Fluoride Increases Superoxide Production and Impairs the Respiratory Chain in ROS 17/2.8 Osteoblastic Cells

It is known that fluoride produces oxidative stress. Inflammation in bone tissue and an impairment of the respiratory chain of liver have been described in treatments with fluoride. Whether the impairment of the respiratory chain and oxidative stress are related is not known. The aim of this work was to study the effects of fluoride on the production of superoxide radical, the function of the respiratory chain and the increase in oxidative stress in ROS 17/2.8 osteoblastic cells. We measured the effect of fluoride (100 µM) on superoxide production, oxygen consumption, lipid peroxidation and antioxidant enzymes activities of cultured cells following the treatment with fluoride. Fluoride decreased oxygen consumption and increased superoxide production immediately after its addition. Furthermore, chronic treatment with fluoride increased oxidative stress status in osteoblastic cells. These results indicate that fluoride could damage bone tissue by inhibiting the respiratory chain, increasing the production of superoxide radicals and thus of the others reactive oxygen species.


Introduction
Fluoride (F-) is a typical double-edged weapon for human beings. On one hand, its daily administration prevents from tooth cavities and has a mitogenic action on osteoblasts [1]. On the other hand, F-chronic exposure has been demonstrated to be toxic and to cause fluorosis. Several studies have described an increase in bone mineral density in treatments with F- [2]. However, the bone is poorly mineralized and exhibits inflammatory foci [3], which could explain the lack of beneficial effects in treatments with sodium fluoride (NaF). It has been demonstrated that 100 mM of F-decreased the proliferation of osteoblasts and induced apoptosis through the production of reactive oxygen species (ROS) [4].
The generation of ROS, lipid peroxidation and altered antioxidant defence systems are considered to play an important role in the toxic effects of F-. Although damaging effects of F-and ROS production are well documented, the cellular mechanisms by which F-induces ROS formation in bone tissue is still unknown [5].
Mitochondria are considered to be the major source of intracellular reactive oxygen species [6]. The mitochondrial electron transport chain is a major site of superoxide radicals' production, followed by formation of hydrogen peroxide (H 2 O 2 ), which can be converted into the reactive hydroxyl free radical causing oxidative damage [7]. Most oxygen consumed (98%) by cells is used in mitochondria [8] so a key parameter of mitochondrial function is the value of oxygen uptake rate (VO 2 ).
Changes in the oxygen availability or alterations in the electron transport can increase superoxide production.
Previous studies performed in bacteria have demonstrated that F-can be extracelularly protonated to form hydrofluoric acid that freely diffuses through the membrane [9]. Therefore, F-could enter mitochondria following a similar mechanism. We have previously demonstrated that the treatment with F-produced oxidative stress and decreased VO 2 in liver [10]. However the link between the two processes was not found and there is scarce evidence about the effects of therapeutically used concentrations of F-on bone tissue or cells. The aim of this study was to assess the effects of F-on the production of superoxide radical, oxygen consumption and oxidative stress in ROS 17/2.8 osteoblastic cells. Fluoride concentrations used in the experiments described in this paper were within the range of plasma concentrations (10-100 mM) found after the intake of a therapeutic dose of F-(3-20 mgF-/Kg bw. day) [11,12] or water with high fluoride concentration [13].

Fluoride Treatments
Acute experiments. These experiments were carried out to study the effects of F-when it immediately contacts with cells and mitochondria. For this purpose, F-was added to cells or isolated mitochondria while were respiring. Briefly, subconfluent ROS 17/ 2.8 cells were trypsinized and immediately transferred to an oxygen chamber at a density of 1.6610 6 cells/mL. Basal VO 2 was measured for 1 min. Afterwards, the effect of F-(10, 50 or 100 mM) on VO 2 was measured. The experiment was repeated seven times for each F-concentration. Then, isolated mitochondria from subconfluent ROS 17/2.8 cells were obtained and resting mitochondrial VO 2 (state 4) and active VO 2 (state 3) were measured before and after the addition of 100 mM of F-. Finally, superoxide production was studied in isolated mitochondria before and after the addition of 100 mM of F-.
Chronic experiment. This treatment mimics a situation where daily doses of F-are chronically consumed. It has been reported that after an oral F-dose there is a peak plasma concentration of approximately 100 mM followed by a fast return to basal concentration levels [11,16]. In order to reproduce more precisely the in vivo situation, ROS 17/2.8 cells were exposed to a daily 15-min pulse of NaF (100 mM, F-treated cells) or water (control cells) for three consecutive days. After each pulse, culture medium was removed and cells were further incubated for 24 h in fresh medium. After three days, cells were released by trypsinization and used for VO 2 measurements in intact cells, activity of respiratory chain complexes and oxidative stress indexes. The volume of F-solution (treated cells) or distilled water (control cells) was 0.1% of the total culture medium. The chronic experiment was repeated three times.
Oxidative stress indexes were measured in Control and Fgroups. Isolated mitochondria were obtained from control and Ftreated groups (see below) to measure respiratory complexes activities.
The results of the chronic experiment are expressed as percentage of the respective control group (water-exposed), considered as 100%.
Measurement of Oxygen Uptake Rate (VO 2 ) VO 2 was measured in cell suspensions and in isolated mitochondria using a hermetically sealed oxygen measure chamber equipped with a Clark-type electrode maintained at 37uC [17]. Data were recorded with software designed in the laboratory (Biomedical data acquisition suite 1.0). VO 2 from cell suspensions was measured in growth medium. VO 2 from isolated mitochondria was measured in a reaction medium consisting of 0.23 M mannitol, 0.07 M sucrose, 20 mM Tris-HCl, 1 mM EDTA, 3 mM MgCl 2 and 5 mM KH 2 PO 4 /K 2 HPO 4 (pH 7.40). Succinate 6 mM was used as substrate to measure resting respiration (state 4) and 1 mM ADP was added to measure active respiration (state 3). Respiratory control was calculated as the relationship between state 3 respiration and state 4 respiration. To be sure mitochondrial respiration was being measured; 25 mM of KCN (complex IV inhibitor) was added at the end of the measurement.

Isolation of Mitochondria
Mitochondria were isolated from ROS 17/2.8 cells as previously described by Boveris A. [18]. Briefly, cells were lysed by sonication in a MSTE medium consisting of 0.23 M mannitol, 0.07 M sucrose, 10 mM Tris-HCl, 1 mM EDTA, pH 7.40 (3 pulses, 5 seconds, 30% amplitude) using a Vibra-Cell VCX130   Fluoride effects on superoxide production of submitochondrial particles before (Basal) and after adding 100 mM of F-(NaF). Bars and segments represent mean and SD respectively. *Significantly different of NaF compared to Basal. Paired Student's t test, n = 8, p,0.05. doi:10.1371/journal.pone.0100768.g003 device (Sonics & Materials, Newton, CT, USA). Resulting lysates were centrifuged at 500 g for 10 min to discard nuclei and cell debris. The sediment was discarded and the supernatant was centrifuged at 11000 g for 10 min to obtain the enriched mitochondrial fraction. Purity of isolated mitochondria was assessed by determining lactate dehydrogenase activity; only mitochondria with less than 10% impurity were used [19,20]. The total protein concentration of the obtained fractions was measured using a commercial kit based on the red pyrogallolmolibdate method [21] (ProtiU/LCR, Wiener Lab, Rosario, Argentina).

Preparation of Submitochondrial Particles
Submitochondrial particles were obtained for measurement of the activity of respiratory complexes and superoxide production. For this purpose, previously obtained mitochondria were resuspended in MSTE buffer, frozen and thawed three times, and homogenized through the passage of the suspension through a 27G needle 15 mm in length and 0.1 mm in outer diameter [19,20].

Activity of Respiratory Complexes
The determination of the activities of NADH-cytochrome c reductase (Complex I-III) and succinate-cytochrome c reductase (Complex II-III) was based on the reduction of cytochrome c 3+ to cytochrome c 2+ and was followed spectrophotometrically at 550 nm for 2 min at 30uC. The reaction mixture was composed of 100 mM buffer H 2 KPO 4 /HK 2 PO 4 pH 7.40, mitochondrial membranes (0.02 mg protein/ml), 0.5 mM KCN (to inhibit the activity of complex IV), 25 mM cytochrome c 3+ , and 5 mM NADH or 0.2 mM succinate. The activities were calculated as mmol.min 21 .mg protein 21 [22].
The determination of the activity of cytochrome oxidase (Complex IV) was based on the oxidation of cytochrome c 2+ to cytochrome c 3+ and followed spectrophotometrically at 550 nm for 1 min at 30uC. The reaction mixture was composed of 100 mM buffer KH 2 PO 4 /K 2 HPO 4 pH 7.40, mitochondrial membranes (0.1 mg protein/ml) and 40 mM cytochrome c 2+ . As this reaction follows a pseudo-first order kinetic mechanism, the constant (k = min 21 ) of the one-phase exponential decay was used as a measure of the activity of the complex, expressed as min 21 .mg protein 21 [23].

Measurement of Superoxide Production by Submitochondrial Particles
Superoxide production was measured by the SOD-inhibitable oxidation of adrenaline to adrenochrome [24]. The assay medium (pH 7.40) contained 0.23 M mannitol, 0.07 M sucrose, 20 mM Tris-HCl, 1 mM adrenaline and 0.5 mg/mL submitochondrial particles. Superoxide production was iniciated with 7 mM succinate. Formation of adrenochrome was followed spectrophotometrically at 480 nm (e = 4.0 mM/cm) for 2 min.
For the acute experiment, superoxide production was measured before and after the addition of 100 mM of F-.

Measurement of Glutathione Peroxidase (GPx) and Catalase (CAT) Activities
Glutathione peroxidase (GPx) activity was determined in cell lysates following the method of Lawrence and Burk [25] employing 0.25 mM hydrogen peroxide as substrate. The assay medium contained 50 mM buffer KH 2 PO 4 /K 2 HPO 4 pH 7.40, 1 mM EDTA, 1 mM NaN 3 , 0.2 mM NADPH, 1 mM GSH, and 1 U/ml glutathione reductase. The consumption of NADPH was followed spectrophotometrically at 340 nm. The resulting activity was expressed in mmol.min-1.mg of protein-1.
Catalase (CAT) activity was evaluated following Aebi's method [26]. The principle of the assay is based on the determination of the rate constant of hydrogen peroxide decomposition by catalase enzyme. The decomposition of the substrate hydrogen peroxide was monitored spectrophotometrically at 240 nm for 3 min in a reaction medium consisting of 100 mM phosphate buffer (pH 7.40) and 20 mM hydrogen peroxide.

Measurement of Malondialdehyde (MDA)
MDA, the marker of extent lipid peroxidation, was estimated as thiobarbituric acid reactive substances (TBARS) level by the method of Ohkawa [27]. The principle of the method is based on the spectrophotometric measurement of the complex that appeared during thiobarbituric acid's reaction with MDA. Briefly,

Measurement of Oxidized/Total Glutathione Ratio
Oxidized/total glutathione ratio (GSSG/TGSH) was calculated as the quotient between the intracellular oxidized (GSSG) and total glutathione (TGSH) contents determined spectrophotometrically by the enzymatic recycling procedure of Tietze [28], as  modified by Griffith [29]. Briefly, cells were scraped in phosphate saline buffer (125 mM, pH 7.50) supplemented with EDTA (6.3 mM) and lysed by sonication (3 pulses, 5 seconds, 30% amplitude). Resulting lysates were subjected to deproteinization with sulfosalicylic acid (10%) and used for the determination of TGSH or derivatized by incubation with 2-vinylpyridine and triethanolamine (3 ml and 5 ml for 100 ml of deproteinized supernatant respectively) for 1 h at 30uC. Derivatized samples were used for GSSG. Reactions were carried out in phosphate saline buffer-EDTA as above described. NADPH (0.21 M) and dithionitrobenzoic acid (0.6 mM) were used as substrates. 2 and 20 ml of sample were used for TGSH and GSSG respectively and reactions were started by addition of glutathione reductase (0.5 units). TGSH and GSSG contents were quantified through the rate of formation of a coloured derivative of dithionitrobenzoic acid measured spectrophotometrically at 412 nm.

Statistical Analysis
Differences among groups were analysed with One-way analysis of variance (ANOVA) followed by Bonferroni's post test. Students t-test was used to compare two independent samples. Differences were considered significant if p,0.05. All data in text and figures are provided as mean 6 SD. Shapiro and Wilk test demonstrated normal distribution of the samples and Fligner test demonstrated equality of variances of each group. Therefore, parametric tests were used. Data analyses were performed with the package agricolae [30] for R.2.14.1 [31].

Acute Experiments
After the addition of 10, 50 or 100 mM of F-a significant decrease in cells VO 2 was observed for all F-concentrations ( Figure 1). The most important inhibition (40.4%) was assessed when cells were exposed to 100 mM of F-. On the basis of the results obtained, we decided to use the highest F-concentration in subsequent experiments.
The addition of 100 mM of F-to active mitochondria produced a significant decrease in mitochondrial respiration, both in state 4 and state 3 ( Figure 2). These results coincided with previous VO 2 inhibition observed in the osteoblasts and demonstrated that Finhibits the respiratory chain immediately after its addition.
Finally, superoxide production in the absence or presence of Fwas assessed to verify whether fluoride-inhibition of the respiratory chain increases the production of superoxide radicals. The results shown in Figure 3 demonstrate a significant increase in superoxide production after the addition of 100 mM of F-to submitochondrial particles.

Chronic Experiment
F-decreased VO 2 of osteoblastic cells after 3 pulses of NaF 100 mM of 15 min duration (Figure 4).
The analysis of mitochondrial complexes revealed that there was a decrease of all respiratory complexes activities after the third pulse of F-in F-group. A significant inhibition along time was observed for complex IV reaching an 80% inhibition in F-group (Table 1).
Finally, oxidative stress measurements revealed a significant increase in the levels of TBARS of F-exposed cells compared to Control group ( Table 2). The GSSG/TGSH ratio was also increased in cells exposed to F-. Finally, no changes in antioxidants enzymes activities were observed after three pulses with F-.

Discussion
The increase in oxidative stress damage caused by F-is well documented [32], but the mechanisms involved in ROS generation are still unknown. One possible explanation is that F-could trigger oxidative stress via inhibition of the pentose phosphate oxidative pathway [33]. In addition, F-induced apoptosis by oxidative stress-induced lipid peroxidation, causing the release of cytochrome c through HL-60 cells mitochondria [34]. Presently, no mechanism for mitochondrial ROS generation by F-in osteoblasts has been proposed. The present contribution has been aimed at investigating whether F-could modify the activity of the respiratory chain in osteoblasts-like cells, changing the rate of production of oxygen reactive species.
We observed that as soon as F-reaches the cells (acute experiment), it significantly inhibits their respiration measured as VO 2 . Also, F-inhibits mitochondrial VO 2 in both state 4 and state 3 and increases superoxide production. According with these results, the oxygen uptake rate also decreased in chronic treatments with F-. The increase in superoxide radicals as soon as F-reaches the mitochondrion could explain the increase in GSSG/TGSH ratio due to an augmented amount of peroxides reacting with reduced glutathione and producing higher amounts of oxidized glutathione [35]. The antioxidant enzymes activities were not modified after three pulses of F-and the decreased in GSSG/TGSH could explain the increase in lipid peroxidation after three pulses of F-.
Recently it has been reported that the constant exposure to Ffor 72 h is able to induce apoptosis in osteoblasts through increasing oxidative stress. In that work F-was administered to osteoblasts in constant concentration for 72 h [36]. In the experiments described in our paper, osteoblasts are exposed for a few minutes to a high concentration of F-, as it occurs in vivo. Although F-levels returned to basal levels 24 h after each pulse of F-, an inhibition of oxygen consumption and respiratory complexes activities and an increase in oxidative stress status were observed.
The results shown in this paper also demonstrate that chronic administration of F-produces a significant decrease in mitochondrial respiratory chain activity. F-treatment significantly inhibited complex IV and partially decreased (30% inhibition) complex I-II and complex I-III activities. As complex I and III are the main sites of superoxide radical synthesis [37], F-could enhance its production by inhibiting the mitochondrial activity at respiratory chain level. F-inhibition at complex IV level could enhance the formation of ubiquinone radical which in turns can react with molecular oxygen increasing superoxide radical production [38]. When the respiratory chain is inhibited (in complex IV in this case), the electron supply reduces the ubiquinone (Q) pool and in the presence of large proton motive force, the electrons are forced back from reduced ubiquinone (QH 2 ) into complex I (by electron reverse transport), which has two possible sites of superoxide production: the flavin in the NADH-oxidizing site and the ubiquinone-reducing site [39]. Recently it has been demonstrated that when succinate is used as electron donor, most superoxide is produced at the ubiquinone reduction site [40].
Taking all these results into account, we are able to describe the effect of F-on mitochondrial ROS production and its relationship with oxidative stress and inflammation ( Figure 5). After an oral dose, F-may inhibit the respiratory chain, increasing the production of superoxide radical (by the possible mechanism explained before) and thereby of hydroxide peroxide and peroxynitrite [7]. Antioxidant enzymes activities cannot prevent increased free radical formation. Therefore, there is an increase in ROS that finally produce oxidations in membranes and damage the cell macromolecules (as seen by the increase in lipid peroxidation) and may be the cause of the inflammatory foci observed in the bone. It has already been demonstrated that ROS production induced inflammatory gene expression in alveolar macrophages [41], fibroblasts [42] and kidney [43]. Therefore, bone inflammatory foci could be enhanced via ROS-dependent activation of pro-inflammatory genes.

Supporting Information
Data S1 Raw data from each figure and