Preparation of caffeine solutions
The powder of caffeine (Sigma, St. Louis, MO, USA) were purchased and diluted in phosphate buffered solution (Sigma, St. Louis, MO, USA). In the first part of this study, the effects of various concentrations of caffeine on bone cell activities were evaluated by using MTT assay as described below. Seven different concentrations (100, 50, 10, 5, 1, 0.5, 0.1 mM) were tested for 1 day, 3 days, 7 days and 14 days period.
Osteoblast cell culture
Sequential digestion of newborn Wistar-rat calvaria was performed by using modification of the methods previously described . To subculture, the cells were washed with sterile PBS followed by treatment with 1:1 mixture of 0.03% collagenase and 0.05% trypsin (Sigma, St. Louis, MO, USA) for 20 minutes at 37°C in 5% C02. The resulting cell suspension was then passed and centrifuged at 1500 rpm for five minutes to pellet the cells. The supernatant was removed and the pellet re-suspended in α-minimal essential media (α-MEM; Sigma, St. Louis, MO, USA) as described below. Unambiguous identification of cell populations as osteoblasts is complex and none of the parameters used for defining osteoblasts-like cells are unique to this cell types. The presence of alkaline phosphatase, an early marker of osteoblasts , is used to assess the osteoblastic character of the isolated cells .
Colorimetric assay for cell viability 
The mitochondria activity of the bone cells after exposure to various concentrations of caffeine was determined by colorimetric assay which detects the conversion of 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (MTT, Sigma Co., St. Louis, MO, USA) to formazan. For the assay, 2.5 × 104 cells per well were incubated (5% CO2, 37°C) in the presence of various concentration of caffeine. After various time intervals the supernatant was removed, 100 μl per well of MTT solution (1 mg/ml in test medium) was added and the wells were incubated at 37°C for 4 h to allow the formation of formazan crystal. All crystals were dissolved, the plates were read on Micro Elisa reader (Emax Science Corp., Sunnyvale, California, USA) at wavelength of 570 nm against a reference wavelength of 690 nm.
Osteoblasts cultured in the media in the presence of dexamethasone have been shown to be capable of synthesizing and mineralizing an extracellular matrix and to form alkaline phosphatase in vitro . To test the differentiation of osteoblasts, a concentration of 1 × 105 cells/100 μl was added to 35 mm wells of a 6-well plate. The osteoblasts were incubated at 37°C in 5% C02 for 48 hours. After 48 hours, the media were changed and the cells were incubated in α-MEM supplemented with 10% fetal calf serum (FCS; Gibco BRL, Rockville, MD, USA), antibiotics (gentamicin 50 μg/ml, penicillin G 100 μg/ml [Gibco BRL, Rockville, MD, USA]), L-ascorbic acid (50 μg/ml Gibco BRL, Rockville, MD, USA), supplemented with 5 mM β-glycerophosphate (Sigma, St. Louis, MO, USA) and 10-8 M dexamethasone (Sigma, St. Louis, MO, USA). The day of changing specific medium was day zero. From day zero of culture, 10 mM caffeine solution was added. The medium was changed every 3–4 days; alkaline phosphatase (ALP) staining, von Kossa stain for mineralized nodules and biochemical parameters including alkaline phosphatase, lactate dehydrogenase, prostaglandin E2 and total protein were performed at day 1, 3, and 7.
Alkaline Phosphatase (ALP) staining
After fixing the cells, the dishes were incubated for 30 minutes in TRIS Buffer (0.2 M, pH 8.3) with AS-MX phosphate (Sigma, St. Louis, MO, USA) as a substrate and Fast Blue (Sigma, St. Louis, MO, USA) as a stain. The ALP positive cells stained blue/purple. For each experiment, a minimum of three dishes was counted and the experiments were repeated three times.
The von-Kossa staining on mineralized nodules formation
Mineralization of the nodules in the cultures was assessed using von-Kossa stain. The matrix was washed with PBS, and cultures were treated with 5% silver nitrate solution 100 μL/well in the dark at 37°C for 30 minutes. The excess silver nitrate solution was then completely washed away using double-distilled H2O and the culture plate was exposed to sodium carbonate/formaldehyde solution for few minutes to develop color. The von Kossa-stained areas were viewed by light microscopy. For each experiment, a minimum of three dishes was counted and the experiments were repeated three times.
Analysis of alkaline phosphatase, lactate dehydrogenase, prostaglandin E2 and total protein in culture medium
Alkaline phosphatase (ALP), lactate dehydrogenase (LDH) activities and total protein released from the cells into the medium were measured with a commercially available assay kit (ALP: Procedure no. ALP-10; Procedure no. 435, LDH: Procedure no. 228-UV, LDL-10, TP: Procedure no.690-A, Sigma Co., St. Louis, MO, USA). The production of prostaglandin E2 (PGE2) in culture medium was also analyzed with a commercially available assay kit (Cayman Chemical Company, MI, USA).
Analysis of intracellular ALP, LDH, PGE2 and total protein
At the end of the experimental period, ALP, LDH, PGE2 and TP activities were determined following lysis of the cells with the detergent Triton X-100 (Sigma, St Louis, MO, USA). Intracellular ALP, LDH, PGE2 and TP values were determined as the methods described for the measurements of culture media.
All data were expressed as mean ± standard deviation and were analyzed by analysis of variance. Statistical significance was determined by Bonferroni's t-test. Probability values less than 0.05 were considered significant.
DNA degradation analysis
For the DNA fragmentation, a concentration of 1 × 106 cells/100 μl was added to 90 mm disc. From day one of culture, six different concentrations of caffeine solution (0, 0.5, 1.0, 2.5, 5.0, 10.0 nM) were added. The medium was changed every 3–4 days, the DNA fragmentation analyses were performed at day 1, 3, and 7. For the test, floating and adherent cells from each culture condition were combined, centrifuged, pelleted at 400 × g for 5 min, and washed twice with PBS. The pellet was resuspended in 0.2 ml lysis buffer [100 mM NaCl, 10 mM Tris (pH 8.0), 1 mM EDTA, 0.5% sodium dodecyl sulfate, 0.20 mg/ml proteinase K, 200 μg/ml ribonuclease A]. The cell lysates were then incubated at 37°C for 2 h. The genomic DNA was extracted by two separations, with phenol/chloroform and then with chloroform only. The DNA pellet was then washed in 70% ethanol and resuspended in 1 mM EDTA, 10 mM Tris-HCl (pH 8.0) at a final concentration of 20 μg/ml. The DNA fragmentation analysis was performed using a 1.5% agarose gel in 1 mM EDTA, 40 mM Tris acetate (pH 7.6) to visualize the laddering of the samples.