Human studies were accomplished according to the 1964 Declaration of Helsinki in its latest amendment. The study protocol was approved by the ethics committee of the Medical Faculty of the Eberhard-Karls-Universität and University clinic Tübingen (ethical vote: 538/2016BO2). All participants completed a written consent form in the study. Tissues from patients with (potential) tumor or infectious diseases were excluded from this study.

Medical consultation was mandatory before harvesting bone tissue. All bone samples were obtained from the Department of Traumatology, BG Unfallklinik Tübingen. Bone fragments were isolated from cancellous bone samples and digested with 0.07 % w/v Collagenase II (Serva, Heidelberg, Germany) in PBS for 1 h. Released hOBs were cultured in Dulbecco's Modified Eagle's Medium (DMEM)-high glucose 5 % v/v fetal calf serum (FCS), 50 µmol/L L-ascorbate-2-phosphate, 100 U/mL penicillin and 100 µg/mL streptomycin, 50 µmol/L β-glycerol-phosphate after washing with PBS. Medium change was performed every 4 days. Culture medium was replaced by osteogenic medium (DMEM-high glucose, 1 % v/v FCS, 5 mM β-glycerol-phosphate, 200 µM L-ascorbate-2-phosphate, 1.5 mM CaCl₂, 25 mM HEPES, and 100 nmol/L dexamethasone) to induce osteogenic differentiation (Ehnert et al., 2018[14]).

Delphinol^® is a standardized MBE product that consists of a minimum of 25 % delphinidins, which has a high antioxidant capacity (Gillespie et al., 2007[18]). Delphinol^® is a highly water-soluble powder extract and therefore was dissolved in cell culture medium (without FCS) and diluted to different concentrations (Nagaoka et al., 2019[36]). MBE was sterilized by a 0.22 μm pore filter and then used in the cell cultures.

One commercial cigarette (Marlboro, Philip Morris, New York City, USA) was continuously bubbled through a 25 mL pre-warmed plain DMEM-high glucose medium without FCS in a standard gas wash bottle. A peristaltic pump was used to generate negative pressure at a speed of 95-100 bubbles/ min. The CSE was normalized using a plate reader (BMG Labtech, Ortenberg, Germany) by the optical density at 320 nm (OD₃₂₀). 0.7 of OD₃₂₀ was considered as 100 % CSE. The normalized CSE was diluted with osteogenic differentiation medium after sterile filtration (0.22 μm pore filter) into 5 % CSE, which is associated with smoking 10 cigarettes/day (Sreekumar et al., 2018[46]).

To evaluate the effects of CSE and MBE, and their combined effects on hOBs, the present study was grouped as follows: Control group (Crl): hOBs were cultured in osteogenic medium without supplement up to 21 days; Pre-incubation group (Pre): hOBs were cultured in osteogenic medium with MBE for 7 days and CSE was added subsequently till day 21; Co-incubation group (Co): hOBs were exposed to CSE and MBE in osteogenic medium for 21 days; Post-incubation group (Post): hOBs were cultured in osteogenic medium with CSE for 7 days, and MBE was added subsequently till day 21; Maqui group (MBE): hOBs were incubated with MBE in osteogenic medium for 21 days; CSE group (CSE): hOBs were incubated with CSE in osteogenic medium for 21 days. Osteogenic medium was changed twice a week, and hOBs were exposed to CSE and MBE by changing 10 % volume of medium every day to maintain a certain concentration (Figure 1(Fig. 1)). Measurements were performed on day 4, 7 and 14 for AP activity, cell viability assays, and analysis of gene and protein expression, and on day 21 for the matrix mineralization measurement (late osteogenic marker).

Mitochondrial activity of cells was detected by Resazurin conversion assay. Cells were incubated with 0.025 % w/v Resazurin solution (Sigma-Aldrich, Darmstadt, Germany) (1:10 diluted in PBS) for 30 min at 37 °C. Formed resorufin was measured by fluorescence test (ex/em = 540/590 nm) using a plate reader (BMG Labtech, Ortenberg, Germany), and Resazurin solution without incubation with cells was considered as the background control (solvent mixture without cells) (Ehnert et al., 2016[11]).

SRB staining is a measurement for total protein content of cells. Cells were directly fixed by ethanol for at least 60 min at −20 °C after washing with PBS. After fixation, cells were incubated with 0.4 % w/v SRB for 30 min at room temperature protected from light. Unbound SRB was removed by washing with 1 % v/v acetic acid, and bound SRB was resolved with 10 mmol/ L unbuffered TRIS solution (pH = 10.5). SRB staining results were determined by absorbance test (λ = 565 nm) with a plate reader (Aspera-Werz et al., 2018[3]).

Living cells were determined by calcein AM (intracellular esterase activity) and nucleus were determined by Hoechst 33,342 staining. hOBs were incubated with 2 µM calcein AM and 3.5 µM Hoechst 33,342 (diluted in PBS) at 37 °C for 30 min after washing with PBS. A fluorescence microscope (Epifluorescence: EVOS FL, life technologies, Darmstadt, Germany) was used to detect living cells (green) and cell nucleus (blue).

2′,7′-dichlorofluorescein-diacetate (DCFH-DA) as a fluorescent dye can reveal ROS production in cells. hOBs were incubated with 10 µM DCFH-DA for 30 min at 37 °C. After washing with PBS twice, cells were directly exposed to different concentrations of MBE or 5 % CSE. hOBs which were exposed to 0.001 % v/v H₂O₂ were considered as the positive control. The increase in fluorescence was detected by a plate reader (ex/em=  485/520 nm) (Chen et al., 2020[6]).

Matrix mineralization, which is mediated by the well-differentiated osteoblasts, was measured by Alizarin red staining. Cells were washed with PBS and then directly fixed with ethanol at − 20 °C for at least 60 min. After fixation, cells were washed by using tap water three times and incubated with Alizarin red solution (0.5 % w/v, pH = 4.0) for 30 min at room temperature. Stained matrix (red) can be detected by microscope after washing with tap water at this stage. The bound staining was resolved with 10 %w/v cetylpyridinium chloride solution, and the quantification of alizarin staining was determined by measuring the absorbance (λ = 562 nm) using a plate reader (Ehnert et al., 2015[12]).

AP activity was correlated to early osteoblastic differentiation. Cells were washed with PBS and incubated with AP reaction solution (100 mM TRIS, 0.2 % 4-nitrophenyl-phosphate, 50 mM glycine, 1 mM MgCl₂, pH = 10.5) at 37 °C for 15 min. AP activity results were measured photometrically (λ = 405 nm), and AP reaction solution without incubation with cells was used for control. Results were normalized to cell numbers of hOBs determined by SRB staining (Ehnert et al., 2017[15] (85 V for 40 - 50 min) was carried out to separate the gel).

PCR measurements were used to determine the gene expression. Trifast reagent (38 % v/v phenol, 0.4 mM ammonium thiocyanate, 0.8 mM guanidine thiocyanate, 0.68 mM glycerol, and 3 M sodium acetate solution) was used to isolate total RNA, and quantified using a plate reader. cDNA synthesis was performed by using the cDNA kit according to the manufacturer (Thermo Fisher Scientific, Massachusetts, the USA). Biozym Red HS Taq Master Mix (Vienna, Austria) was used to perform PCR reactions by the guide of manufacturer's manual. The PCR conditions and primer sequences for the investigated genes are shown in Table 1(Tab. 1). 18S was used for housekeeping gene. RNA samples were loaded onto a 1.8 % w/v agarose gel with ethidium bromide. Gel electrophoresis (85 V for 40 - 50 min) was carried out to separate the gel (Ehnert et al., 2017[13]). As a reference for the amplification size of DNA, pUC19/ Msp I (Carl Roth, Karlsruhe, Germany) was used as a DNA marker. Intensity of the bands was measured with ImageJ software (NIH, Bethesda, USA).

Cell lysates of hOBs were prepared by using RIPA buffer, and micro Lowry assay was used to quantify total protein of hOBs. Cell lysates were diluted to obtain a protein content of 1 ng/ μL in ddH₂O. 20 ng protein was transferred to a nitrocellulose membrane via a Dot blotter (Carl Roth, Karlsruhe, Germany). Ponceau staining was performed to confirm the transfer of proteins. Membranes were washed with Tris-buffered saline with Tween 20 (TBS-T) to remove ponceau and then blocked with 5 % BSA in TBS-T for 1 h at ambient temperature. Membranes were incubated with primary antibodies (see detailed information in Table 2(Tab. 2)) overnight at + 4 °C, and GAPDH was used for the housekeeper. The next day membranes were incubated with the secondary antibodies for 2 h at ambient temperature after washing with TBS-T. ECL substrate solution was used for chemiluminescent signal development of membranes. Signals of proteins were detected by a ChemCam imager (INTAS, Göttingen, Germany), and the intensity of signals was quantified by using the ImageJ software (Weng et al., 2020[52]).

GraphPad Prism Software 8.0 (El Camino Real, USA) was used for statistical calculations and figure plotting in the study. Results are obtained from experiments that have been repeated at least three times and are expressed as mean ± SEM. The number of biological (N) and technical (n) replicates is shown in the figure legends. Data sets were compared by non-parametric Kruskal Wallis test followed by Dunn's multiple comparison test using p < 0.05 was taken as minimum level of significance.

We determined cytotoxicity of MBE on hOBs and its ability to induce oxidative stress with concentrations of MBE ranging from 0 - 25 μg/mL. MBE was shown to be toxic to hOBs at concentrations ≥ 25 μg/mL, decreasing resazurin conversion and SRB staining significantly. on day 1, 3 and 7 (Figure 2A and 2B(Fig. 2)). Calcein AM/Hoechst staining also showed that treatment with 25 μg/mL MBE resulted in a significant reduction in the viability of hOBs on day 7 (Figure 2C(Fig. 2)). Additionally, we found high concentrations of MBE significantly increased ROS levels in hOBs (Figure 2D and 2E(Fig. 2)). Low concentrations of MBE had no negative effect on hOBs cell viability and did not increase the ROS level. Among the tested concentrations, 1.56 μg/mL MBE demonstrated the highest hOBs cell viability, which is consistent with the physiological concentrations in humans (Schon et al., 2018[45]). Based on this result, a concentration of 1.56 μg/mL MBE was determined to be used in subsequent experiments.

Nitrotyrosine is a ROS-mediated tyrosine nitration product, which can be used to indicate the level of oxidative stress in cells (Knight et al., 2018[23]). Therefore, ROS levels and nitrotyrosine levels were measured to determine the oxidative stress in the present study. As shown in Figure 3(Fig. 3), both ROS and nitrotyrosine levels were increased significantly by CSE in hOBs. Incubation with MBE could significantly decrease the levels of ROS and nitrotyrosine induced by CSE. Pre-incubation with MBE had the most pronounced effect in reducing elevated nitrotyrosine by CSE.

Nrf2 controls the basal and induced expression of an array of antioxidant response element-dependent genes to regulate the intracellular anti-oxidative system (Ma, 2013[29]). Phosphorylated Nrf2 (pNrf2) is the activated form of native Nrf2 (Chen et al., 2016[5]). Nrf2 gene expression was determined by PCR and pNrf2 protein levels were determined by Dot blot in the present study. As shown in Figure 4(Fig. 4), CSE and MBE both increased Nrf2 expression compared to the control group, and combined incubations with CSE and MBE showed higher Nrf2 expression. Similar holds for the activation of pNrf2 protein. Among three combined incubation methods, pre-incubation with MBE showed the most pronounced activation effect on Nrf2.

Nrf2-regulated anti-oxidative enzymes SOD2, CAT, and GPX were evaluated at the gene and protein level (Amaral et al., 2019[2]). As shown in Figure 5(Fig. 5), incubation with CSE or MBE increased the expression of antioxidant enzymes at the gene and protein level compared to the control group. Combined incubations of CSE and MBE showed higher expression of antioxidant enzymes at the gene and protein level. Specifically, pre-incubation with MBE showed the highest activation effect on the selected enzymes' expression among all groups.

Cell viability, cell differentiation and matrix mineralization were evaluated to investigate the consequences of oxidative damage on hOBs. Resazurin conversion assay, SRB staining, and live/dead staining were used to determine hOB cell viability. Our results showed that CSE exposure significantly decreased hOB cell viability compared to the control and MBE groups; pre-incubation with MBE could significantly prevent the decreased cell viability caused by CSE. AP is an important marker for osteoblast activity during early osteoblast differentiation (Li et al., 2019[27]). As shown in Figure 6D(Fig. 6), CSE suppressed AP activity significantly compared to control and MBE groups, while pre-incubation with MBE significantly enhanced AP activity that was reduced by CSE. Alizarin-red staining was used to determine hOB mineralization (Takanche et al., 2020[48]). Similar to AP activity, CSE exposure decreased hOB matrix mineralization compared to control and MBE groups, and was significantly compensated by pre-incubation with MBE (Figure 6(Fig. 6)).

See also the Supplementary data.