Potential for Drug Repositioning of Midazolam as an Inhibitor of Inflammatory Bone Resorption

Abstract

Drug repositioning is a method for exploring new effects of existing drugs, the safety and pharmacokinetics of which have been confirmed in humans. Here, we demonstrate the potential drug repositioning of midazolam (MDZ), which is used for intravenous sedation, as an inhibitor of inflammatory bone resorption. We cultured a mouse macrophage-like cell line with or without MDZ and evaluated its effects on the induction of differentiation of these cells into osteoclasts. For in vivo investigations, we administered lipopolysaccharide (LPS) together with MDZ (LPS+MDZ) to the parietal region of mice and evaluated the results based on the percentage of bone resorption and calvaria volume. Furthermore, we examined the effects of MDZ on the production of reactive oxygen species (ROS) in cells and on its signaling pathway. MDZ inhibited osteoclast differentiation and bone resorption activity. In animal studies, the LPS+MDZ group showed a decreasing trend associated with the rate of bone resorption. In addition, the bone matrix volume in the LPS+MDZ group was slightly higher than in the LPS only group. MDZ inhibited osteoclast differentiation by decreasing ROS production and thereby negatively regulating the p38 mitogen-activated protein kinase pathway. Thus, we propose that MDZ could potentially be used for treating inflammatory bone resorption, for example, in periodontal disease.

Keywords: bone resorption; drug repositioning; inflammation; midazolam; reactive oxygen species; signal transduction.

Figure 1

Effect of MDZ on osteoclast differentiation of RAW264 cells. (A) Structure of MDZ. (B) TRAP activity in the absence of RANKL and MDZ (RANKL(−)) and in the presence of 300 ng/mL RANKL together with MDZ (0, 5, 10, 20, and 40 μM). For each group, the experiment was carried out at the same time and repeated six times (n = 6). The TRAP activity was determined in triplicate for each group of samples. The asterisks (*) indicate significant differences compared with the activity at 0 µM MDZ (* p < 0.05, nonparametric Steel’s test). (C) Sigmoid curve created based on dose–response data of TRAP activity used for EC₅₀ calculation. The activity values in the 5, 10, 20, and 40 μM MDZ treatments are shown relative to that in the absence of MDZ (0 μM), which was set to 100%. The calculated EC₅₀ value is shown in red. (D) Representative optical microscopic images of TRAP-stained RAW264 cells on day 3 of culture after the addition of MDZ. Observations were performed using a 10× objective lens. The formation of multinucleated osteoclasts was suppressed in the presence of 20 and 40 μM MDZ (Scale bar = 100 μm). MDZ: midazolam; RANKL: receptor activator of nuclear factor kappa B ligand; TRAP: tartrate-resistant acid phosphatase.

Figure 2

Effect of MDZ on the expression of osteoclast differentiation marker genes in RAW264 cells. RAW264 cells were cultured in the absence of RANKL and MDZ (Cont) or in the presence of 300 ng/mL RANKL together with MDZ (0, 10, and 20 μM). Nfatc1: nuclear factor of activated T cells 1; Trap: tartrate-resistant acid phosphatase; and Ctsk: cathepsin K. The level of each mRNA was normalized to that of the reference gene glyceraldehyde-3-phosphate dehydrogenase (Gapdh), and the relative quantification data for Nfatc1, Trap, and Ctsk mRNA levels in RAW264 cells were generated based on a mathematical model for relative quantification in the qPCR system. The asterisk (*) indicates a significant difference in values between each concentration of MDZ (* p < 0.05, Steel–Dwass test). MDZ: midazolam.

Figure 3

Effect of MDZ on osteoclast-mediated bone resorption as determined using the pit formation assay. (A) Representative images of pits. Images of the bottom of CaP-coated plate were captured with a scanner and the resorption pits (black regions) were observed. The formation of resorption pits was suppressed in the presence of 20 and 40 μM MDZ. (B) Percentage area of resorption pits. The area of the black regions was summed using the ImageJ software. For each group, the experiment was carried out at the same time and repeated six times (n = 6). The asterisks (*) indicate significant differences compared with the value obtained at 0 µM MDZ (* p < 0.05, nonparametric Steel’s test). MDZ: midazolam.

Figure 4

Morphological observations of the effect of MDZ on LPS-induced calvarial mouse model. (A) Changes in the body weight of mice during the treatment period. Values are the means ± standard error of the mean (SEM) for six mice (* p < 0.05, Steel–Dwass test). (B) Representative micro-CT 3D reconstruction images of mouse calvaria (Scale bar = 1 mm). (C) Representative stereomicroscopic images of TRAP staining showing osteoclastogenesis in each sample. Cont: PBS-alone dosage group (i.e., control); LPS: LPS-only dosage group; LPS+MDZ: combination of LPS and MDZ dosage group (Scale bar = 1 mm). LPS: lipopolysaccharide; MDZ: midazolam; micro-CT: micro-computed tomography; TRAP: tartrate-resistant acid phosphatase.

Figure 5

Validation of suture width on bone destruction in an LPS-induced calvarial mouse model. (A,B) Locations of lambdoidal and sagittal sutures, respectively, with red lines indicating measurement sites. Scale bars = 5 mm for (A) and 1 mm for (B). (C,D) Average width of lambdoidal and sagittal sutures for each group. Measurements were obtained at ten (lambdoidal) and three (sagittal) sites per suture using the ImageJ software (v.1.52a). Six mice per group (n = 6). Significant differences are indicated by asterisks (* p < 0.05, Steel–Dwass test). Cont: PBS-alone (i.e., control); LPS: LPS-only; LPS+MDZ: LPS and MDZ combination. LPS: lipopolysaccharide; MDZ: midazolam; PBS, phosphate-buffered saline.

Figure 6

Validation of bone matrix volume on bone destruction in LPS-induced calvarial mice. (A,D) Representative images showing the location of lambdoidal (A) and coronal (D) sutures for measurement of bone matrix volume. For each mouse, a block of lambdoidal (1120 µm long × 4000 µm wide) and coronal (1600 µm long × 1600 µm wide) suture in the area indicated by the yellow dotted box was created using a 3D reconstruction software (TRI/3D-BON-FCS64, RATOC, Tokyo, Japan). Scale bars = 2 mm for (A) and = 1 mm for (D). (B,E) Representative images of the upper part block of lambdoidal (B) and coronal (E) sutures. Bone matrix area is shown in white or gray and seam or bone destruction areas are shown in light blue. Scale bars = 2 mm for (B) and = 1 mm for (E). (C,F) Representative images of the lateral block of lambdoidal (C) and coronal (F) sutures. The thickness of each block, indicated by the red arrow, was 365 µm for lambdoidal suture and 231 µm for coronal suture. Scale bars = 2 mm for (C) and = 1 mm for (F). (G,H) Average bone matrix volume of lambdoidal (G) and coronal (H) sutures in each group. The average bone matrix volume for each group was calculated using the 3D reconstruction software. The measurement for each group was performed independently for six mice (n = 6). The asterisks (*) indicate significant differences between each group (* p < 0.05, Steel–Dwass test). Cont: PBS-alone dosage group (i.e., control); LPS: LPS-only dosage group; LPS+MDZ: combination of LPS and MDZ dosage group. LPS: lipopolysaccharide; MDZ: midazolam.

Figure 7

Effects of MDZ on reactive oxygen species (ROS) production and phosphorylated p38 levels in RAW264 cells. (A) Intracellular ROS levels in RAW264 cells. Fluorescence signals were measured using an ROS Detection Cell-Based Assay Kit (DHE) (Ex/Em = 480/570 nm). RANKL: receptor activator of nuclear factor kappa B ligand; MDZ: midazolam; LPS: lipopolysaccharide; AA: antimycin A. Cont: control (i.e., RAW264 cells cultured alone without any substances); RANKL only, LPS only, and AA only: RAW264 cells cultured with RANKL alone, LPS alone, and AA alone, respectively. ROS analysis was performed via six independent experiments (n = 6). The asterisks (*) indicate significant differences between each group (* p < 0.05, nonparametric Steel’s test). (B) Western blotting images showing the levels of p-p38 (top), p38 (middle), and Gapdh (bottom) in RAW264 cells treated with LPS. p38: p38 mitogen-activated protein kinase; p-p38: phosphorylated p-38; Gapdh: glyceraldehyde-3-phosphate dehydrogenase. LPS: lipopolysaccharide; MDZ: midazolam; SB: SB203580. LPS(−): RAW264 cells cultured without LPS; LPS only: RAW264 cells cultured with LPS alone. (C) Analysis of the relative level of p-p38 and p38 normalized against Gapdh levels in LPS-treated RAW264 cells using densitometry of Western blots, employing the ImageJ software (v.1.52a). Western blot analysis was performed via three independent experiments (n = 3). The asterisks (*) indicate significant differences between the LPS-only and each group (* p < 0.05, Student’s t-test). LPS: lipopolysaccharide; MDZ: midazolam; SB: SB203580. LPS(−): RAW264 cells cultured without LPS; LPS only: RAW264 cells cultured with LPS alone.

Figure 8

Schematic diagram illustrating the mechanism of inhibition of RANKL- or LPS-induced osteoclast differentiation by MDZ. MDZ suppresses the production of ROS in mitochondria via PBR and negatively regulates the TRAF6-p38 pathway induced by RANKL or LPS. As a result, differentiation into osteoclasts is inhibited. RANKL: receptor activator of nuclear factor kappa B ligand; RANK: receptor activator of nuclear factor kappa B ligand; TRAF6: tumor necrosis factor receptor-associated factor 6; p38: p38 mitogen-activated protein kinase; MDZ: midazolam; PBR: peripheral benzodiazepine receptor; ROS: reactive oxygen species; LPS: lipopolysaccharide; TLR4: Toll-like receptor 4.