Introduction

Monocarboxylate transporters (MCTs), a family of proton-linked plasma membrane transporters that carry monocarboxylates across the membrane, have been classified into 14 different subtypes1, though only MCT1, MCT2, MCT3, and MCT4 are reported to be involved in the transport of lactate and pyruvate2. In mammalian cells, these 4 subtypes have functions related to uptake and efflux of monocarboxylic acids including lactate and pyruvate across the membrane. Lactic acid and ketone bodies are taken up for oxidation into heart and red muscle cells, as well as neurons, through MCT1 which is ubiquitously expressed, or MCT2, expressed in kidney and brain tissues1,2. Furthermore, lactic acid is also taken up by cells in the liver and kidneys through MCT1 or MCT2 for gluconeogenesis. On the other hand, efflux of monocarboxylic acid from cells occurs via MCT1, MCT3, and MCT4. Lactic acid is produced by glycolysis under hypoxic conditions in nearly all types of cells, then released to the outside via MCT1 to suppress a decrease in intracellular pH and prevent stoppage of glycolysis by accumulating intracellular lactic acid. MCT4 is also involved in the efflux of lactic acid from cells that produce energy by glycolysis even under normoxic conditions3,4. MCT3 has more specific activities, and is expressed only in retinal pigment and choroid plexus epithelium, where it is responsible for transportation of lactic acid to the outside of those cells1.

Based on these backgrounds, the MCT1-4 subtypes have been considered to have functions generally related to maintaining the energy metabolism of cells, as well as homeostasis of intracellular pH by the transportation of lactate and other monocarboxylates5. However, we recently reported that altering the activity of MCT has other significant effects on cells6,7,8. More specifically, MCT1 was required for cell death induced by IL-1β in mouse chondrocyte-like ATDC5 cells6. IL-1β-treated ATDC5 cells expressed 2 different enzymes, namely, inducible NO synthase (iNOS) and phagocyte-type NADPH oxidase (NOX-2). Peroxynitrite, a reaction product of NO and superoxide produced by iNOS and NOX-2, respectively, caused mitochondrial dysfunction, resulting in cell death7. We found that Mct1 silencing reduced intracellular ROS production and late-phase activation of NF-κB, which suppressed NOX-2 expression and cell death6.

In another study, we found that silencing of Mct1 by siRNA strongly suppresses osteoblast differentiation8. Using a culture system in which C2C12 cells, a mouse myoblast cell line, were stimulated with BMP-2 to induce differentiation into osteoblasts, Mct1 silencing suppressed the expression of Runx2 and Sp7, major transcription factors for osteoblast differentiation, as well as that of Tnap, a differentiation marker of mature osteoblasts. These changes were caused by increased expression of Trp53 in C2C12 cells induced by Mct1 silencing. Together, these findings indicate that MCT1 is involved not only in inflammatory cell death but also normal differentiation of osteoblasts, thus suggesting possible involvement of MCT in bone formation and remodeling.

On the other hand, the roles of MCTs in osteoclasts, the counterpart of osteoblasts for maintaining bone metabolism, have not been elucidated. In the present study, we investigated the roles of MCT subtypes in osteoclast differentiation and function.

Results

α-cyano-4-hydroxycinnamic acid (CHC) promoted osteoclast differentiation from BMMs

Bone marrow macrophages (BMMs), obtained by culturing mouse bone marrow cells (BMCs) in the presence of M-CSF, were further cultured for 72 hours in the presence of M-CSF and RANKL to induce their differentiation into osteoclasts. Addition of CHC known to inhibit MCT1, MCT2, MCT3, and MCT4, promoted RANKL-induced osteoclast formation (Fig. 1a). Additionally, CHC (0.3 mM) increased TRAP activity (Fig. 1b) and osteoclast size (Fig. 1c,d), whereas it had no effect on either number of osteoclasts (Fig. 1e) or proliferation of BMMs (Fig. 1f, red line). Hence, the increased TRAP activity by CHC (0.3 mM) in the presence of various concentrations of RANKL (Fig. 1g) was a result of something other than the number of BMMs. On the other hand, addition of CHC at 1.0 mM inhibited osteoclast formation (Fig. 1a), decreased TRAP activity (Fig. 1b), and suppressed BMM proliferation (Fig. 1f, blue line), whereas it did not have a significant effect on the number of osteoclasts (Fig. 1e).

Figure 1
figure 1

Effect of CHC on RANKL-induced differentiation of BMMs into osteoclasts. (a–e) BMMs were cultured for 72 hours in the presence of M-CSF, RANKL, and various concentrations of CHC. (a) Osteoclasts were detected by TRAP activity staining. (b) TRAP activity in the cultures was determined. (c–e) The area of osteoclasts (c), total osteoclast area (d), and the number of osteoclasts (e) per well were quantified using the software package pre-installed in the optical microscope. (f) BMMs were cultured for 24, 48, and 72 hours in the presence of M-CSF and the indicated concentrations of CHC, which was followed by determination of cell number to evaluate BMM proliferation. (g) BMMs were cultured for 72 hours with M-CSF and various concentrations of RANKL in the presence or absence of CHC (0.3 mM). TRAP activity in the cultures was determined. (h) BMMs were cultured for 48 hours in the presence of M-CSF, RANKL, and various concentrations of CHC. Expression levels of Rank, Nfatc1, Dcstamp, Trap, Ctsk, and Irf8 are expressed as relative to that of Gapdh. (a) Scale bars, 500 μm. (b–h) Values are expressed as the mean ± SD. * and ** indicate significant difference at p < 0.05 and p < 0.01, respectively. n.s., not significant.

Full size image

CHC changed expression of genes for osteoclast differentiation markers in BMMs

To analyze the effect of CHC on expression of genes for osteoclast differentiation markers induced by RANKL, various concentrations of CHC were added simultaneously with RANKL to the BMM cultures. CHC at 0.1 mM increased the RANKL-induced expression of Nfatc1, Trap, and Ctsk as compared to the control group (Fig. 1h), which suggested that CHC (0.1 mM) promoted osteoclast differentiation from BMMs. On the other hand, CHC at 1.0 mM decreased the expression levels of Dcstamp, Trap, and Ctsk, while that strongly increased the expression of Irf8, a negative regulator of osteoclastgenensis9. CHC at 1.0 mM also increased the expression of Rank (Fig. 1h).

CHC reduced the number of mature osteoclasts on dentine

Since CHC was shown to modulate osteoclast differentiation of BMMs, we next examined the effect of CHC on bone resorption by mature osteoclasts using pit assays. Osteoclasts obtained by co-cultivation of UAMS32 cells with BMCs were seeded on dentin disks and allowed to form resorption pits. Following osteoclast attachment to the disk, CHC was added to the medium, which resulted in reduction of the number of resorption pits in a concentration-dependent manner (Fig. 2a,c). Scanning electron microscopy findings also confirmed that addition of CHC reduced the number of resorption pits (Fig. 2b). In addition, CHC decreased the number of osteoclasts obtained in co-cultures of BMCs and UAMS32 cells in a concentration-dependent manner (Fig. 2d,e). CHC also reduced the number of mature osteoclasts obtained by culturing BMM in the presence of RANKL (Fig. 2f,g). CHC also decreased TRAP activity in osteoclasts that had differentiated from BMMs in calcium phosphate-coated plates in the presence of M-CSF and RANKL (Supplementary Fig. S1a). Furthermore, CHC inhibited the formation of actin rings, required for attachment of osteoclasts to the bone surface for bone resorption (Supplementary Fig. S1b, red arrowhead). On the other hand, CHC did not induce apoptosis in osteoclasts (Supplementary Fig. S1c). These results suggested that CHC reduced the number of mature osteoclasts at least in part through inhibition of the attachment of mature osteoclasts to the plates and the surface of dentin discs.

Figure 2
figure 2

Effect of CHC on the survival of mature osteoclasts attached to the dentin surface. (a–e) Osteoblastic UAMS32 cells and BMCs were co-cultured in the presence of 1α,25(OH)2D3 (10 nM) and PGE2 (1 μM) for 8 days in collagen-coated plates. Cells were re-plated on dentin discs (a–c) or normal plastic plates (d,e), then cultured for an additional 24 hours in the presence of various concentrations of CHC. (a) Resorption pits formed on dentin discs were stained with toluidine blue. (b) The surface of each dentin disc was observed using scanning electron microscopy. Magnification, 45 × (upper panels) and 250 × (lower panels). (c) The area of resorption pits stained with toluidine blue was quantified using the software package pre-installed in the optical microscope. (d) The number of osteoclasts was determined under a microscope. (e) TRAP activity staining of cultures in normal plastic plates was performed. (f,g) BMMs were cultured for 72 hours in the presence of M-CSF and RANKL to induce differentiated into osteoclasts in normal plastic plates. After counting the initial number of osteoclasts (Start), they were further cultured for 48 hours in the presence or absence of the indicated concentrations of CHC, and the number of osteoclasts remained on the plates was counted (f). Osteoclasts were detected by TRAP activity staining (g). (a,e,g) Scale bars, 500 μm. (c,d,f) Values are shown as the mean ± SD. *Value significantly smaller than that obtained in the control without CHC (p < 0.05).

Full size image

BMMs and osteoclasts expressed Mct1, Mct2, and Mct4

In order to clarify the gene expression profile of MCT subtypes in BMMs and osteoclasts, RNA samples were collected every 24 hours after stimulation with RANKL, and real-time PCR analysis was performed for MCT subtypes. Mct1 expression was increased at 24 hours after stimulation with RANKL and then decreased with the progress of osteoclast differentiation. That of Mct2 was low in BMMs, then increased with the progression of osteoclast differentiation, while the expression of Mct4 was high in BMMs and then decreased with differentiation. Mct3 expression was not detected under the present experimental settings (Fig. 3a). It has been reported that monocarboxylic acid is taken up through MCT1 or MCT2, then flowed out by MCT1, MCT3, or MCT41,2. The present results indicate that BMMs mainly use MCT4 to release monocarboxylic acid from cells, while mature osteoclasts use MCT2 to take up monocarboxylic acid into cells. In addition, it has also been reported that RANKL promotes the glycolytic system in BMMs10. Thus, our findings suggest that the transient increase of Mct1 expression following RANKL stimulation is involved in extracellular release of lactic acid to suppress intracellular acidification and prevent stoppage of glycolysis by accumulation of intracellular lactic acid.

Figure 3
figure 3

Effects of Mct1, Mct2, and Mct4 gene silencing in BMMs on osteoclast differentiation. (a) Expression levels of Mct1, Mct2, Mct3, and Mct4 during the process of RANKL-induced osteoclast differentiation from BMMs. Values are shown as relative to that of Gapdh. ND, not detected. (b–f) Mct1, Mct2, or Mct4 siRNA, or control siRNA was introduced into BMMs. (b) At 24 hours after introduction of each siRNA, expression of mRNA for Mct1, Mct2, and Mct4 was quantitatively analyzed by real-time RT-PCR. Values are normalized to that for Gapdh and shown as relative to the control. (c–e) After introduction of siRNA, cells were cultured for 72 hours in the presence of M-CSF and RANKL. Osteoclasts were detected by TRAP activity staining (c) and the number of osteoclasts per well was counted (d). TRAP activity in the cultures was determined (e). (f) BMMs introduced with each siRNA were cultured for 48 hours in the presence of M-CSF and RANKL. The expressions of Rank, Nfatc1, Dcstamp, Trap, Ctsk, and Irf8 were normalized to that of Gapdh, and expressed as relative values. (a,b,d–f) Values are shown as the mean ± SD. * and ** indicate a significant difference at p < 0.05 and p < 0.01, respectively. (c) Scale bar, 500 μm.

Full size image

Osteoclast formation from BMMs promoted by Mct1 siRNA and suppressed by Mct2 siRNA

Next, we introduced specific siRNAs for Mct1, Mct2, and Mct4 to BMMs to clarify their individual roles in osteoclast differentiation (Fig. 3b). None of the subtype siRNAs had effects on BMM cell proliferation (data not shown). Mct1 siRNA promoted and Mct2 siRNA strongly suppressed osteoclast formation, while Mct4 siRNA had no effect (Fig. 3c,d). Promotion of osteoclast differentiation by Mct1 siRNA and its suppression by Mct2 siRNA was also confirmed by determination of TRAP activities (Fig. 3e).

Introduction of siRNAs for Mct1 and Mct2 into BMMs altered expression of osteoclast differentiation marker genes

The effects of introduction of individual siRNAs for Mct1, Mct2, and Mct4 into BMMs on the expression of genes related to osteoclast differentiation induced by RANKL (Fig. 3f) were investigated. Mct1 siRNA increased the expression of Nfatc1, a transcription factor essential for osteoclast differentiation11, as well as that of Trap and Ctsk, representative marker genes of osteoclasts12,13. These results indicate that MCT1 suppresses osteoclast differentiation through, at least in part, inhibition of Nfatc1 expression. While Mct2 siRNA increased the expression of the Rank gene, it also augmented the expression of Irf89, a transcription factor that negatively regulates osteoclast differentiation, indicating the possibility that MCT2 functions as a promoter of osteoclast differentiation via suppression of Irf8. On the other hand, Mct4 siRNA did not have any effects on the expression of Rank, Nfatc1, Dcstamp, Trap, Ctsk, or Ifr8.

Mct2 gene silencing in mature osteoclasts suppressed their survival/attachment to bone

We also examined the effects of silencing of the genes for MCT1 and MCT2 in mature osteoclasts obtained after stimulation of BMMs for 72 hours by RANKL in order to clarify the role of these transporters in bone resorption. siRNA for Mct1 as well as that for Mct2 suppressed the expression of the corresponding genes in mature osteoclasts (Fig. 4a), whereas the expression of Mct4 in osteoclasts was not suppressed by addition of its siRNA (data not shown). Mct1 siRNA did not have any effect on the resorption of calcium phosphate by osteoclasts. On the other hand, that of Mct2 significantly suppressed calcium phosphate resorption by mature osteoclasts (Fig. 4b,c). Furthermore, introduction of Mct2 siRNA into mature osteoclasts that had been obtained by culturing BMM in the presence of RANKL reduced their number (Fig. 4d,e). These results indicate that MCT2 plays an important role in the osteoclastic bone resorption by supporting the survival of mature osteoclasts or their attachment to the bone surface.

Figure 4
figure 4

Effects of Mct1 and Mct2 gene silencing on the survival/attachment of mature osteoclasts. (a–c) BMMs were cultured in calcium phosphate-coated plates in the presence of M-CSF and RANKL for 72 hours, then Mct1, Mct2, or control siRNA was introduced, and culture continued for 24 hours. (a) At 24 hours after the introduction of siRNAs into osteoclasts, the expression of mRNA for Mct1 and Mct2 was quantitatively analyzed by real-time RT-PCR. Values were normalized by that of Gapdh and shown as relative to the control. (b) Representative photographs of plates at 24 hours after the introduction of siRNA. Lighter portions are areas of resorption. (c) Areas of resorption were quantified using the software package pre-installed in the optical microscope. (d,e) BMMs were cultured for 72 hours in the presence of M-CSF and RANKL to induce differentiated into osteoclasts in normal plastic plates. After counting and the initial number of osteoclasts (Start), they were introduced with Mct1, Mct2, or control siRNA, further cultured for 48 hours, and the number of osteoclasts was counted again (d). Osteoclasts were detected by TRAP activity staining (e). (f) Graphical depiction of a protocol to determine the effect of inhibition of MCTs on osteoclast differentiation and bone-resorbing activities. (g) Graphical depiction of a protocol to determine the effect of silencing of Mct1 and Mct2 genes on osteoclast differentiation and bone-resorbing activities. (a,c,d) Values are shown as the mean ± SD. * and ** indicate a significant difference at p < 0.05 and p < 0.01, respectively.

Full size image

Discussion

The results obtained in this study are presented in a schematic form in Fig. 4f,g. CHC at a low concentration was shown to promote differentiation of BMMs into osteoclasts, whereas that at a high concentration suppressed it. Furthermore, CHC reduced the number of mature osteoclasts and hence suppressed the resorption of calcium phosphate in a concentration-dependent manner (Fig. 4f). Additionally, results obtained in experiments using specific siRNAs for MCT subtypes indicate that MCT1 is a negative regulator and MCT2 a positive regular of osteoclast differentiation of BMMs. The present study also found that MCT2, the expression of which is up-regulated during the process of osteoclast differentiation, is required for their survival or attachment to the surface of the plates or dentin discs (Fig. 4g).

CHC has been reported to inhibit of the 4 MCT subtypes (MCT1-4) that function to transport monocarboxylates14. Among those, the Mct1, Mct2, and Mct4 genes are expressed in BMMs, though their expression varies following RANKL stimulation in a time-dependent manner, making it difficult to identify which is actually inhibited by CHC in BMMs. Nevertheless, our results showed silencing by the specific siRNA of each of the subtypes, thus indicating that the inhibitory effect of CHC towards BMMs is mainly via MCT1. On the other hand, in mature osteoclasts, the subtype inhibited by CHC is considered to be MCT2, as the mRNA of that was most highly expressed among the 3 subtypes at 72 hours after RANKL stimulation, while the introduction of Mct2 siRNA to mature osteoclasts reduced the area of calcium phosphate resorption.

Previously, bone metastatic malignant tumor cells were shown to release lactic acid by an anaerobic metabolism function, which is then taken up by adjacent osteoclasts via MCT1, leading to lactic acid-induced bone resorption15. However, there may be differences regarding the expression of MCT subtypes between humans and mice, though that report noted that the bone resorption of osteoclasts was increased by lactic acid taken in via MCT1, while osteoclast differentiation was not affected. Our findings also indicate that exogenous lactic acid is necessary for survival of mature osteoclasts, because CHC decreased the number of mature osteoclasts and thereby suppressed the amount of bone being resorbed by them. MCT2 was shown to be required to take up monocarboxylic acid into cells16, which was also supported by our finding that Mct2 expression increases with osteoclast differentiation. Furthermore, CHC caused a reduction in the number of mature osteoclasts (Fig. 2d–g). Since the subtype of MCTs mainly expressed in mature osteoclasts was MCT2, it is conceivable that CHC inhibited MCT2. Therefore, MCT2 is considered to be necessary for the survival of mature osteoclasts or their attachment to the surface of dentin discs and calcium phosphate-coated plates.

On the other hand, lactate dehydrogenase (LDH) was found to be activated in RANKL-stimulated BMMs, and stimulated the glycolytic and mitochondrial respiratory metabolisms17. In the present experiments, a decreased concentration of intracellular lactate was observed in BMMs following stimulation by RANKL (data not shown). Thus, an increase in endogenous lactate concentration based on a change in energy metabolism may promote osteoclast differentiation17.

A technical limitation of this study was our inability to determine the movement of lactic acid or other monocarboxylic acids across the cell membrane. If glycolytic progression could be observed18, as shown by NAD+/NADH ratio19, the resultant findings might indicate a correlation of energy metabolism with osteoclast differentiation and function20. Another technical limitation of this study was the inability to quantify the change in the bone-resorbing activity of a single osteoclast after the addition of CHC or the introduction of Mct2 siRNA. Hence, we could not discriminate between the effects of these treatments on differentiation/survival and those on the bone-resorbing activity of osteoclasts.

The primary clinical and scientific implications of the present results are identification of the expressions and roles of MCT subtypes in osteoclasts. Additionally, we found that inhibition or silencing of transporters provides control of the differentiation and functsion of osteoclasts, indicating that transporters on the cell membrane of osteoclasts may serve as therapeutic targets for treatment of osteoporosis and other diseases related to bone metabolism.

Methods

Ethical approval statement

All of the present experimental protocols were approved by the ethical board for animal experiments of Showa University (approval number 17052), and conducted in accordance with Japanese Governmental Law (No. 10069) for the care and use of laboratory animals. The authors declare that all experiments were performed in accordance with relevant guidelines and regulations.

Reagents

Recombinant human macrophage-colony stimulating factor (M-CSF) (Leukoprol®) and recombinant mouse receptor activator of nuclear factor κB (RANK) ligand (RANKL) were purchased from Kyowa Hakko Kogyo Co., Ltd. (Tokyo, Japan) and R&D Systems Inc. (Minneapolis, MN, USA), respectively. CHC (CAS number 28166-41-8), an inhibitor of MCT1, MCT2, MCT3, and MCT4, was purchased from Sigma-Aldrich (St. Louis, MI, USA). Collagenase, 1α,25-dyhydroxyvitamin vitamin D3 [1α,25(OH)2D3], prostaglandin E2 (PGE2), ScreenFectTM siRNA, and Cellmatrix® were purchased from FUJIFILM Wako Pure Chemical Co. (Osaka, Japan). StealthTM siRNAs for mouse Mct1, Mct2, and Mct4, and a negative control siRNA, as well as Alexa FluorTM 488 and 4’6-diamidino-3-phenylondole (DAPI) were obtained from Invitrogen (Carlsbad, CA, USA).

BMMs

BMCs were isolated from the femurs and tibias of 5- to 8-week-old male ddY mice (Japan SLC, Shizuoka, Japan), and cultured for 72 hours in α-minimal essential medium (αMEM) (FUJIFILM Wako Pure Chemical Co.) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich), antibiotics, and M-CSF (4000 U/mL). Adherent cells were regarded as BMMs.

Induction of osteoclast differentiation in BMM cultures

BMMs (1 × 104 cells/well) were plated in the center of the bottom surface of 48-well normal plates. For some experiments, BMMs (2 × 104 cells/well) were plated in 96-well normal plates or 96-well calcium phosphate-coated Osteo Assay Surface plates (Corning Inc., Lowell, MA, USA). They were cultured for 72 hours in αMEM supplemented with 10% FBS and antibiotics, in the presence of M-CSF (4000 U/mL) and RANKL (30 ng/mL), with or without various concentrations of CHC. To obtain mature osteoclasts, BMMs (1 × 107 cells/well) were cultured for 72 hours at 37 °C in 10-cm temperature-sensitive RepCell® plates (CellSeed Inc., Tokyo, Japan) in the presence of M-CSF (4000 U/mL) and RANKL (30 ng/mL). After incubation for 30 minutes at 4 °C, cells were collected as a suspension in the medium using centrifugation and used in the present experiments to examine the effects of CHC on mature osteoclasts.

Induction of osteoclast differentiation in co-cultures of osteoblasts and BMCs

Mouse osteoblastic UAMS32 cells (1 × 106 cells/well) and BMCs (2 × 107 cells/well) were co-cultured in 10-cm dishes coated with collagen (Nitta Gelatin, Yao, Japan) for 8 days in αMEM plus 10% FBS and antibiotics, in the presence of 1α,25(OH)2D3 (10 nM) and PGE2 (1 μM). Cells were detached by incubation for 10 minutes at 37 °C with collagenase (0.4%) and dispase (0.2%) in αMEM, plus 10% FBS and antibiotics. Cells were collected by centrifugation and used for experiments to examine the bone resorption by osteoclasts.

Introduction of siRNAs into BMMs and osteoclasts

StealthTM siRNAs for mouse Mct1, Mct2, and Mct4, as well as negative control siRNA were introduced into BMMs and osteoclasts using ScreenFectTM siRNA by forward transfection. To induce osteoclast differentiation of BMMs, M-CSF and RANKL (50 ng/mL) were added to the medium at the same time as each siRNA was introduced.

Bone resorption of osteoclasts

Osteoclasts obtained from BMM cultures in temperature-sensitive RepCell® plates were re-plated in calcium phosphate-coated plates (Osteo Assay Surface, Corning Inc., Lowell, MA, USA), then cultured for 24 hours. Resorption pits were observed and quantified under an optical microscope (BZ-9000, Keyence, Osaka, Japan). Cells containing osteoclasts obtained in co-cultures of UAMS32 cells and BMCs were plated on dentin discs with a diameter of 6 mm, and cultured for 24 hours in 24-well plates. The dentin discs were sonicated in 5% ammonia in PBS to remove cells, then rinsed with water, air-dried, stained with 1% toluidine blue, and washed again with water. Resorption pits stained with toluidine blue were observed and quantified using an optical microscope (BZ-9000, Keyence).

Real-time RT-PCR

Total RNA was extracted from cells using TRIzol® reagent (Invitrogen), according to the manufacturer’s instructions. Reverse transcription reactions were performed using ReverTra ACE RT qPCR master Mix (TOYOBO Co. Ltd., Osaka, Japan). Quantitative real-time RT-PCR was performed using a TaqManTM Gene Expression Assay (Life Technologies, Carlsbad, CA, USA) and a StepOne Real-time PCR System (Applied Biosystems, Carlsbad, CA, USA). Amplification signals from the target genes were normalized against that of glyceraldehyde 3-phosphatase (Gapdh). The assay IDs for the genes of GAPDH, MCT1, MCT2, MCT3, MCT4, RANK, NFATc1, DCSTAMP, TRAP, Cathepsin K, and IRF8 were as follows: Gapdh, Mm99999915_g1; Mct1, Mm01306379_m1; Mct2, Mm00441442_m1; Mct3, Mm00445115_ m1; Mct4, Mm00446102_m1; Rank, Mmoo437135m1; Nfatc1, Mm00479445_m1; Dcstamp, Mm04209236_m1; Trap, Mm00475698_m1; Ctsk, Mm00484039_m1; and Irf8, Mm00492567_m1, respectively.

Cell proliferation assay

BMMs were plated at a density of 1 × 104 cells/well in 96-well plates, and cultured for various periods up to 72 hours in the presence of M-CSF (4000 U/mL) with or without CHC (0.1, 0.3, 1.0 mM). At the end of each culture period, cells were fixed with 4% paraformaldehyde in PBS, then washed with PBS and incubated for 20 minutes with DAPI solution (Invitrogen). After washing with PBS, cells were observed under a fluorescence microscope (BZ-9000, Keyence) and counted using the pre-installed software package.

Tartrate-resistant acid phosphatase (TRAP) activity staining

Cells were fixed with 4% paraformaldehyde in PBS, then washed with PBS, and treated for 1 minute with a mixture of acetone and ethanol (1:1), washed again with PBS, and incubated for 30 minutes at 37 °C in TRAP buffer (pH 5.2) containing 0.2 mg/mL naphthol AS-MX phosphate (Sigma-Aldrich) and 0.6 mg/mL Fast Red Violet LB Salt (Sigma-Aldrich). After washing with pure water, they were observed under a microscope.

Measurement of the number and area of osteoclasts

The number and area of TRAP activity-positive cells with a diameter of 200 μm or more were determined using a microscope (BZ-9000, Keyence) with analysis software (BZ-X Analyzer 1.3.1.1, Keyence).

Determination of TRAP activity

Cells were homogenized in TRAP buffer with 0.1% Triton-X 100 (Simga-Aldrich) under sonication on ice. Cell lysates (20 μL) were added to 100 μL of TRAP buffer containing 3 mg/mL disodium p-nitrophenyl phosphate (FUJIFILM Wako Pure Chemical Co.). After incubation for 30 minutes at 37 °C, reactions were terminated by addition of 80 μL of 0.5 M NaOH, then absorbance of the reaction mixture at 405 nm was read using a microplate reader (SH-1000, Corona Electric, Ibaraki, Japan).

Fluorescent staining of F-actin

Cells were fixed in 3.7% formaldehyde in PBS and treated with 0.5% Triton X-100 in PBS for 3 minutes. Next, cells were stained with 0.2 U/mL of FITC-phalloidin (Sigma-Aldrich) in 0.1% Triton X-100 for 1 hour. After additional staining with DAPI, they were observed under a fluorescence microscope (BZ-9000, Keyence).

Cell viability assay

Live, dead, and apoptotic cells were quantified using a TaliTM Apoptosis kit with Annexin V Alexa FluorTM 488 and propidium iodide (Thermo Fisher Scientific, Waltham, MA, USA), and a TaliTM image-based Cytometer (Thermo Fisher Scientific).

Statistical analysis

Values are expressed as the mean ± SD. Statistical analysis between 2 groups was performed using Mann-Whitney’s U test and between multiple groups using a Steel-Dwass test. P-values less than 0.05 were considered to indicate statistical significance.