2.1.

Cell Culture

Saos-2 human osteoblast-like cells (ATCC® HTB-85™) were cultured in Dulbecco modified Eagle’s medium (DMEM; Sigma-Aldrich, St. Louis, Missouri) enriched with 10% fetal calf serum (Lonza Walkersville Inc., Walkersville), $100\mathrm{IU}/\mathrm{ml}$ penicillin, and $100\mu \mathrm{g}/\mathrm{ml}$ streptomycin solution (Gibco Invitrogen SRL, San Giuliano Milanese, Milan). When confluent, cells were detached with 0.05% (w/v) trypsin and 0.02% (w/v) ethylenediamine tetra-acetic acid (EDTA), counted and seeded into black 24-multiwell tissue culture plates with clear bottoms (STEPBIO S.r.l., Bologna, Italy) at a density of $2.6\times {10}^4\mathrm{cell}/{\mathrm{cm}}^2$. Plates were returned to the controlled humidified incubator (37°C in temperature, 95% air/5% ${\mathrm{CO}}_2$).

2.2.

In Vitro Micro Wound Model

After confluence had been reached, Saos-2 cultures were wounded with a sterile $200\text{-}\mu \mathrm{m}$ Eppendorf tip to create a cell free zone in the monolayer (at baseline, T0, wound area measured $7.2\pm 0.4{\mathrm{mm}}^2$). Cells were extensively washed with sterile phosphate buffered solution (PBS, Gibco Invitrogen SRL, San Giuliano Milanese, Milan), then laser irradiated. To discriminate the contribution of cell proliferation and migration to the process of wound closure, half of the wells were treated with MMC (Sigma-Aldrich) at a concentration of $50\mathrm{ng}/\mathrm{ml}$. Cell cultures were incubated and observed with an inverted microscope (Nikon Eclipse Ti-U, Nikon Italia, Italy) equipped with a digital camera (Sight DS-Fi2, Nikon Italia, Italy) after 4, 24, 48, 72, and 96 h from laser irradiation. Each well was photographed at $4\times$ magnification to cover the wounded area. The image acquisition software (NiS Elements Advanced Research, Nikon Italia, Italy) was used to measure the area of the cell free zone of the artificially created wounds.

2.3.

Laser Irradiation

Cells were exposed to irradiation with a GaAlAs diode laser (Pocket Laser, Orotig s.r.l., Verona, Italy), which has a wavelength of $915\pm 10\mathrm{nm}$ and a maximum power output of $6\mathrm{W}\pm 20\%$. A 100 Hz pulse irradiation mode, a duty cycle of 50%, and a set power of 1 W (corresponding to an output power of 0.575 W, as measured at hand piece aperture) were used for 48, 96, and 144 s. The administered doses were, respectively, of 5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$. The laser beam was delivered perpendicularly to each well by an optical fiber 0.6 mm in diameter that was defocused at the tip by a concave lens to cover the growth area of each well ($1.91{\mathrm{cm}}^2$) at a distance of 19 mm. To avoid overlapping or scattered irradiation, black multiwell plates were used for all the assays. During the period of laser irradiation, the cover plate was removed and DMEM was replaced with PBS to avoid serum interference during irradiation.² Control cells were not irradiated. Both control and laser-treated cells were cultivated under the same experimental conditions.

2.4.

Viability Test and DNA Quantification

Alamar Blue assay (AbD Serotec, Oxford, United Kingdom) was used to evaluate cell viability after 24, 48, and 72 h from laser irradiation. Alamar was added to each culture well (1:10 v/v) for 4 h at 37°C. The colorimetric reaction was measured spectrophotometrically on supernatants at 570 and 625 nm wavelengths with a microplate absorbance reader (iMark, Biorad-Laboratoires Inc., Hercules, USA).

DNA quantification (Quant-iT™ PicoGreen® dsDNA) was performed following manufacturer’s instructions. Briefly, cells were repeatedly washed with PBS, frozen at $-80°\mathrm{C}$, and thawed at room temperature three times. Cell lysis was obtained by adding $100\mu \mathrm{l}$ of Tris-EDTA buffer with sodium dodecyl sulphate 0.01% solution. A working solution of the PicoGreen® reagent was added and incubated with cell lysates in the dark for 3 min at room temperature. The fluorescence was read at 490ex-520em wavelengths, the readings were expressed as relative fluorescence units, and the DNA amount of each sample was calculated above a standard curve.

2.5.

Quantification of mRNA Expression Levels by Quantitative Polymerase Chain Reaction

After 24, 48, and 72 h from laser irradiation, Saos-2 cells grown in the presence of DMEM (10% fetal bovine serum, 1% penicillin-streptomycin and plasmocin) were homogenized and total RNA extraction was performed by the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. Total RNA was eluted with RNase-free water, quantified by NanoDrop 2000 (Thermo Scientific, Waltham, Massachusetts), and kept at $-80°\mathrm{C}$ until reverse transcription.

Each RNA sample (2500 ng) was reverse transcribed to cDNA using the Super Script VILO cDNA Synthesis kit (Invitrogen) according to manufacturer’s instructions and diluted to the final concentration of $5\mathrm{ng}/\mu \mathrm{l}$. Quantification of gene expression for collagen type 1 alpha (COL1A1), transforming growth factor beta 1 (TGFbeta1), interleukin 1 beta (IL1beta), matrix metallopeptidase 1 (MMP1), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (used as reference gene) was performed in a LightCycler Instrument (Roche Diagnostics GmbH, Mannheim, Germany) with the use of the Quanti Tect SYBR Green PCR Master Mix (Qiagen). Primer details are reported in Table 1. The protocol included

Table 1

Primer specifications.

Each sample was tested in duplicate. Data were collected using the LightCycler Software 4.1. Relative quantification was performed using the comparative threshold (Ct) method (ΔCt), where relative gene expression levels equal $2^{-\mathrm{\Delta }\mathrm{\Delta }\mathrm{Ct}}$. Gene expression levels of the target genes were calculated by normalization to the reference gene GAPDH, using the cells untreated as calibrators.

2.6.

Supernatant Enzyme-Linked Immunosorbent Assay Measurements

After 24, 48, and 72 h from laser irradiation, supernatants were collected for collagen type 1 (COLL1), TGFbeta1, and prostaglandin $E_2$ (${\mathrm{PGE}}_2$) determinations by Enzyme-linked Immunosorbent Assay (ELISA) kits following manufacturer’s instructions (R&D Systems, Inc., Minneapolis, Minnesota, for ${\mathrm{PGE}}_2$ and Boster Biological Technology Co, Fremont, California, for other proteins). For TGFbeta1 assays, cell supernatants were chemically activated before testing by two serial steps: 1 N HCl for 10 min followed by 1.2 N NaOH with 0.5 M Hepes for 10 min.

The measured protein concentrations were normalized by DNA content.

2.7.

Statistical Analysis

The Shapiro-Wilk Test was used to assess the normality of data. For scratched-wound area, cell viability, COL1A1 and TGFbeta1 gene expression, the differences between the laser irradiation protocols for each experimental time were analyzed using a one-way multivariate analysis of variance (MANOVA) and Tukey’s honest significant difference post hoc test, while the differences between the three experimental times were evaluated with ANOVA with repeated measures and post hoc tests using the Bonferroni correction. For DNA content, release of COLL1 and TGFbeta1, the differences between the laser irradiation protocols for each experimental time were analyzed using a Kruskall–Wallis H test and a Bonferroni-corrected Mann–Whitney U post hoc test, while the differences between the three experimental times were evaluated with a Friedman test followed by a Bonferroni-corrected Wilcoxon paired sign-rank test for each laser irradiation protocol. All comparisons were performed between laser-irradiated and nonirradiated groups; inter laser-irradiated group comparisons were performed if the former ones were significant. Statistical analyses were performed using the statistical software SPSS for Windows (version 18.0; 2009; SPSS Inc., Chicago, Illinois). The limit for statistical significance was set at $p<0.05$.

3.1.

In Vitro Micro Wound Healing

Cells progressively participated in the healing process enabling a gradual wound closure (Fig. 1). The 5 and $10\mathrm{J}/{\mathrm{cm}}^2$ laser-treated groups were the first to reach complete wound closure after 72 h, followed by the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-treated group after 96 h from irradiation. Nonirradiated controls still showed partial wound healing after 96 h (Fig. 1).

Fig. 1

Representative micrographs of the in vitro scratch wound of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$ or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$) at different times (4, 24, 48, 72, and 96 h). Black lines mark the residual wound area. $\text{Bar}=500\mu \mathrm{m}$.

After 4 h, the 5 and $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups exhibited a significantly decreased wound area compared with the nonirradiated controls ($p<0.0005$, Fig. 2). After 48 and 72 h, all the laser-irradiated groups exhibited a significantly decreased wound area compared with nonirradiated controls ($p<0.0005$ for all, except for $15\mathrm{J}/{\mathrm{cm}}^2$ versus $0\mathrm{J}/{\mathrm{cm}}^2$ at 72 h with $p=0.001$, Fig. 2) with a dose-dependent effect especially at 48 h ($15\mathrm{J}/{\mathrm{cm}}^2$ versus 5 and $10\mathrm{J}/{\mathrm{cm}}^2$ with decreasing significance of $p<0.0005$ and $p=0.001$, respectively, Fig. 2). By analyzing data over the experimental times, the $5\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group and nonirradiated controls significantly decreased the wound area between 4 and 72 h ($p=0.008$, Fig. 2); in addition, control cell cultures significantly decreased the wound area between 4 and 24 h ($p=0.008$, Fig. 2). The $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group showed a significant decrease in wound area at almost all the experimental times (4 h versus 48 and 72 h, $p=0.003$; 48 h versus 72 h, $p=0.002$, Fig. 2).

Fig. 2

Scratched wound area measured at different times (4, 24, 48, and 72 h) of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$) or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$). Data are means; bars are standard deviations. One-way multivariate analysis of variance (MANOVA) and Tukey’s honest significant difference (HSD) post hoc tests: 4 h: ***, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 5 and $10\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$; 48 h: ***, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$, ***, $5\mathrm{J}/{\mathrm{cm}}^2$ versus $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$, **, $10\mathrm{J}/{\mathrm{cm}}^2$ versus $15\mathrm{J}/{\mathrm{cm}}^2$, $p=0.001$; 72 h: ***, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 5 and $10\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$, **, $0\mathrm{J}/{\mathrm{cm}}^2$ versus $15\mathrm{J}/{\mathrm{cm}}^2$, $p=0.001$. ANOVA with repeated measures and post hoc tests using the Bonferroni correction: $0\mathrm{J}/{\mathrm{cm}}^2$: a, 4 h versus 24 and 72 h, $p=0.008$; $5\mathrm{J}/{\mathrm{cm}}^2$: b, 4 h versus 72 h, $p=0.008$; $10\mathrm{J}/{\mathrm{cm}}^2$: c, 4 h versus 48 and 72 h, $p=0.003$, d, 48 h versus 72 h, $p=0.002$.

Similar to other cell cultures, in the presence of MMC, the $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group reached complete wound closure after 72 h and the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group healed after 96 h, while nonirradiated controls still showed partial healing after 96 h from scratch (Fig. 3). The $5\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group, however, reached a complete wound closure after 96 h (with MMC) rather than after 72 h (without MMC). Statistically significant differences between the groups were only slightly less pronounced and more time-delayed in the presence of the cell proliferation inhibitor. The 10 and $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups had decreased wound area compared with nonirradiated controls after 24 h ($p=0.003$ and $p=0.004$, respectively) and 72 h ($p=0.001$) (Fig. 3). Data analysis over the experimental times showed a statistically significant decrease in wound area only for the $5\mathrm{J}/{\mathrm{cm}}^2$ laser-treated group between 4 and 48 h ($p=0.001$, Fig. 3).

Fig. 3

Scratched wound area measured at different times (4, 24, 48, and 72 h) of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$) or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$) in the presence of Mitomycin C (MMC). Data are means; bars are standard deviations. One-way MANOVA and Tukey’s HSD post hoc tests: 24 h: **, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 10 and $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.005$; 72 h: **, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 10 and $15\mathrm{J}/{\mathrm{cm}}^2$, $p=0.001$. ANOVA with repeated measures and post hoc tests using the Bonferroni correction: $5\mathrm{J}/{\mathrm{cm}}^2$: a, 48 h versus 4 h, $p=0.001$.

3.2.

Viability Test and DNA Quantification

There was no statistically significant difference in cell viability between laser-irradiated and nonirradiated groups for each experimental time (Fig. 4). Cell viability showed a statistically significant increase between 24 and 72 h for all laser-irradiated groups ($p<0.0005$ for 5 and $10\mathrm{J}/{\mathrm{cm}}^2$, $p=0.001$ for $15\mathrm{J}/{\mathrm{cm}}^2$), between 24 and 48 h for the 5 and $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups ($p<0.0005$), and between 48 and 72 h for the $5\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group ($p=0.008$).

Fig. 4

Viability results of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$) or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$) after 24, 48, and 72 h from irradiation. Data are means; bars are standard deviations. ANOVA with repeated measures and post hoc tests using the Bonferroni correction: 5 and $10\mathrm{J}/{\mathrm{cm}}^2$: a, 24 h versus 48 and 72 h, $p<0.0005$; $5\mathrm{J}/{\mathrm{cm}}^2$: b, 48 h versus 72 h, $p=0.008$; $15\mathrm{J}/{\mathrm{cm}}^2$: c, 24 h versus 72 h, $p=0.001$.

In agreement with the viability results, no statistically significant difference was found for DNA content between the different treatment protocols or experimental times.

3.3.

Quantification of mRNA Expression Levels by Quantitative Polymerase Chain Reaction

After 24 h, COL1A1 gene expression was increased in all laser-irradiated groups compared with nonirradiated controls ($p<0.0005$, Fig. 5). Moreover, the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group showed an increased COL1A1 gene expression compared with the 5 and $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups ($p=0.001$, Fig. 5). The level of COL1A1 gene expression showed a significant increment at 48 h in the 10 and $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups compared with the nonirradiated group and reached the highest values in the $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group ($p<0.0005$, Fig. 5). After 72 h, the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group showed an increased COL1A1 gene expression compared with the nonirradiated group ($p<0.005$, Fig. 5). Over the experimental times, nonirradiated controls showed an increased COL1A1 gene expression at 72 h compared with 24 h ($p<0.0005$, Fig. 5). The $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group showed an increased COL1A1 gene expression at 48 h compared with 24 h ($p<0.005$, Fig. 5); the same trend was observed in the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group at 72 h compared with 48 h ($p<0.005$, Fig. 5).

Fig. 5

Relative gene expression of COL1A1 of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$) or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$) after 24, 48, and 72 h from irradiation. Data are means; bars are standard deviations. One-way MANOVA and Tukey’s HSD post hoc tests: 24 h: ***, $0\mathrm{J}/{\mathrm{cm}}^2$ versus 5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$; **, $15\mathrm{J}/{\mathrm{cm}}^2$ versus 5 and $10\mathrm{J}/{\mathrm{cm}}^2$, $p=0.001$. 48 h: ***, 10 and $15\mathrm{J}/{\mathrm{cm}}^2$ versus $0\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$; †, $10\mathrm{J}/{\mathrm{cm}}^2$ versus $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$. 72 h: **, $15\mathrm{J}/{\mathrm{cm}}^2$ versus $0\mathrm{J}/{\mathrm{cm}}^2$, $p<0.005$. ANOVA with repeated measures and post hoc tests using the Bonferroni correction: $0\mathrm{J}/{\mathrm{cm}}^2$: a, 24 h versus 72 h, $p<0.0005$; $10\mathrm{J}/{\mathrm{cm}}^2$: b, 24 h versus 48 h, $p<0.005$; $15\mathrm{J}/{\mathrm{cm}}^2$: b, 48 h versus 72 h, $p<0.005$.

The gene expression of TGFbeta1 showed no statistically significant differences after 24 h between the experimental conditions; after 48 h, nonirradiated controls showed an increased expression compared with the $5\mathrm{J}/{\mathrm{cm}}^2$ ($p=0.008$) and the $15\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated groups ($p<0.0005$) (Fig. 6). After 72 h, the $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group exhibited a decreased TGFbeta1 gene expression compared with the nonirradiated group ($p<0.0005$) (Fig. 6). Over the experimental times, nonirradiated controls showed an increased TGFbeta1 gene expression at 24 h compared with 48 h ($p<0.0005$, Fig. 6); at 72 h, the TGFbeta1 gene expression in the $10\mathrm{J}/{\mathrm{cm}}^2$ laser-irradiated group decreased compared with 24 h ($p=0.001$) (Fig. 6).

Fig. 6

Relative gene expression of TGFbeta1 of Saos-2 cells treated with different laser doses (5, 10, and $15\mathrm{J}/{\mathrm{cm}}^2$) or untreated ($0\mathrm{J}/{\mathrm{cm}}^2$) after 24, 48, and 72 h from irradiation. Data are means; bars are standard deviations. One-way MANOVA and Tukey’s HSD post hoc tests: 48 h: **, $0\mathrm{J}/{\mathrm{cm}}^2$ versus $5\mathrm{J}/{\mathrm{cm}}^2$, $p=0.008$; ***, $0\mathrm{J}/{\mathrm{cm}}^2$ versus $15\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$. 72 h: ***, $10\mathrm{J}/{\mathrm{cm}}^2$ versus $0\mathrm{J}/{\mathrm{cm}}^2$, $p<0.0005$. ANOVA with repeated measures and post hoc tests using the Bonferroni correction: $0\mathrm{J}/{\mathrm{cm}}^2$: a, 24 h versus 48 h, $p<0.0005$; $10\mathrm{J}/{\mathrm{cm}}^2$: b, 24 h versus 72 h, $p=0.001$.

Gene expression for IL1beta and MMP1 was undetectable at each experimental time and condition.

3.4.

Supernatant ELISA Measurements

No statistically significant differences were observed between laser-irradiated and nonirradiated groups for COLL1 and TGFbeta1 release. At a protein level, ${\mathrm{PGE}}_2$ was undetectable at each experimental time and condition.