T-75 flasks and culture dishes (100 × 20 mm) were purchased from SPL Life Sciences (Seoul, Korea). 96-well plates were purchased from BD Biosciences (Franklin Lakes, NJ, USA). Dulbecco's modified Eagle medium (DMEM)-low glucose, fetal bovine serum (FBS), and penicillin/streptomycin were purchased from Life Technologies Korea (Seoul, Korea). Retroviral vector pLXSN expression system was purchased from Clontech (Palo Alto, CA, USA). Annexin V-FITC apoptosis detection kit, Lipofectamine 2000, G418, polybrene reagent, Hoechst 33258, propidium iodide, and 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) were purchased from Sigma (St. Louis, MO, USA). A rabbit polyclonal antibody to human ADC was purchased from ATGEN (Seoul, Korea). Polyvinylidene fluoride (PVDF) membrane was purchased from Millipore (Billerica, MA, USA). 4',6-diamidino-2-phenylindole (DAPI) was purchased from Vector Labs (Burlingame, CA, USA). Bicinchoninic acid (BCA) kit and enhanced chemiluminescent (ECL) Western blotting detection reagent were purchased from Thermo Fischer Scientific (Fremont, CA, USA). Rabbit polyclonal antibodies to AKT and cleaved caspase 3 and rabbit monoclonal antibodies to phosphorylated forms of AKT and CREB were purchased from Cell signaling technology (Beverly, MA, USA). A rabbit polyclonal antibody to BDNF, a mouse polyclonal antibody to β actin, and an anti-mouse or rabbit IgG antibody conjugated with horse radish peroxidasae (HRP) were purchased from Abcam (Cambridge, UK). A rabbit polyclonal antibody to caspase 3 and FITC conjugated second antibody were purchased from Millipore (Billerica, MA, USA). Unless specified, all other chemicals were purchased from Sigma (St. Louis, MO, USA).

Human bone marrow mesenchymal stem cells (MSCs) were obtained from Yonsei Cell Therapy center of Severance Hospital. Flow cytometry was performed to confirm the phenotypic characteristics of MSCs. MSCs were seeded in T-75 flasks and cultured at 37℃ in a humid incubator with 95% of O₂ and 5% of CO₂. The following day, the non-adherent cells were removed from the flasks and the adherent cells were fed with fresh Dulbecco's modified Eagle medium (DMEM)-low glucose supplemented with 10% FBS and 100 unit/mL penicillin and 100 µg/mL streptomycin. MSCs were then continuously maintained for 1-4 weeks. Medium was changed every second day. MSCs, of which passage number is over 5, were substantially used in the experiments. H₂O₂ was employed to reproduce oxidative stress condition, which is prevalent in damaged tissues (21). It has unique biochemical properties such as relatively long half-life and high solubility in both lipid and aqueous media (22). The control MSCs and MSCs infected with human ADC or empty vector were treated with 200 µM of H₂O₂ for 6 hr. This regimen for treatment was selected based on MTT data in the control MSCs exposed to H₂O₂, where the viability was approximately 50%-55%.

The retroviral vector carrying human ADC gene (gene bank accession number AY325129) was previously constructed by our group (23). For the experiments, PT67 packaging cells were infected with retroviral vector containing human ADC gene or empty pLXSN vector and then maintained in a humid incubator with 5% of CO₂ at 37℃ for one week. Then, the medium containing retrovirus was filtered using 0.45 µm pore polysulfonic filter. For infection, human MSCs were cultured with the supernatants containing human ADC vector or empty pLXSN vector using polybrene reagent (6 µg/mL). After 24 hr of incubation, the medium was replaced by normal culture medium. The infected MSCs were maintained to perform the experiments for up to one week.

The expression of human ADC protein was investigated in control MSCs, pLXSN-MSCs and ADC-MSCs through immunocytochemistry. Samples were washed three times with PBS and fixed with 100% ethanol for 10 min on ice, and then washed three times with PBS. To increase permeability, samples were incubated with 0.025% Triton, and were blocked for 1 hr at room temperature with blocking solution. The samples were treated with anti-human ADC antibody for 3 hr at room temperature, which was diluted in the incubation buffer (1:500). Then, primary antibody was removed and samples were washed three times for 5 min each with PBS. Then, samples were treated with FITC labeled secondary antibody for 2 hr at room temperature, which was diluted in the incubation buffer (1:500). Finally, samples were stained with 4',6-diamidino-2-phenylindole (DAPI) 10 µL/mL for staining nucleus and then visualized under a confocal microscope (Zeiss LSM700; Zeiss, Thornwood, NY, USA).

In H₂O₂ treated and non-treated groups, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay, which is based on the activity of mitochondrial dehydrogenase (24), was performed to estimate the number of viable cells. Briefly, MSCs were seeded (1.0 × 10⁴ cells/200 µL) in 96-well plates. The cells were maintained in 96-well plates for 2 days and then were treated with H₂O₂ (200 µM for 6 hr). Then, the media were removed from each well and 200 µL of MTT solution was added into each well and the culture plates were incubated for 4 hr at 37℃. Then, dimethyl sulfoxide (DMSO) was added to dissolve formazan crystal in the mitochondria of living cell. The absorbance at 570 nm was measured with a spectrophotometer.

Hoechst 33258 and propidium iodide were employed to stain control MSCs, LXSN-MSCs and ADC-MSCs in this study. Hoechst 33258 enters both live and dead cells, but is distinctly accumulated in apoptotic nucleus. On the other hand, PI enters only dead cells and thus, is used to examine cell membrane integrity. All experimental groups were cultured in 24-well plates at a density of 3 × 10⁴ cells/well for 2 days. After treatment of 200 µM of H₂O₂ for 6 hr, Hoechst 33258 was added to the culture medium (2-3 mg/mL) and then the samples were kept at 37℃ for 30 min. PI solution was then added (2-5 mg/mL) just before observation under the fluorescence microscope. PI/Hoechst-positive and PI/Hoechst-negatives were counted as dead or living cells, respectively.

The proportion of cells undergoing apoptosis was analyzed with Annexin V-FITC Apoptosis Detection Kit according to the manufacturer's protocol. In Brief, cells were harvested and resuspended in binding buffer (10⁶ cells/mL), and then, cells were mixed with 5 µL of annexin V-FITC and 10 µL of propidium iodide (PI), followed by the incubation of cells at room temperature for 15 min in dark. Then, both FITC and PI fluorescence of cells were analyzed with flow cytometer.

The agmatine concentration in control MSCs and MSCs transfected with empty LXSN or human ADC gene carrying vector was measured by HPLC. The harvested cells were homogenized in 10% trichloracetic acid (TCA) and then, centrifuged at 12,000 rpm for 30 min. The supernatant was derivatized with o-phthaladehyde (OPA) and then, injected into the HPLC column with a fluorescence detector. The recovery of agmatine was calculated from the added external standard and the values are expressed as the µM concentration.

Cells from both H₂O₂ treated/non-treated groups were collected and the pellets were resuspended in lysis buffer (1% NP-40, 50 mM Tris [pH 7.5], 150 mM sodium chloride, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM EGTA, 5 mM EDTA, 1 mM PMSF, 1 mM Na orthovanadate, 5 mM Na fluoride, 1 µg/mL pepstatin, 2 µg/mL aprotinin, and 5 µg/mL leupeptin) for 2 hr and then centrifuged at 13,200 rpm for 30 min at 4℃. The supernatants were collected and protein concentration was determined with bicinchoninic acid (BCA) assay. The equal amounts of proteins (40 µg) were separated on 10%-12% SDS-polyacrylamide gels using an electrophoresis unit maintained at the constant voltage (100 V). Separated proteins were then electrotransfered onto PVDF membrane. Later, membranes were incubated with 0.1% bovine serum albumin (BSA) in Tris-buffered saline and 0.3% Tween-20 (TBS-T) to block nonspecific binding for 1 hr at room temperature. After blocking, the membranes were treated with specific primary antibodies including anti-phospho AKT, anti-phospho CREB, anti-caspase 3, anti-cleavage caspase 3, and anti-BDNF at 4℃. All of primary antibodies were diluted (1:1,000) in incubation buffers for treatment. After treatment, membranes were washed 3 times with TBST for every 5 min. Then, membranes were treated for 1 hr at room temperature with anti-mouse or rabbit IgG antibody conjugated with horse radish peroxidase (HRP) in the incubation buffers (1:3,000). Finally, blots were rinsed and proteins of interest were detected with enhanced chemiluminescent (ECL) kit and visualized with LAS 4000 image analyzer. Protein bands detected were quantified as the ratio of the target protein to beta actin using Image J software (National Institutes of Health, Bethesda, MD, USA).

The data are presented as mean ± SD. Statistical analysis of the data was made by one-way ANOVA using SPSS statistical package version 12.0. P values less than 0.05 were considered to be statistically significant and the significant differences were calculated at 95% confidence interval. All experiments were independently carried out at least three times.

The retroviral vectors carrying human arginine decarboxylase (ADC), which was previously constructed by our group (23), were employed to infect human mesenchymal stem cells (MSCs). Western blotting data show that ADC protein was greatly elevated in MSCs infected with vectors containing ADC gene (Fig. 1A). In particular, the increase of approximately 0.7 or 2.5 fold was respectively observed in those MSCs, compared with LXSN-MSCs under normal or H₂O₂ exposure condition (Fig. 1B). And the level of agmatine synthesized by ADC was approximately 2 fold raised in MSCs infected with vectors containing ADC gene under H₂O₂ exposure (Fig. 1C), implying the overexpressed ADC involved this elevation in those MSCs.

The overexpressed human arginine decarboxylase (ADC) preserved mitochondrial integrity in human mesenchymal stem cells (MSCs) treated with H₂O₂. The level of formazan was 56.4% ± 0.9% and 52.8% ± 1.5% of untreated control MSCs in control MSCs or LXSN-MSCs exposed to H₂O₂, respectively (Fig. 2A). However, its level significantly increased into 69.7% ± 2.7% of untreated control MSCs in ADC-MSCs treated with H₂O₂ (Fig. 2A). As well, the overexpressed ADC saved plasma membrane integrity in human MSCs exposed to H₂O₂. The propidium iodide (PI) positive cells clearly appeared in control MSCs and LXSN-MSCs treated with H₂O₂ (Fig. 2B). However, this damage was significantly alleviated in ADC-MSCs. In fact, the numbers of PI positive cells were reduced into 32.9% ± 1.3% of control MSCs in ADC-MSCs under H₂O₂ exposure. This protective effect of the overexpressed ADC was further investigated with FACS analysis. The distribution of FITC positive cells was obviously reduced in ADC-MSCs compared to control and LXSN MSCs under H₂O₂ exposure (Fig. 2C).

The overexpressed human arginine decarboxylase (ADC) suppressed apoptotic events in human mesenchymal stem cells (MSCs) exposed to H₂O₂. The level of pro-caspase 3 was not quite different between control MSCs and ADC-MSCs under normal or H₂O₂ exposure (Fig. 3A). However, the level of cleaved caspase 3 increased in control MSCs treated with H₂O₂, whereas its level obviously decreased in ADC-MSCs exposed to H₂O₂ (Fig. 3A). The similar trend also appeared in nuclear integrity. No difference of nuclear morphology was found between control MSCs and ADC-MSCs under normal condition (Fig. 3B). However, a lot of condensed nuclei were observed in control MSCs exposed to H₂O₂, while their numbers apparently decreased in ADC-MSCs exposed to H₂O₂ (Fig. 3B). This protective effect of the overexpressed human arginine decarboxylase (ADC) was further substantiated in human mesenchymal stem cells (MSCs) treated with H₂O₂. That is, the expressions of pro-survival proteins P-AKT, P-CREB, and BDNF obviously increased in ADC-MSCs, compared with control MSCs under H₂O₂ exposure (Fig. 4). However, their expressions were not differed between control MSCs and ADC-MSCs under normal condition (Fig. 4).