Technologies) and were maintained at 37℃ without CO₂ for 1 h. Oxygen consumption rate (OCR) was evaluated by Seahorse XF kit and Seahorse XFe96 Analyzer (Agilent Technologies), and normalized by protein concentration. The cycles (three times for 3 min) were run for every measurement, and the Mitostress kit (Agilent Technologies) was used that contained 2 μM oligomycin (ATP synthase inhibitor), 2 μM FCCP (mitochondrial uncoupler), 0.5 μM rotenone (respiratory chain complex I inhibitor), and antimycin A (complex III inhibitor).

ROS generation was evaluated using 5-(-6)-chloro-methyl-2’,7’-dichlorodihydrofluorescein diacetate, acetyl ester (CM-H₂DCFDA; Invitrogen, Eugene, OR, USA). After CM-H₂DCFDA incubation, fluorescence was measured at the excitation and emission wavelengths of 485 and 538 nm, respectively, by using a fluorescence microplate reader (Flexstation II, Molecular devices).

Data are presented as the mean±standard error of the mean. To compare group means, Student’s t-test and one-way analysis of variance were used, followed by the Scheffe’s test. Any difference was considered statistically significant at *p<0.05, **p<0.005, and ***p<0.0001.

Representative positive cell surface antigens of MSCs are CD73, CD90, and CD105 (22), and approximately 99% of positive cells in both BCs and BPs at passage 3 (P3) expressed all these antigen (Fig. 1A). Furthermore, there was no statistically significant difference in mean fluorescence intensities (MFI) of CD73, CD90, and CD105 in BCs and BPs, suggesting comparable expression of cell surface antigens on these BMSCs (Fig. 1B).

To compare the proliferation potential of BCs and BPs, we analyzed the average PDT at P3, P4, and P5, the early passages, and calculated the total cumulative cell number obtained after serial passage until the PDT of each BMSC reached 150 h. Besides, we analyzed the mean expression levels of Nanog, Oct4, and Sox2, stemness regulators in P3, P4, and P5. The mean PDT of BPs (65.3±11.1 h) at the early passage was significantly shorter than that of BCs (83.7±9.2 h), but the average total cumulative number of cells was higher in BCs (5.97×10⁷ cells) than in BPs (4.90±10⁷ cells); however, the difference was not significant (Fig. 2A∼C). Furthermore, no differences in expressions of stemness regulators between BCs and BPs were observed in the early passage (Fig. 2D∼F). To evaluate whether the lower PDT in BPs is a common phenomenon in cirrhosis disease, we investigated the colony forming unit–fibroblast (CFU-F) in normal and cirrhotic SD rat. Interestingly, much more colony formation was observed in the bone marrow of thioacetamide-induced cirrhotic rats than normal rats, and colony size was larger in cirrhotic rats than in normal rats (Fig. 2G). These results suggest that pathologic conditions with fibrosis and inflammation, such as cirrhotic changes, may promote the proliferation of stem cells.

At P3, senescent cells formed 11.22±7.03% and 19.06±3.55% in BCs and BPs, respectively (Fig. 3A). Consistently, p21 and p53 expression was significantly higher in BPs than in BCs (Fig. 3B). Despite the shorter PDT in the early passage, there were 1.7 times more aging cells in the BP group.

As the passage of MSCs progressed, the differentiation ability of adipocytes decreases gradually (23). We compared the differentiation capability of BCs and BPs in the early passage into adipocytes. More lipid droplets following differentiation were microscopically observed in BCs than in BPs (Fig. 4A). On measuring the degree of differentiation by the absorbance of Oil Red O extracted with isopropanol, the absorbance of BCs was significantly higher than that of BPs. When leptin, PPARγ, and C/EBPα expressed in adipocytes were identified by qPCR, the expression level of adipocyte-related genes in both groups was not significant (Fig. 4B).

Mitochondria is bioenergetically important for ATP production (24) and also generates ROS and activates apoptosis (25, 26). To compare the mitochondrial activities of BCs and BPs, we analyzed mtDNA copy number, MMP, OCR, and ROS generation. There was no significant difference between mtDNA copy numbers in BCs and BPs, and in fact increased variations were observed in individual BPs (Fig. 5A). The JC-1 fluorescence ratio of J-aggregates and monomer reflects the MMP, which was slightly lower in BPs (1.99±0.28) than in BCs (2.24±0.18) but not significant (Fig. 5B). FCCP-induced depolarization was similar between BCs (0.69±0.10) and BPs (0.68±0.11). ATP production was significantly higher in BCs (43.72±5.95 nmol/mg protein) than in BPs (30.70±8.45 nmol/mg protein) (Fig. 5C). The expression levels of mitochondrial proteins including MT-CO1, Tfam, and SDHB were not different between BCs and BPs (Fig. 5D∼F). As a sensitive indicator of mitochondrial activity, we measured mitochondrial OCRs in the basal state and maximal stimulation. Mitochondrial respiratory activities were not significantly different between BCs and BPs (Fig. 5G∼I). Additionally, although not significant, the expression of antioxidant genes, CAT, GPX1, and MnSOD, were slightly higher in BPs than in BCs (Fig. 5J∼L). Unlike the expression of antioxidants, ROS generation in the basal state and after H₂O₂ treatment was lower in BPs than in BCs, implying less oxidative stress in BPs (Fig. 5M).