Zoledronic Acid Modulation of TRPV1 Channel Currents in Osteoblast Cell Line and Native Rat and Mouse Bone Marrow-Derived Osteoblasts: Cell Proliferation and Mineralization Effect

Bisphosphonates (BPs) reduce bone pain and fractures by balancing the osteoblast/osteoclast ratio. The behavior of ion channels in the presence of BPs is not known. To investigate this, the effect of zoledronic acid BP (ZOL) (3 × 10−8 to 5 × 10−4 M) treatment, on ion channels, cell proliferation, and mineralization, has been investigated on preosteoclast-like cells, RAW264.7, preosteoblast-like cells MC3T3-E1, and rat/mouse native bone marrow-derived osteoblasts. In whole-cell patch clamp on cell line- and bone marrow-derived osteoblasts, ZOL potentiated outward currents. On RAW264.7, ZOL (10−4 M)-evoked current was reduced by the Kv channel blocker tetraethylammonium hydrochloride (TEA), but not by the selective TRPV1-channel antagonist capsazepine. On MC3T3-E1 cells and bone marrow-derived osteoblasts, ZOL-evoked current (5 × 10−8 to 10−4 M) was reduced by capsazepine, whereas the selective TRPV1-channel agonist capsaicin potentiated the control current. In the cell proliferation assay, 72 h incubation of RAW264.7 and MC3T3-E1 cells with ZOL reduced proliferation, with IC50 values of 2.62 × 10−7 M and 2.02 × 10−5 M, respectively. Mineralization of MC3T3-E1 cells and bone marrow-derived osteoblasts was observed in the presence of capsaicin and ZOL (5 × 10−8–10−7 M); ZOL effects were antagonized by capsazepine. In summary, the ZOL-induced activation of TRPV1 channel mediates the mineralization of osteoblasts and counterbalances the antiproliferative effects, increasing the IC50. This mechanism is not operative in osteoclasts lacking the TRPV1 channel.


Introduction
Bone tissue is one of the preferred sites for metastasis in patients with advanced malignancy, especially in case of prostate, breast cancer, and multiple myeloma. Tumor cells prefer to migrate into heavily vascularized areas of the skeleton, such as the red marrow of the long bones, sternum, pelvis, ribs, and vertebrae [1], because of the support of their microenvironment. The bone microenvironment is considered a fertile "soil", containing cytokines and growth factors that stimulate the growth of in reducing dehydrogenase activity in CCK8 assay, as evaluated using one-way ANOVA analysis between drugs (F = 1.123). The maximal efficacy against RAW264.7 was, however, in favor of ZOL vs. the other BPs, with ZOL being more effective in inhibiting cell proliferation than ALE, as evaluated by Student t-test comparing the Emax percentage (%) of the different drugs (p < 0.05) ( Table  1). Also, in preosteoblast-like cells MC3T3-E1, the three compounds were equally capable of reducing intracellular dehydrogenase activity in the micromolar concentration range, as evaluated using oneway ANOVA analysis between drugs (F = 1.111). The Hill coefficient was <1 for all the compounds in RAW264.7, whereas a slope >1 was calculated for MC3T3-E1. In MC3T3-E1 cells, all BPs caused a mild but not significant increase of dehydrogenase activity in the nanomolar concentration range (3 × 10 −8 to 10 −7 M) (Figure 1a,b).   In the osteoclastogenesis assay performed on differentiated RAW264.7 cells, ALE, RIS and ZOL were capable of inhibiting osteoclastogenesis in the micromolar concentration range after 7 days of Then, BP effects in the nanomolar concentration range were investigated using the mineralization-related osteoblastogenesis assay, to evaluate the capability of drugs to modify calcium phosphate nodule formation. On differentiated MC3T3-E1 cell line, ZOL, RIS, and ALE were capable of inducing significant osteoblastogenesis and mineralization in the nanomolar concentration range (3 × 10 −8 to 5 × 10 −8 M). In our previous work, these concentrations were effective in inducing mineralization in cell lines [6]. At 5 × 10 −8 M, ZOL was the most effective drug in inducing mineralization, which increased by +136.08% ± 21.48% compared to controls (number of replicates = 5) as determined by Student t-test (p < 0.05). At this concentration, RIS and ALE were less effective than ZOL in inducing nodule formation, causing an increase of +65.63% ± 5.22% and +58.78% ± 6.08% vs. controls group (p < 0.05) (number of replicates = 3), respectively. Nodule formation of calcium phosphate precipitate was visible after 10-15 days of incubation of cells with drugs in the mineralized medium ( Figure 3). Instead, no effect of these drugs was observed in the micromolar concentration (data not shown). Mineralization assay with alizarin red S staining for calcium nodules after 15 days of incubation on MC3T3-E1 cells after treatments with alendronate (ALE), risedronate (RIS), and zoledronic acid (ZOL). Cells were treated with (a) normal medium, (b) mineralized medium, mineralized medium in the presence of (c) 3 × 10 −8 M ALE, +38.68% ± 2.18% vs. mineralized medium in b, (d) 5 × 10 −8 M ALE, +58.78% ± 6.08% vs. mineralized medium in b, (e) 3 × 10 −8 M RIS, +45.13% ± 4.12% vs. mineralized medium in b, (f) 5 × 10 −8 M RIS, +65.63% ± 5.22% vs. mineralized medium in b, (g) 3 × 10 −8 M ZOL, +99.18% ± 31.28% vs. mineralized medium in b, (h) 5 × 10 −8 M ZOL, +136.08% ± 21.48% vs. mineralized medium in b. Then, BP effects in the nanomolar concentration range were investigated using the mineralization-related osteoblastogenesis assay, to evaluate the capability of drugs to modify calcium phosphate nodule formation. On differentiated MC3T3-E1 cell line, ZOL, RIS, and ALE were capable of inducing significant osteoblastogenesis and mineralization in the nanomolar concentration range (3 × 10 −8 to 5 × 10 −8 M). In our previous work, these concentrations were effective in inducing mineralization in cell lines [6]. At 5 × 10 −8 M, ZOL was the most effective drug in inducing mineralization, which increased by +136.08% ± 21.48% compared to controls (number of replicates = 5) as determined by Student t-test (p < 0.05). At this concentration, RIS and ALE were less effective than ZOL in inducing nodule formation, causing an increase of +65.63% ± 5.22% and +58.78% ± 6.08% vs. controls group (p < 0.05) (number of replicates = 3), respectively. Nodule formation of calcium phosphate precipitate was visible after 10-15 days of incubation of cells with drugs in the mineralized medium ( Figure 3). Instead, no effect of these drugs was observed in the micromolar concentration (data not shown). Then, BP effects in the nanomolar concentration range were investigated using the mineralization-related osteoblastogenesis assay, to evaluate the capability of drugs to modify calcium phosphate nodule formation. On differentiated MC3T3-E1 cell line, ZOL, RIS, and ALE were capable of inducing significant osteoblastogenesis and mineralization in the nanomolar concentration range (3 × 10 −8 to 5 × 10 −8 M). In our previous work, these concentrations were effective in inducing mineralization in cell lines [6]. At 5 × 10 −8 M, ZOL was the most effective drug in inducing mineralization, which increased by +136.08% ± 21.48% compared to controls (number of replicates = 5) as determined by Student t-test (p < 0.05). At this concentration, RIS and ALE were less effective than ZOL in inducing nodule formation, causing an increase of +65.63% ± 5.22% and +58.78% ± 6.08% vs. controls group (p < 0.05) (number of replicates = 3), respectively. Nodule formation of calcium phosphate precipitate was visible after 10-15 days of incubation of cells with drugs in the mineralized medium ( Figure 3). Instead, no effect of these drugs was observed in the micromolar concentration (data not shown).  Based on these results, ZOL appeared to be the most effective compound in modulating cell activity both in osteoblast and osteoclast cell lines. In fact, the calculated low IC 50 MC3T3-E1/IC 50 RAW264.7 ratio of ZOL of 77, and the osteoclastogenesis assay, revealed a strong selectivity of ZOL for osteoclasts with regard to the reduction of proliferation and the differentiation process. More remarkably, ZOL was able not only to increase dehydrogenase activity in preosteoblast-like cells MC3T3-E1 but, also, to induce notable mineralization in the nanomolar concentration range. Therefore, since ZOL appeared to be the most effective drug in all the performed assays, ZOL actions were further investigated on ion channel currents in whole-cell patch clamp on the target cells.

Characterization of Cation Channel Currents of RAW264.7 and MC3T3-E1 Cell Lines in Controls and in the Presence of ZOL
We first characterized the macroscopic currents on preosteoclast-like cells RAW264.7 in whole-cell patches. Using physiological K + concentrations in the bath and pipette solutions, a hyperbolic current-voltage relationship can be recorded in RAW264.7 cells. The resting potential (Vm) of these cells was −44.49 ± 3.40 mV (number of cells = 48). The application of iberiotoxin (Ibtx), a selective BK channel blocker, to a final concentration of 4 × 10 −7 M, reduced the cationic currents inducing cell depolarization. The co-incubation of the Ibtx with 5 × 10 −3 M TEA did not induce a further reduction of the cationic currents, suggesting that BK channels carried at least 27.9% of the control currents. The co-incubation of the cells with Ibtx, TEA, and 5 × 10 −3 M Ba 2+ caused an almost complete reduction of the control currents, suggesting the contribution of inward-rectifying potassium (Kir) channels and outward-rectifying K + channels to the control currents ( Figure 4a).
The application of 10 −4 M ZOL on the RAW264.7 cells induced a sustained enhancement of the cationic currents both at positive and negative membrane potentials. The co-incubation of the ZOL solution with 5 × 10 −3 M TEA caused a partial reduction of the ZOL-activated currents, suggesting that Kv/BK channels could carry the ZOL-activated currents. The co-incubation of the cells with ZOL, TEA. and 5 × 10 −3 M Ba 2+ caused a further reduction of the currents below controls (Figure 4b). The application of 10 −6 M capsazepine, a selective TRPV1 channel antagonist, failed to reduce ZOL-evoked currents (Figure 4c).
In agreement with our functional observation, the mRNA levels and protein of TRPV1 gene was not detected by others in this cell line [28][29][30].
Next, we characterized the macroscopic currents recorded in the preosteoblast-like cells MC3T3-E1 in whole-cell patches in the same experimental conditions. The Vm of these cells in the control condition was −29.45 ± 2.37 mV (number of cells = 45). At −80 mV (Vm) the current amplitude was −18.2 ± 0.49 pA (number of cells = 45), while a large outward current of +105.1 ± 24.9 pA (number of cells = 45) was recorded at positive membrane potentials. TEA at 5 × 10 −3 M reduced the current amplitude, and co-incubation with 5 × 10 −3 M Ba 2+ caused an incomplete reduction of voltage-dependent currents, suggesting the presence of other channels, beside BK and Kv channels, contributing to the control current ( Figure 5a). The application of 10 −5 M capsaicin, a well-known transient receptor potential vanilloid 1 (TRPV1) channel agonist, elicited large outward currents. Ruthenium red (RR), a non-selective TRP channel blocker, completely reduced the capsaicin-induced currents at 10 −5 M, supporting the idea that the TRPV1 channel is functionally active in these cells (Figure 5b).
The effects of ZOL were investigated on leak currents recorded following incubation of the cells with the unselective K + -channel blockers Ba 2+ (5 × 10 −3 M) and TEA (5 × 10 −3 M) in the Cl − -free bath solution. We found that the application of ZOL (10 −4 M), even in the presence of these blockers, potentiated the outward currents that were antagonized by ruthenium red (10 −5 M), indicating that ZOL-activated currents were carried not only by K + channels, but also by other different cationic channels, possibly TRP ( Figure 5g).
The effects of ZOL were investigated on leak currents recorded following incubation of the cells with the unselective K + -channel blockers Ba 2+ (5 × 10 −3 M) and TEA (5 × 10 −3 M) in the Cl − -free bath solution. We found that the application of ZOL (10 −4 M), even in the presence of these blockers, potentiated the outward currents that were antagonized by ruthenium red (10 −5 M), indicating that ZOL-activated currents were carried not only by K + channels, but also by other different cationic channels, possibly TRP ( Figure 5g).   Since ZOL, at 10 −7 M concentration, did not produce notable activating effects on macroscopic currents following acute application of the drug solution, we also tested the effects of ZOL at this concentration as a function of time in the same MC3T3-E1 cells. After about 30 min of incubation time, ZOL (10 −7 M) caused an enhancement of the outward currents (number of cells = 10). In the majority of the investigated cells, the ZOL-evoked current was markedly reduced by TEA (5 × 10 −3 M) by −66.36% at +100 mV (Vm); a further and full reduction of the current was observed with capsazepine (10 −6 M) (Figure 6a). Cells responsive to ZOL (10 −7 M) were also responsive to capsaicin (10 −5 M). Instead, no activation was observed in cells not responsive to capsaicin (data not shown) after a long-term incubation time with ZOL (10 −7 M). These findings indicate that currents activated by ZOL (10 −7 M) involved TRPV1 and/or Kv/BK channels.
Since ZOL, at 10 −7 M concentration, did not produce notable activating effects on macroscopic currents following acute application of the drug solution, we also tested the effects of ZOL at this concentration as a function of time in the same MC3T3-E1 cells. After about 30 min of incubation time, ZOL (10 −7 M) caused an enhancement of the outward currents (number of cells = 10). In the majority of the investigated cells, the ZOL-evoked current was markedly reduced by TEA (5 × 10 −3 M) by −66.36% at +100 mV (Vm); a further and full reduction of the current was observed with capsazepine (10 −6 M) (Figure 6a). Cells responsive to ZOL (10 −7 M) were also responsive to capsaicin (10 −5 M). Instead, no activation was observed in cells not responsive to capsaicin (data not shown) after a long-term incubation time with ZOL (10 −7 M). These findings indicate that currents activated by ZOL (10 −7 M) involved TRPV1 and/or Kv/BK channels.
Similar effects were observed after the application of ZOL ( In agreement with this observation, the mRNA level of the TRPV1 gene was found to be elevated in this cell line and osteoblasts [31,32].  In agreement with this observation, the mRNA level of the TRPV1 gene was found to be elevated in this cell line and osteoblasts [31,32]. We next investigated whether ZOL was also able to activate currents in native mesenchymal stem cells (MSCs) isolated from both mouse and rat bone marrow [33]. Firstly, macroscopic currents were pharmacologically characterized in MSCs isolated from mouse bone marrow cells. Using asymmetrical, physiological K + concentration in the bath and pipette solutions, a hyperbolic current was observed that crossed the voltage axis −40 ± 8.28 mV (number of cells = 15) and −36.08 ± 3.78 mV (Vm) (number of cells = 12), representing the resting potential of the mouse and rat cells, respectively. TEA (5 × 10 −3 M) as well as capsazepine (10 −6 M) reduced current amplitudes to control values suggesting the contribution of Kv/BK and TRPV1 channels to the total current in either cell population (Figure 7a-d). The application of capsaicin (10 −5 M) potentiates the hyperbolic current, supporting the idea that TRPV1 channels are functionally active in these cells (Figure 7b-e). Both mouse MSCs and rat MSCs, therefore, showed similar properties regarding the response to channel modulators, even if mouse cells showed lower current amplitude in control conditions. On these cells, ZOL (10 −4 M) elicited large, outward currents that were blocked by co-application of capsazepine (10 −6 M) and TEA (5 × 10 −3 M), further confirming ZOL's capability of enhance TRPV1 and Kv/BK channel currents (Figure 7c-f). Especially on mouse MSCs, ZOL (10 −4 M) enhanced current amplitude by +5.37% at −30 mV, +29.82% at +30 mV, and +47.69 at +50 mV vs. controls, and on rat MSCs by +6.94 at −30 mV, +101.37% at +30 mV, and +212.67% at +50 mV. On rat, TEA (5 × 10 −3 M) caused a reduction of −75.35%, and capsazepine (10 −6 M) completely closed ZOL-evoked currents, supporting the idea that Kv/BK channels are responsible for most of the ZOL-activated currents in these cells (Figure 7g,h). ZOL (10 −4 M) also enhanced the inward current by +81.9% at −90 mV in mice and +54.3% in rat, and these effects were fully inhibited by TEA and capsazepine (Figure 7c-f).

Effects of Ion Channel Modulators on ZOL-Induced Mineralization on Osteoblasts
The effects of ion channels modulators on the mineralization process induced by ZOL was evaluated using the mineralization assay both in mouse and rat bone marrow-derived osteoblasts (Figure 8a,b) and MC3T3-E1 cells (Figure 8c). The extracellular matrix Ca 2+ deposits used for mineralized nodule formation were stained with alizarin red S dye, which combines with Ca 2+ ions to stain calcified nodules bright red, after 15 days of culture in the mineralized medium in the presence of ZOL (5 × 10 −8 to 10 −7 M) with or without channel inhibitors. and (l) ZOL-evoked instantaneous current in rat MSCs. (m) Image of murine MSCs as observed by Olympus CX41 4× magnification after crystal violet staining.

Effects of Ion Channel Modulators on ZOL-Induced Mineralization on Osteoblasts
The effects of ion channels modulators on the mineralization process induced by ZOL was evaluated using the mineralization assay both in mouse and rat bone marrow-derived osteoblasts (Figure 8a,b) and MC3T3-E1 cells (Figure 8c). The extracellular matrix Ca 2+ deposits used for mineralized nodule formation were stained with alizarin red S dye, which combines with Ca 2+ ions to stain calcified nodules bright red, after 15 days of culture in the mineralized medium in the presence of ZOL (5 × 10 −8 to 10 −7 M) with or without channel inhibitors. Moreover, capsaicin (10 −6 M) alone on murine bone marrow cells was responsible for inducing strong mineralization leading to an increase of +209.19% ± 45.33% (number of experiments = 3) compared to the control conditions, suggesting a key role for the TRPV1 channel in mineralization and the osteoblastogenesis process. NaF (10 −3 M) a well-known chemical inducing mineralization agent that led to a mineralization increase of +215% ± 58.39% (number of experiments = 3) compared Moreover, capsaicin (10 −6 M) alone on murine bone marrow cells was responsible for inducing strong mineralization leading to an increase of +209.19% ± 45.33% (number of experiments = 3) compared to the control conditions, suggesting a key role for the TRPV1 channel in mineralization and the osteoblastogenesis process. NaF (10 −3 M) a well-known chemical inducing mineralization agent that led to a mineralization increase of +215% ± 58.39% (number of experiments = 3) compared to the control condition on mouse bone marrow-derived osteoblast cells (Figure 8a). The fact that ZOL at 10 −7 M and 10 −8 M concentrations showed equivalent effects in inducing mineralization, and capsazepine is effective against ZOL (10 −7 M), can be explained by the fact that the threshold concentration of ZOL for inducing TRPV1-mediated mineralization is 10 −7 M. This is also the threshold concentration required to activate TRPV1 and Kv/BK channels in osteoblasts in our experiments. Maximal mineralization is observed at 10 −7 M of ZOL and, under these conditions, capsazepine is fully effective.
The capability of ZOL in inducing mineralization was further compared to NaF (10 −3 M) on mouse bone marrow-derived osteoblast cells. We found that ZOL at 5 × 10 −8 and 10 −7 M concentrations was less effective than NaF (10 −3 M) in inducing mineralization. In these experiments, capsazepine was more effective than TEA in antagonizing the ZOL-induced mineralization on mouse bone marrow-derived osteoblast cells (Figure 8b).
A similar capability of TEA and capsazepine in antagonizing ZOL-induced mineralization was observed in MC3T3-E1 cells (Figure 8c).
It is evident that TEA alone at high concentrations caused a reduction of cell proliferation, thereby affecting mineralization after 15 days of incubation time. TEA at 5 × 10 −8 M did not significantly affect, per se, mineralization of bone marrow cells. Also, the incubation of the rat and mouse bone marrow cells with capsazepine (5 × 10 −8 to 10 −7 M) reduced the proliferation at high concentrations while, at 5 × 10 −8 M, this compound did not significantly affect proliferation.

ZOL Effects on Oocytes Transfected with TRPV1 Channel Clone
Finally, the effects of ZOL on TRPV1 channels were evaluated by using excised inside-out macropatches in Xenopus oocytes; it was found that ZOL was capable of activating TRPV1 currents when directly applied to the excised patch ( Figure 9). In particular, ZOL (5 × 10 −8 M) enhanced TRPV1 currents of +51.25% at +30 mV, +21.58% at +60 mV, and +32.06% at +100 mV (Vm) with respect to the current activated by capsaicin (10 −5 M) at the same voltages. Therefore, ZOL at nanomolar concentration activates the TRPV1 channel current expressed in Xenopus oocytes, in excised macropatches, at negative and positive membrane voltages. to the control condition on mouse bone marrow-derived osteoblast cells (Figure 8a). The fact that ZOL at 10 −7 M and 10 −8 M concentrations showed equivalent effects in inducing mineralization, and capsazepine is effective against ZOL (10 −7 M), can be explained by the fact that the threshold concentration of ZOL for inducing TRPV1-mediated mineralization is 10 −7 M. This is also the threshold concentration required to activate TRPV1 and Kv/BK channels in osteoblasts in our experiments. Maximal mineralization is observed at 10 −7 M of ZOL and, under these conditions, capsazepine is fully effective. The capability of ZOL in inducing mineralization was further compared to NaF (10 −3 M) on mouse bone marrow-derived osteoblast cells. We found that ZOL at 5 × 10 −8 and 10 −7 M concentrations was less effective than NaF (10 −3 M) in inducing mineralization. In these experiments, capsazepine was more effective than TEA in antagonizing the ZOL-induced mineralization on mouse bone marrow-derived osteoblast cells (Figure 8b).
A similar capability of TEA and capsazepine in antagonizing ZOL-induced mineralization was observed in MC3T3-E1 cells (Figure 8c).
It is evident that TEA alone at high concentrations caused a reduction of cell proliferation, thereby affecting mineralization after 15 days of incubation time. TEA at 5 × 10 −8 M did not significantly affect, per se, mineralization of bone marrow cells. Also, the incubation of the rat and mouse bone marrow cells with capsazepine (5 × 10 −8 to 10 −7 M) reduced the proliferation at high concentrations while, at 5 × 10 −8 M, this compound did not significantly affect proliferation.

ZOL Effects on Oocytes Transfected with TRPV1 Channel Clone
Finally, the effects of ZOL on TRPV1 channels were evaluated by using excised inside-out macropatches in Xenopus oocytes; it was found that ZOL was capable of activating TRPV1 currents when directly applied to the excised patch ( Figure 9). In particular, ZOL (5 × 10 −8 M) enhanced TRPV1 currents of +51.25% at +30 mV, +21.58% at +60 mV, and +32.06% at +100 mV (Vm) with respect to the current activated by capsaicin (10 −5 M) at the same voltages. Therefore, ZOL at nanomolar concentration activates the TRPV1 channel current expressed in Xenopus oocytes, in excised macropatches, at negative and positive membrane voltages.

Discussion
In the present work, we show that zoledronic acid, at nanomolar concentrations, potentiates outward currents and induces cell proliferation and mineralization of MC3T3-E1 preosteoblast-like cells, and rat and mouse native bone marrow-derived osteoblasts. This drug at micromolar concentrations also has antiproliferative effects on RAW264.7 preosteoclast-like cells and MC3T3-E1 preosteoblast-like cells.
The zoledronic acid-induced mineralization, observed at nanomolar concentrations, is mediated by the activation of TRPV1 channel with the contribution of TEA-sensitive voltage-dependent K + channels. These findings were supported by the fact that ZOL is capable of potentiating the whole-cell currents recorded in MC3T3-E1 cell and the mouse and rat bone marrow-derived osteoblasts, and these currents were partially antagonized by the selective TRPV1 blocker capsazepine and by the unselective Kv/BK channel blocker TEA. The ZOL activating action of the outward currents was observed in physiological conditions, as well as in Cl − -free bath solution, supporting the notion that the currents were carried by cations. Furthermore, the zoledronic acid-induced mineralization was prevented by capsazepine in 8 out of 10 experiments by −29.5%. TEA was also effective in 4 out of 10 experiments in antagonizing the mineralization induced by ZOL by −29.4%. These findings suggest that Kv/BK and TRPV1 channels contribute to the ZOL-induced mineralization by about 60%, while the residual 40% of mineralization is sustained by mechanisms involving, for instance, the activation of connexin 43 and the inhibition of intracellular protein tyrosine phosphatases [5,8,12,13]. Also, it was not possible to use specific BK channel blockers, such as Ibtx or paxilline, because of their irreversible blocking action of the BK channel associated with antiproliferative effects [19,34]. Also, it should be noted that other potent and selective TRPV1 antagonist are available; in our experiments, we use capsazepine to block TRPV1 channel because of the its well-known pharmacology and the lack of action on cell proliferation and mineralization. Despite that in our experiments we could not clearly identify the Kv channel type activated by ZOL, the best candidate is the BK channel. Functional coupling between TRPV1 and BK channels has indeed been reported in neurons, contributing to the modulation of pain [35].
The action of ZOL (5 × 10 −8 M) on TRPV1 channel appears to be mediated by the direct interaction with the channel subunits, as demonstrated by the fact that this drug was capable of activating, in excised macropatches, the TRPV1 channel expressed in Xenopus oocytes. Currently, we do not know the location of the binding sites of ZOL on the channel subunits modulating channel function. It can be speculated that the binding site is located on the cytosolic face of the TRPV1 channel. This hypothesis is supported by the fact that ZOL, at nanomolar concentrations, is more effective in activating the TRPV1 current of a clone expressed in Xenopus oocyte in excised macropatches, rather than the TRPV1 whole-cell current recorded in the osteoblasts. Also, the activating effect of ZOL on the TRPV1 channel current at nanomolar concentrations is rescued following the long-term incubation of the osteoblasts with the drug.
At micromolar concentrations, ZOL and other BPs show antiproliferative effects associated with the inhibition of intracellular target/s, such as the human recombinant enzymes hFPPS and/or hGGPPS [6]. Here, we showed that, in preosteoblast-like cells MC3T3-E1, the BPs under investigation were equally capable of reducing intracellular dehydrogenase activity. The IC 50 MC3T3-E1/IC 50 RAW264.7 ratio used as a cell-selective indicator against RAW264.7 for RIS, ZOL, and ALE were 36.8, 77.0, and 169.0 in favor of RIS; while ZOL was the most effective drug in inducing the maximal reduction of cell proliferation on RAW264.7 vs. other BPs. The concentration-response curves of these drugs were relatively flat, as also demonstrated by the slope coefficient that was <1 for all the compounds in RAW264.7, suggesting a negative cooperativity between drugs at the binding site, whereas a slope >1 was calculated for MC3T3-E1, supporting the involvement of additional target-modulating BP effects on osteoblasts, such as TRPV1 channel.
By contrast, on RAW264.7 cells, the ZOL-evoked currents were not responsive to capsazepine, suggesting that TRPV1 channels are not functionally active in these cells. Previous reports failed to show any evidence of TRPV1 channels in osteoclasts, both using immunoblotting analysis and RT-PCR [28][29][30], whereas RT-PCR analysis revealed that TRPV1 mRNA is expressed in MC3T3-E1 cells [31,32].
Moreover, in the nanomolar concentration range (3 × 10 −8 to 10 −7 M), the BPs failed to increase intracellular dehydrogenase activity on RAW264.7 cells in the cell proliferation assay. These findings suggest that the observed ZOL-activating action of TRPV1 channel is associated with cell proliferation and mineralization of osteoblasts with cytoprotection, explaining its selective action as an antiproliferative drug against the osteoclasts lacking TRPV1 channel current.
We propose the following cascade of events: ZOL-induced transient activation of TRPV1 at nanomolar concentrations may lead to the influx of Ca 2+ ions and osteoblast depolarization with activation of Kv and BK channels in the cells that maintain the membrane potentials with cytoprotection. This mechanism triggers the proliferation and mineralization of osteoblasts. It is expected that TRPV1, once activated by capsaicin and ZOL, would rapidly desensitize in response to persistent activation.
At micromolar concentrations, ZOL induces inhibition of intracellular hFPPS and/or hGGPPS enzymes, with apoptosis in osteoblasts and osteoclasts leading to antiproliferative effects.
In osteoblasts, the ZOL-induced activation of the TRPV1 channel counterbalances the antiproliferative effects associated with the inhibition of intracellular hFPPS and/or hGGPPS enzymes, thereby shifting the concentration-response relationship of the drug to the right on the log concentration axis. This mechanism is not operative in osteoclasts lacking the TRPV1 channel, and explains the osteoclast selective antiproliferative effect of ZOL on these cells.

Ex-Vivo Culture of Native Mouse and Rat Mesenchymal Stem Cells
Native mouse and rat mesenchymal stem cells were obtained as previously described [33]. Briefly, cells were isolated, under sterile condition, from the tibia and femoral marrow compartments and cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 1% L-glutamine, and 1% antibiotics (penicillin-streptomycin) under standard conditions, at 37 • C, in a humidified atmosphere containing 5% CO 2 . Non-adherent cells were carefully removed after 3 h of incubation time. After primary cultures had become almost confluent, they were treated using Trypsin-EDTA Solution 1× (0.25% trypsin, 0.02% EDTA) for 2 min at room temperature. The same protocol was applied 3 times, so that a purified population of MSCs was obtained 3 weeks after the initiation of culture, and then used for experiments. Mouse and rat bone marrow-derived osteoblasts for the mineralization assay were cultured in the mineralization medium.

Ethical Statements
In accordance to 3R RULE (DIRECTIVE 2010/63/EU) Replace, Reduce, Refine, bone marrow cells were collected both from wild-type male mice C57BL sacrificed for other scientific purposes

CCK-8 Intracellular Dehydrogenase Assay
The activity of the intracellular dehydrogenases was evaluated by using the Cell Counting Kit-8 (CCK-8) (Sigma-Aldrich), a highly sensitive test that utilizes water-soluble tetrazolium salt [34,39]. WST-8 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt can be reduced by dehydrogenases in cells, producing an orange colored formazan dye soluble in the culture medium. The cells were counted by using Scepter™ 2.0 (Merck Millipore Corporation, New York, NY, USA) cell counter, seeded in 96-well plates at a density of 8 × 10 3 cells/well, and pre-incubated for 24 h under standard conditions. Then, cells were treated for 72 h with different concentrations of the drug solution. Finally, the activity of the intracellular dehydrogenases was evaluated by adding CCK-8 solution (10 µL) to each well and measuring the absorbance at λ = 450 nm after 2 h of incubation with CCK-8 solution, using the microplate reader Victor ® 3V (PerkinElmer ® , Waltham, MA, USA). The changes of the cell viability were expressed as percentage (%) changes of intracellular dehydrogenase activity induced by the drug with respect to the control.

Mineralization Assay
This assay was performed to evaluate cells' capability to mineralize, that is, their ability to create calcium deposits or nodules after the expression of a mature osteoblast phenotype, as previously described [6]. Briefly, the cells (1 × 10 6 cells/mL/well) were cultured in the mineralization medium using 12-and 24-well plates in the presence of different concentrations of the drug solution for approximately 10-15 days under standard conditions. The medium was refreshed every 3 days. After mineralized nodules have become visible, cells were stained with the alizarin red S staining solution 2% (m/v), able to form a bright red stained calcium complex, and washed 3 times with PBS to avoid non-specific dying. For the quantification, 10% acetic acid solution was added to each well to completely elute the dye, and then the absorbance at λ = 405 nm was measured using the microplate reader Victor ® 3V. The changes of mineralization ability were expressed as the percentage (%) change of mineralization induced by drugs with respect to the control or with respect to 10 −3 M NaF, a well-known chemical compound inducing mineralization [13]. For co-incubation experiments, the capability of channel antagonists to inhibit ZOL-induced mineralization was expressed as percentage (%) changes of mineralization induced with respect to ZOL-induced mineralization. For taking pictures, cells were incubated in 6-well plates containing slide glasses; the staining of randomly selected areas was evaluated by using Olympus CX41 microscope, and thereby photographed at the end of the experiment.

Osteoclastogenesis Assay
The TRAP kit (Sigma-Aldrich, St. Louis, MO, USA) was used to evaluate the ability of drugs to inhibit osteoclast proliferation and activity by detecting the presence of tartrate-resistant acid phosphatase in mature osteoclasts. RAW264.7 cells (8 × 10 4 cells/well) were incubated in 6-well plates containing glass slides to allow photos to be taken at the end of the experiment in a specific medium of differentiation containing 50 ng/mL RANKL. The medium was refreshed every 2 days. After 7 days of differentiation with RANKL, cells were treated with different concentrations of the drug solutions for 72 h; then, they were fixed with 10% formaldehyde/PBS for 15 min, and the plate was washed using deionized water and then stained for TRAP, according to the manufacturer's instructions. The staining of randomly selected areas was evaluated at the microscopy and thereby photographed.
For macropatch experiments on Xenopus oocytes expressing TRPV1 channel, the same solution was used both for the pipette and external recording solution: 150 mM NaCl, 10 mM EGTA, 2 mM MgCl 2 , and 10 mM HEPES (pH = 7.4) [44].
Stock solutions of Ibtx, TEA, Ba 2+ , and Capsz were prepared by dissolving the drugs in dimethyl sulfoxide (DMSO), RR was solubilized in water, and Caps was dissolved in ethanol, at concentrations of 118.6 × 10 −3 M. BPs were provided as a stock solution in phosphate-buffered saline (PBS) at 1 or 10 mM. Stock solutions were stored in the refrigerator, whereas dilute solutions were directly prepared on the day of the experiments and maintained at room temperature. For cell proliferation assays, diluted solutions of drugs were prepared using the medium. For patch clamp experiments, microliter amounts of the stock solutions were added to the bath solutions as needed. DMSO did not exceed 0.07%; at this concentration, this solvent does not normally affect current or cell proliferation.

Whole-Cell Recordings in the Cells
Drug actions on the channel currents recorded during instantaneous I/V relationships were investigated by applying a depolarization protocol, in a range of potentials going from −100 to +100 mV (Vm) for RAW264.7 cells, and from −80 to +120 mV (Vm) for MC3T3-E1 cells in 10 mV steps, starting from HP = −60 mV (Vm), in physiological conditions, with asymmetrical K + ion concentrations (int K + : 132 × 10 −3 M; ext K + : 2.8 × 10 −3 M) using whole-cell patch clamp technique. For mesenchymal stem cells from mouse and rat bone marrow cells, the current was obtained in response to voltage pulses from −90 to +70 mV (Vm) in 20 mV steps. Currents are expressed as densities (pA/pF) to control for cell size/capacitance differences. All experiments were performed at room temperature (20-22 • C) and sampled at 2 kHz (filter = 1 kHz) using an Axopatch-1D amplifier equipped with a CV-4 head-stage (Axon Instruments, Foster City, CA, USA). Patch pipettes were pulled from PG52165-4 Patch Clamp #8250 Glass Capillaries, 1.65mm o.d., 1.1 mm i.d. (World Precision Instruments) with a vertical puller (PP-82 Narishige Tokyo, Japan) to give a resistance of 8-10 MΩ. Data acquisition and analysis was performed using pCLAMP 10 software suite (Axon Instruments, Foster City, CA, USA) as previously described [45][46][47]. Seal resistance was continuous monitoring during the experiment; cells not showing a stable seal (R >> 1 GΩ) were not selected for further analysis.

Heterologous Protein Expression Electrophysiology
For patch clamp experiments in Xenopus oocytes, the transcribed cRNA for TRPV1 channel was injected into oocytes using a Nanoliter 2010 WPI (World Precision Instruments, Sarasota, FL, USA) microinjector. Then, they were incubated in ND96 (96 mM NaCl, 2 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.6) solution at 18 • C for 2-3 d after injection [48]. Patch pipettes were pulled in a horizontal pipette puller (Sutter Instruments) from borosilicate capillary glass (World Precision Instruments; 1B150F-4), and pipette tips were fire-polished using a heating filament under a microscope to promote high-resistance seal formation [48,49].
Inside-out excised patch recordings were performed at 20 • using patch pipettes with a tip size of 15-25 µm (pipette resistance <1 MΩ after polishing). Currents were obtained in response to voltage pulses from −60 to +260 mV (Vm) in 10 mV steps and were recorded with an Axopatch 200B system (Axon Instruments) using Clampex 10 (Axon Instruments) acquisition software. Leak subtraction was performed based on a P/4 protocol [50]. Electrophysiological data were analyzed with Clampfit 10.5 software (Molecular devices corporation, Sunnyvale, CA, USA).

Data Analysis and Statistics
Data were collected and analyzed using Excel software (Microsoft Office 2010); curve fitting was made using SigmaPlot 10.0. The statistical results are presented as mean ± SEM. The number of replicates relative to each experimental dataset was reported in the results paragraph and in the figure description. The Student t-test was used to evaluate the significance of differences between the means of two groups. p values < 0.05 were considered to indicate statistical significance.
For patch clamp experiments on cells, the percentage of activation was calculated as (I drug − I CTRL)/(I max − I CTRL), where I max was evaluated at the maximum voltage applied in the control (CTRL) condition. For patch clamp experiments on Xenopus oocytes, the percentage of activation of ZOL was calculated as (I drug − I CTRL)/(I Caps − I CTRL), referring to the amount of currents activated by the TRPV1-agonist capsaicin (10 −5 M) at the same voltages.

Conclusions
TRPV1 channel can be a novel target for zoledronic acid and structurally related BP compounds, mediating the proliferative cytoprotective action of these drugs on osteoblasts. Further investigations are needed to clarify the binding site/s of this class of drugs on the TRPV1 channel. This mechanism may contribute to the anti-osteoporotic effect of zoledronic acid, as well as the pain-relieving effect. Beneficial effects of zoledronic acid can be expected in skeletal muscle where TRPV1 channel appear to play a role in activating hypertrophic signaling and cytoprotection [51,52].

Patents
Currently no patents are resulting from the work reported in this manuscript.