Aesculetin Inhibits Osteoclastic Bone Resorption Through Blocking Ruﬄed Border Formation and Lysosomal Traﬃcking

International Journal of Molecular Sciences

Article Aesculetin Inhibits Osteoclastic Bone Resorption through Blocking Ruﬄed Border Formation and Lysosomal Traﬃcking

Woojin Na †, Eun-Jung Lee †, Min-Kyung Kang, Yun-Ho Kim, Dong Yeon Kim, Hyeongjoo Oh, Soo-Il Kim, Su Yeon Oh and Young-Hee Kang *

Department of Food Science and Nutrition and The Korean Institute of Nutrition, Hallym University, Chuncheon 24252, Korea; [email protected] (W.N.); [email protected] (E.-J.L.); [email protected] (M.-K.K.); [email protected] (Y.-H.K.); [email protected] (D.Y.K.); [email protected] (H.O.); [email protected] (S.-I.K.); [email protected] (S.Y.O.) * Correspondence: [email protected]; Tel.: +82-33-248-2132; Fax: +82-33-254-1475 These authors contributed equally to this work. † Received: 22 September 2020; Accepted: 10 November 2020; Published: 13 November 2020

Abstract: For the optimal resorption of mineralized bone matrix, osteoclasts require the generation of the ruffled border and acidic resorption lacuna through lysosomal trafficking and exocytosis. Coumarin-type aesculetin is a naturally occurring compound with anti-inflammatory and antibacterial effects. However, the direct effects of aesculetin on osteoclastogenesis remain to be elucidated. This study found that aesculetin inhibited osteoclast activation and bone resorption through blocking formation and exocytosis of lysosomes. Raw 264.7 cells were differentiated in the presence of 50 ng/mL receptor activator of nuclear factor-κB ligand (RANKL) and treated with 1–10 µM aesculetin. Differentiation, bone resorption, and lysosome biogenesis of osteoclasts were determined by tartrate-resistance acid phosphatase (TRAP) staining, bone resorption assay, Western blotting, immunocytochemical analysis, and LysoTracker staining. Aesculetin inhibited RANKL-induced formation of multinucleated osteoclasts with a reduction of TRAP activity. Micromolar aesculetin deterred the actin ring formation through inhibition of induction of αvβ3 integrin and Cdc42 but not cluster of differentiation 44 (CD44) in RANKL-exposed osteoclasts. Administering aesculetin to RANKL-exposed osteoclasts attenuated the induction of autophagy-related proteins, microtubule-associated protein light chain 3, and small GTPase Rab7, hampering the lysosomal trafficking onto ruffled border crucial for bone resorption. In addition, aesculetin curtailed cellular induction of Pleckstrin homology domain-containing protein family member 1 and lissencephaly-1 involved in lysosome positioning to microtubules involved in the lysosomal transport within mature osteoclasts. These results demonstrate that aesculetin retarded osteoclast differentiation and impaired lysosomal trafficking and exocytosis for the formation of the putative ruffled border. Therefore, aesculetin may be a potential osteoprotective agent targeting RANKL-induced osteoclastic born resorption for medicinal use.

Keywords: aesculetin; bone resorption; lysosomes; microtubules; osteoclast; ruﬄed border

1. Introduction Old and damaged bones are replaced with newly formed ones during a highly controlled bone remodeling process that is crucial for preserving skeleton integrity and mineral homeostasis [1,2]. Normal bone remodeling entails a tight coupling of osteoclastic bone resorption to osteoblastic bone formation, maintaining optimal bone mass or quality [2,3]. However, this physiological process

Int. J. Mol. Sci. 2020, 21, 8581; doi:10.3390/ijms21228581 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2020, 21, 8581 2 of 17 can be deranged by various factors such as menopause-associated hormonal changes, age-related factors, changes in physical activity, drugs, and secondary diseases [2]. Numerous studies have elucidated the cellular and molecular mechanisms that regulate the finely coordinated bone remodeling involving activation, resorption, reversal, formation, and termination [4,5]. The key signaling pathways regulating osteoclastogenesis and osteoblastogenesis are known to be receptor activator of nuclear factor-κB (RANK)/RANK ligand (RANKL)/osteoprotegerin (OPG) and canonical Wnt signaling [6–8]. Additionally, bone remodeling involves several paracrine and endocrine regulators such as cytokines and growth factors [1,9]. Understanding diverse mechanisms underlying bone remodeling provides targets for pharmacological interventions [10–12]. Currently available osteoanabolic therapies are the parathyroid hormone and its related peptide synthetic analogues, and antiresorptive therapies include bisphosphonates or denosumab [10,12]. However, definite data are needed to confirm the efficacy and safety of such therapeutic interventions. Osteoclasts are multinucleated hematopoietic cells that are specialized for bone resorption [7,8]. Emerging findings reveal evidence that the ruffled border of mature osteoclasts in contact with the bone surface has discrete subdomains that are operated by several membrane trafficking pathways in osteoclasts [13]. The ruffled border is formed by massive fusion of secretory lysosomes filled with resorptive enzymes [14,15]. Bone resorption by osteoclasts highly depends on proteases, acid hydrolases and acid phosphatases including tartrate resistant acid phosphatase (TRAP), matrix metalloproteinases (MMP), and cathepsin K [13]. Additionally, the bone surface acidification is responsible for digesting inorganic materials of bone matrices [15,16]. Osteoclastic function of resorbing the mineralized bone matrix depends on formation of resorption lacuna presenting specific proteases and low pH [13]. Defect of proteases such as TRAP, MMP, and cathepsin K fails to digest organic components of bone matrices [17]. The inactivation of ruffled border proteins essential for the extracellular acidification results in osteopetrosis manifested by dense bones [17–19]. The molecular machinery governing positioning and peripheral distribution of lysosomes has recently emerged [13,20]. Several intersecting endocytic, secretory, transcytotic, and autophagic pathways take place during the formation discrete subdomains of the ruffled border. Lysosome trafficking depends on regulation of cytoskeletal and lysosomal proteins of Pleckstrin homology domain-containing protein family member 1 (PLEKHM1) and lissencephaly-1 (LIS1) [21,22]. The lysosomal membrane proteins of synaptotagmin VII (Syt VII) and lysosome-associated membrane protein 2 (LAMP2) regulate cathepsin K secretion and formation of the ruffled border in osteoclasts. Loss of connectivity of lysosomes to microtubules impairs bone resorption and lysosomal distribution in osteoclasts [20]. In addition, formation of the ruffled border entails several autophagy-related proteins (Atg) in concert with microtubule-associated protein light chain 3 (LC3) for actin ring formation, cathepsin K release, and bone resorption of osteoclasts [23,24]. However, the molecular mechanisms regulating lysosomal biogenesis and function in osteoclasts remain unclear. Distinct vesicle transport and protrusion along microtubules in osteoclasts may be exploited for the development of new therapies for metabolic bone disorders. There have been considerable efforts in developing novel therapeutic agents targeting bone loss with natural plant-derived compounds that do not exhibit any undesirable effects [25,26]. However, the mechanistic efficacy of these compounds in counteracting bone loss remains elusive. Gossypetin, a bioactive compound of Hibiscus sabdariffa, antagonized bone resorption through inhibiting the ruffled border formation [27]. On the basis of the evidence that bone resorption in osteoclasts relies on a molecular apparatus linking lysosomes to microtubules, the present study examined whether aesculetin (Figure1A) and its glucoside aesculin (Figure2A) inhibited bone resorption through disturbing the lysosomal intracellular trafficking to the ruffled border in RANKL-exposed murine Raw 264.7 macrophages. Int. J. Mol. Sci. 2020, 21, 8581 3 of 17 Int. J. Mol. Sci. 2020, 21, x 3 of 19

Figure 1. Chemical structure of aesculetin (A), cytotoxicity of Raw 264.7 macrophages by 1–15 µM Figure 1. Chemical structure of aesculetin (A), cytotoxicity of Raw 264.7 macrophages by 1–15 μM aesculin in the absence (B) and presence of receptor activator of nuclear factor-κB ligand (RANKL) (C), aesculin in the absence (B) and presence of receptor activator of nuclear factor-κB ligand (RANKL) and inhibitory effects of aesculetin on tartrate-resistance acid phosphatase (TRAP) staining (D) and (C), and inhibitory effects of aesculetin on tartrate-resistance acid phosphatase (TRAP) staining (D) TRAP activity (E). Raw 264.7 cells were cultured for 3 days with 1–15 µM aesculetin in the absence and TRAP activity (E). Raw 264.7 cells were cultured for 3 days with 1–15 μM aesculetin in the absence and presence of 50 ng/mL RANKL. Cell viability was measured by 3-(4,5-dimetylthiazol-yl)-diphenyl and presence of 50 ng/mL RANKL. Cell viability was measured by 3-(4,5-dimetylthiazol-yl)-diphenyl tetrazolium bromide (MTT) assay (B,C). Bar graphs for viability (mean SEM, n = 3) are expressed tetrazolium bromide (MTT) assay (B,C). Bar graphs for viability (mean ± ±SEM, n = 3) are expressed as as percentage cell survival compared to untreated cells. For the TRAP staining and activity (D,E), percentage cell survival compared to untreated cells. For the TRAP staining and activity (D,E), Raw Raw 264.7 cells were cultured for 5 days with 50 ng/mL RANKL in the absence or presence of 264.7 cells were cultured for 5 days with 50 ng/mL RANKL in the absence or presence of 1–10 μM 1–10 µM aesculetin. TRAP-positive multinucleated osteoclasts were observed under light microscopy aesculetin. TRAP-positive multinucleated osteoclasts were observed under light microscopy (four (four separate experiments). Scale bar = 20 µm. TRAP activity (mean SEM, n = 3) was measured separate experiments). Scale bar = 20 μm. TRAP activity (mean ± SEM, n± = 3) was measured at λ = 405 at λ = 405 nm. Respective values in bar graphs not sharing a small letter are significantly different at nm. Respective values in bar graphs not sharing a small letter are significantly different at p < 0.05. p < 0.05.

Int. J. Mol. Sci. 2020, 21, 8581 4 of 17 Int. J. Mol. Sci. 2020, 21, x 4 of 19

µ Figure 2.2. ChemicalChemical structure structure of of aesculin aesculin (A ),(A cytotoxicity), cytotoxicity of Raw of 264.7Raw macrophages264.7 macrophages by 1–40 byM 1–40 aesculin μM (aesculinB,C), and (B inhibitory,C), and einhibitoryffects of aesculin effects onof tartrate-resistanceaesculin on tartrate-resistance acid phosphatase acid (TRAP) phosphatase activity ((TRAP)D) and µ activitystaining ( (DE).) and Raw staining 264.7 cells (E were). Raw cultured 264.7 cells for 3 were days cultured with 1–40 forM 3 aesculindays with in the1–40 absence μM aesculin and presence in the κ absenceof 50 ng /mLand receptorpresence activator of 50 ng/mL of nuclear receptor factor- activatorB ligand of (RANKL).nuclear factor- Cell viabilityκB ligand was (RANKL). measured Cell by 3-(4,5-dimetylthiazol-yl)-diphenyl tetrazolium bromide (MTT) assay (B,C). Bar graphs for viability viability was measured by 3-(4,5-dimetylthiazol-yl)-diphenyl tetrazolium bromide (MTT) assay (B,C). (mean SEM, n = 3) are expressed as percentage cell survival compared to untreated cells. For the Bar graphs± for viability (mean ± SEM, n = 3) are expressed as percentage cell survival compared to TRAP activity and staining (D,E), Raw 264.7 cells were cultured for 5 days with 50 ng/mL RANKL untreated cells. For the TRAP activity and staining (D,E), Raw 264.7 cells were cultured for 5 days in the absence or presence of 1–40 µM aesculin. TRAP activity (mean SEM, n = 3) was measured with 50 ng/mL RANKL in the absence or presence of 1–40 μM aesculin.± TRAP activity (mean ± SEM, at λ = 405 nm. TRAP-positive multinucleated osteoclasts were observed under light microscopy n = 3) was measured at λ = 405 nm. TRAP-positive multinucleated osteoclasts were observed under (four separate experiments). Scale bar = 20 µm. Respective values in bar graphs not sharing a small light microscopy (four separate experiments). Scale bar = 20 μm. Respective values in bar graphs not letter are significantly different at p < 0.05. sharing a small letter are significantly different at p < 0.05. 2. Results 2. Results 2.1. Blockade of RANKL-Induced Osteoclast Differentiation by Aesculetin 2.1. Blockade of RANKL-Induced Osteoclast Differentiation by Aesculetin Mature osteoclasts are multinucleated giant cells differentiated by the RANKL/RANK system fromMature their precursors osteoclasts [6 ,7are]. Tomultinucleated determine the giant inhibitory cells differentiated effect of nontoxic by the aesculetin RANKL/RANK on the osteoclast system fromdifferentiation, their precursors Raw 264.7[6,7]. To macrophages determine the were inhibitory treated effect with 1–10of nontµMoxic aesculetin aesculetin in on the the presence osteoclast of differentiation,50 ng/mL RANKL. Raw RANKL 264.7 macrop promotedhages the were differentiation treated with of Raw 1–10 264.7 μM macrophagesaesculetin in intothe presence TRAP-positive of 50 ng/mLmultinucleated RANKL. osteoclasts RANKL promoted (Figure1 D,E).the differentiation However, there of wasRaw a264.7 marked macrophages inhibition into of the TRAP-positive formation of TRAP-positivemultinucleated osteoclastsosteoclasts in(Figure the presence 1D,E). However, of 5 µM there aesculetin was a (Figure marked1D). inhibition Consistently, of the theformation TRAP ≥ ofactivity TRAP-positive enhanced osteoclasts by RANKL in wasthe presence reduced of in ≥5 μ5Mµ Maesculetin aesculetin-added (Figure 1D). cells Consistently, (Figure1E). the On TRAP the ≥ activityother hand, enhanced 1–40 µ byM RANKL aesculin was did notreduced reduce in the≥ 5 μ TRAPM aesculetin-added activity significantly cells (Figure in RANKL-loaded 1E). On the other Raw hand,264.7 cells1–40 (Figure μM aesculin2D). In did addition, not reduce nontoxic the aesculinTRAP activity did not significantly influence the in formation RANKL-loaded of TRAP-positive Raw 264.7 cellsmultinucleated (Figure 2D). osteoclasts In addition, by RANKLnontoxic (Figureaesculin2E). did Accordingly, not influence aesculetin the formation but not of aesculin TRAP-positive within multinucleatedthe limits of cytotoxicity osteoclasts may by inhibitRANKL the (Figure differentiation 2E). Accordingly, of osteoclasts aesculetin from common but not progenitoraesculin within cells theincluding limits of macrophages. cytotoxicity may inhibit the differentiation of osteoclasts from common progenitor cells including macrophages.

2.2. Inhibition of Lacunar Acidification and Bone Resorption by Aesculetin

Int. J. Mol. Sci. 2020, 21, 8581 5 of 17

2.2. Inhibition of Lacunar Acidification and Bone Resorption by Aesculetin The carbonic anhydrase II (CAII) offers the proton source for extracellular acidification, and the Cl−/HCO3− exchanger anion exchange protein 2 (Ae2) can serve as cellular entry mechanism for Cl− during bone resorption [16,18]. RANKL promoted CAII induction, while 5µM aesculetin attenuated ≥ any such induction (Figure3A). In addition, the RANKL induction of Ae2 protein was suppressed by administering 5 µM aesculetin to Raw 264.7 macrophages (Figure3A). On the other hand, the chloride ≥ channel 7 (ClC-7) and the proton-pumping vacuolar-type H(+)-ATPase (V-ATPase) localized to the ruffled membrane contribute to lacunar acidification of the osteoclast-bone interface [16,19]. Similarly, aesculetin reduced the cellular levels of ClC-7 and V-ATPase enhanced in the presence of 50 ng/mL RANKL (Figure3B), indicating that aesculetin deterred the acidification of resorptive microenvironment Int.in osteoclasts.J. Mol. Sci. 2020, 21, x 6 of 19

FigureFigure 3. 3. InhibitionInhibition of induction of induction of carbonic of carbonic anhydras anhydrasee II (CAII), II (CAII), anion anionexchange exchange protein protein2 (Ae2), 2 chloride(Ae2), chloride channel channel 7 (ClC-7), 7 (ClC-7),vacuolar-type vacuolar-type H(+)-ATPase H(+)-ATPase (V-ATPase), (V-ATPase), and cathapsin and cathapsin K (A,B), Kmatrix (A,B), metalloproteinasematrix metalloproteinase 9 (MMP-9) 9 (MMP-9) secretion secretion (C), and (C bone), and resorption bone resorption (D) by ( Daesculetin.) by aesculetin. Raw 264.7 Raw 264.7cells werecells werecultured cultured in minimum in minimum essential essential medium medium alpha alpha medium medium (α-MEM) (α-MEM) with with 50 50 ng/mL ng/mL receptor receptor activatoractivator of of nuclear nuclear factor- factor-κκBB ligand ligand (RANKL) (RANKL) in in the the absence absence or or presence presence of of1–10 1–10 μMµM aesculetin aesculetin for for 5 days.5 days. Whole-cell Whole-cell lysates lysates were weresubject subject to SDS-PAGE to SDS-PAGE and Western and Western blot with blot a specific with a specificantibody antibody against CAII,against Ae2, CAII, ClC-7 Ae2, V-ATPase, ClC-7 V-ATPase, and cathapsin and cathapsin K (A,B). KAn (A equal,B). An volume equal volumeof culture of medium culture medium was subject was tosubject sodium to dodecyl sodium sulphate-polyacrylamide dodecyl sulphate-polyacrylamide gel electrophoresis gel electrophoresis (SDS-PAGE) (SDS-PAGE) and Western and blot Western with ablot specific with antibody a specific against antibody MMP-9 against (C). β MMP-9-Actin was (C). usedβ-Actin as internal was used control. as internal The bar control. graphs (mean The bar ± SEM,graphs n = (mean 3) representSEM, quantitativen = 3) represent results quantitativeof blots obtained results from of a blots densitometer. obtained fromRespective a densitometer. values in ± double-barRespective graphs values innot double-bar sharing a small graphs letter not sharingare significantly a small letter different are significantly at p < 0.05. The diff erentosteoclast at p < bone0.05. resorptionThe osteoclast was boneassayed resorption using a wascommercially assayed using available a commercially bone resorption available assay bone kit (( resorptionD), three separate assay kit experiments).((D), three separate Attached experiments). cells were Attachedremoved, cellsand wereresorption removed, pits andon the resorption plate were pits visualized on the plate under were lightvisualized microscopy. under Scale light microscopy.bar = 20 μm. Scale bar = 20 µm.

OsteoclastsOsteoclasts secrete secrete lytic lytic enzymes enzymes of of proteases proteases and and acid acid hydrolases, hydrolases, as as well well as as TRAP-degrading TRAP-degrading calcifiedcalciﬁed bone matrixmatrix inin the the resorption resorption lacunae lacunae [6, 7[6,7].]. Cathepsin Cathepsin K is K a lysosomalis a lysosomal cysteine cysteine protease protease which which plays an essential role in bone resorption [28]. RANKL enhanced the cellular expression of cathepsin K in Raw 264.7 macrophages, which was markedly attenuated by administering 10 μM aesculetin (Figure 3C). In addition, the secretion of MMP-9 was upregulated by adding RANKL to macrophages (Figure 3C). However, the administration of 10 μM aesculetin to RANKL-loaded macrophages reduced the secretion of MMP-9. In fact, MMP-9 is highly induced in osteoclasts by RANKL signals and degrades bone collagens in concert with cathepsin K in the osteoclastic microenvironment [29]. To examine the effect of aesculetin on the bone-resorbing activity of osteoclasts, Raw 264.7 cells were cultured on calcium phosphate-coated plates in the presence of 50 ng/mL RANKL and treated with 1–10 μM aesculetin. The osteoclasts created several resorption pits, but aesculetin was effective in diminishing the pit areas (Figure 3D). Collectively, aesculetin may retard bone resorption through hampering serine proteases and MMP including cathepsin K and MMP-9.

Int. J. Mol. Sci. 2020, 21, 8581 6 of 17 plays an essential role in bone resorption [28]. RANKL enhanced the cellular expression of cathepsin K in Raw 264.7 macrophages, which was markedly attenuated by administering 10 µM aesculetin (Figure3C). In addition, the secretion of MMP-9 was upregulated by adding RANKL to macrophages (Figure3C). However, the administration of 10 µM aesculetin to RANKL-loaded macrophages reduced the secretion of MMP-9. In fact, MMP-9 is highly induced in osteoclasts by RANKL signals and degrades bone collagens in concert with cathepsin K in the osteoclastic microenvironment [29]. To examine the effect of aesculetin on the bone-resorbing activity of osteoclasts, Raw 264.7 cells were cultured on calcium phosphate-coated plates in the presence of 50 ng/mL RANKL and treated with 1–10 µM aesculetin. The osteoclasts created several resorption pits, but aesculetin was effective in diminishing the pit areas (Figure3D). Collectively, aesculetin may retard bone resorption through hampering serine proteases and MMP including cathepsin K and MMP-9. Int. J. Mol. Sci. 2020, 21, x 7 of 19 2.3. Inhibitory Effects of Aesculetin on Actin Ring Formation The2.3. Inhibitory actin ring Effects formation of Aesculetin of on osteoclasts Actin Ring Formation is a critical prerequisite for optimal osteoclastic resorptionThe [30 actin]. The ring formation formation of of actin osteoclasts rings was is a elevatedcritical prerequisite in RANKL-treated for optimal macrophages, osteoclastic resorption as evidenced by red[30]. staining The formation of rhodamine of actin rings phalloidin was elevated (Figure in4 RANKL-treatedA). When 5 µmacrophages,M aesculetin as wasevidenced administered by red to ≥ RANKL-loadedstaining of rhodamine macrophages, phalloidin the formation (Figure 4A). of When actin ≥ rings 5 μM wasaesculetin highly was attenuated. administered to RANKL- loaded macrophages, the formation of actin rings was highly attenuated.

Figure 4. Blockade of receptor activator of nuclear factor-κB ligand (RANKL)-induced actin ring Figure 4. Blockade of receptor activator of nuclear factor-κB ligand (RANKL)-induced actin ring formation (A) and induction of integrins, cluster of differentiation 44 (CD44) and Cdc42 (B,C) in formation (A) and induction of integrins, cluster of differentiation 44 (CD44) and Cdc42 (B,C) in mature osteoclasts. Raw 264.7 cells were cultured for 5 days with 50 ng/mL RANKL in the absence mature osteoclasts. Raw 264.7 cells were cultured for 5 days with 50 ng/mL RANKL in the absence or µ or presencepresence of of 1–10 1–10 μMM aesculetin. Differentiated Differentiated cells cells were were fixed fixedin 4% informaldehyde 4% formaldehyde for 10 min, for and 10 min, and fluorescentfluorescent rhodamine rhodamine phalloidin phalloidin was wasadded added to fixed to cells fixed and cells hence and incubated hence incubatedfor 20 min at for 4 °C 20 (A min). at 4 ◦C(FluorescentA). Fluorescent images images were were taken taken with with an Axiomager an Axiomager Optical Optical fluorescence fluorescence microscope. microscope. Original Original magnificationmagnification of microscopic of microscopic images images (n (=n =3). 3). Scale Scale barbar = 2020 μµm.m. After After whole-cell whole-cell extracts extracts were were subject subject to sodiumto sodium dodecyl dodecyl sulphate-polyacrylamide sulphate-polyacrylamide gel electrophoresisgel electrophoresis (SDS-PAGE) (SDS-PAGE) and and Western Western blot blot analysis with aanalysis primary with antibody a primary against antibodyαv integrin, againstβ3 integrin,αv integrin, CD44, β and3 integrin, Cdc42 (CD44,B,C). Representative and Cdc42 (B blot,C). data were obtainedRepresentative from blot three data independent were obtained experiments, from three independent and β-actin experiments, protein was and used β-actin as an protein internal was control. The barused graphs as an internal (mean control.SEM) The represent bar graphs quantitative (mean ± SEM) results represent obtained quantitative from a densitometer. results obtained Respective from a densitometer. Respective± values in different-colored bar graphs not sharing a small letter indicate a values in different-colored bar graphs not sharing a small letter indicate a significant difference at p < 0.05. significant difference at p < 0.05. The αvβ3 integrin abundantly expressed in osteoclasts is localized in the sealing zone, known as The αvβ3 integrin abundantly expressed in osteoclasts is localized in the sealing zone, known a specializedas a specialized cell–matrix cell–matrix adhesion adhesion structure structure [31 [3,321,32].]. ThisThis study determined determined whether whether aesculetin aesculetin suppressed the induction of αvβ3 integrin at the podosome cloud of RANKL-exposed macrophages. The induction of αv integrin and β3 integrin was markedly promoted by administering 50 ng/mL RANKL to Raw 264.7 cells (Figure 4B). In contrast, ≥5 μM aesculetin markedly inhibited the induction of both integrin proteins. On the other hand, the cluster of differentiation 44 (CD44) is localized in the sealing zone of osteoclasts and involved in actin cytoskeletal organization by linking to podosome cores [32,33]. RANKL enhanced the induction of core-linking CD44, but aesculetin did not influence the cellular CD44 induction of osteoclasts (Figure 4B). Consequently, aesculetin may suppress formation of the sealing zone required for effective osteoclastic bone resorption through inhibiting the induction of diffuse cloud-associated αvβ3 integrin. Furthermore, the small GTPase Cdc42 regulates assembly and organization of podosomes into the sealing zone of osteoclasts [34]. The

Int. J. Mol. Sci. 2020, 21, 8581 7 of 17 suppressed the induction of αvβ3 integrin at the podosome cloud of RANKL-exposed macrophages. The induction of αv integrin and β3 integrin was markedly promoted by administering 50 ng/mL RANKL to Raw 264.7 cells (Figure4B). In contrast, 5 µM aesculetin markedly inhibited the induction ≥ of both integrin proteins. On the other hand, the cluster of differentiation 44 (CD44) is localized in the sealing zone of osteoclasts and involved in actin cytoskeletal organization by linking to podosome cores [32,33]. RANKL enhanced the induction of core-linking CD44, but aesculetin did not influence the cellular CD44 induction of osteoclasts (Figure4B). Consequently, aesculetin may suppress formation of the sealing zone required for effective osteoclastic bone resorption through inhibiting the induction of diffuse cloud-associated αvβ3 integrin. Furthermore, the small GTPase Cdc42 regulates assembly and organization of podosomes into the sealing zone of osteoclasts [34]. The cellular Cdc42 induction was highlyInt. enhancedJ. Mol. Sci. 2020 in, 21 RANKL-loaded, x Raw 264.7 cells, while 5 µM aesculetin attenuated its8 induction of 19 ≥ (Figure4C). cellular Cdc42 induction was highly enhanced in RANKL-loaded Raw 264.7 cells, while ≥ 5 μM 2.4. Disruptionaesculetin attenuated of Osteoclastic its induction Cytoskeletal (Figure Arrangement 4C). by Aesculetin 2.4.Microtubules, Disruption of Osteoclastic one of the Cyto elementsskeletal Arrangement of cytoskeletons, by Aesculetin act as a transport system that delivers vesicles andMicrotubules, lysosomes one to of the the ru elementsffled border, of cytoskelet illustratingons, act the as importance a transport ofsystem microtubule that delivers networks in osteoclastvesicles and function lysosomes [20 to,35 the]. ruff Theled formationborder, illustrating of actin the rings import andance microtubule of microtubule cytoskeleton networks in was elevatedosteoclast in RANKL-treated function [20,35]. macrophages, The formation asof evidencedactin rings and by immunocytochemicalmicrotubule cytoskeleton staining was elevated with green FITC-conjugatedin RANKL-treatedα-tubulin macrophages, (Figure 5as). evidenced When 5 byµ Mimmunocytochemical aesculetin was administered staining with to green RANKL-loaded FITC- ≥ macrophages,conjugated theα-tubulin microtubule (Figure 5). cytoskeleton When ≥ 5 μ withinM aesculetin actin was rings administered almost completely to RANKL-loaded disappeared. Accordingly,macrophages, aesculetin the microtubule disturbed cyto theskeleton microtubule within cytoskeletonactin rings almost involved completely in lysosomal disappeared. transport withinAccordingly, osteoclasts, aesculetin leading todisturbed osteoclastic the microtubul bone resorption.e cytoskeleton involved in lysosomal transport within osteoclasts, leading to osteoclastic bone resorption.

FigureFigure 5. 5.Suppressive Suppressive effects effects of aesculetin of aesculetin on receptor on receptor activator of activator nuclear factor- of nuclearκB ligand factor- (RANKL)-κB ligand (RANKL)-inducedinduced microtubular microtubular formation formation in actin inring-bearing actin ring-bearing osteoclasts. osteoclasts. Raw 264.7 Raw macrophages 264.7 macrophages were werecultured cultured in inminimum minimum essential essential medium medium alpha medium alpha medium (α-MEM) ( αwith-MEM) 50 ng/mL with RANKL 50 ng/mL in the RANKL in theabsence absence and andpresence presence of 1–10 of μM 1–10 aesculetinµM aesculetin for 5 days. for RANKL-differentiated 5 days. RANKL-di cellsfferentiated were fixed in cells 4% were fixedparaformaldehyde in 4% paraformaldehyde for 10 min, and for immunocytoch 10 min, andemical immunocytochemical analysis was conducted analysis with was a primary conducted withantibody a primary of α antibody-tubulin and of αsecondary-tubulin antibody and secondary of mouse antibody immunoglobulin of mouse G (IgG) immunoglobulin conjugated with G (IgG) conjugatedfluorescein with isothiocyanate fluorescein isothiocyanate (FITC). Subsequent (FITC).ly, Subsequently, nuclear counterstaining nuclear counterstaining was done with was done4′,6- with diamidino-2-phenylindole (DAPI). Fluorescent images were taken with a fluorescence microscope. 40,6-diamidino-2-phenylindole (DAPI). Fluorescent images were taken with a fluorescence microscope. Original magnification of microscopic images (n = 3). Scale bar = 20 μm. Original magnification of microscopic images (n = 3). Scale bar = 20 µm. 2.5. Blockade of Lysosome Positioning to Microtubules by Aesculetin Bone resorption by osteoclasts depends on vacuoles including lysosomes filled with acid phosphatase [20]. How secretory lysosomes are directed to fuse with the ruffled border is still enigmatic. Osteoclast SytVII, a lysosome-associated calcium sensor protein that regulates exocytosis, regulates transport and secretion of matrix-degrading molecules into the resorptive lacuna [21]. This study found that RANKL prompted the SytVII induction in Raw 264.7 macrophages and that such induction was attenuated in 10 μM aesculetin-supplied osteoclasts (Figure 6A). LAMP2 colocalizes with cathepsin K and Syt VII and is involved in the process of vesicular trafficking [21,36]. RANKL

Int. J. Mol. Sci. 2020, 21, 8581 8 of 17

2.5. Blockade of Lysosome Positioning to Microtubules by Aesculetin Bone resorption by osteoclasts depends on vacuoles including lysosomes filled with acid phosphatase [20]. How secretory lysosomes are directed to fuse with the ruffled border is still enigmatic. Osteoclast SytVII, a lysosome-associated calcium sensor protein that regulates exocytosis, regulates transport and secretion of matrix-degrading molecules into the resorptive lacuna [21]. This study found that RANKL prompted the SytVII induction in Raw 264.7 macrophages and that such induction was attenuated in 10 µM aesculetin-supplied osteoclasts (Figure6A). LAMP2 colocalizes with cathepsin Int.Kand J. Mol. Syt Sci. VII 2020 and, 21, x is involved in the process of vesicular trafficking [21,36]. RANKL stimulated9 of the 19 LAMP2 induction in Raw 264.7 macrophages (Figure6A). When 1–10 µM aesculetin was administered stimulated the LAMP2 induction in Raw 264.7 macrophages (Figure 6A). When 1–10 μM aesculetin to RANKL-induced cells, the LAMP2 induction was reduced. was administered to RANKL-induced cells, the LAMP2 induction was reduced.

Figure 6. Western blot data showing inhibition of induction of synaptotagmin VII (SytVII), Figurelysosome-associated 6. Western blot membrane data showing protein inhibition 2 (LAMP2), of induction lissencephaly-1 of synaptotagmin (LIS1), and VII Pleckstrin (SytVII), lysosome- homology associateddomain-containing membrane protein protein family 2 (LAMP2), member lissencep 1 (PLEKHM1)haly-1 (LIS1), (A,B). and Raw Pleckstrin 264.7 cells homology were cultured domain- in containingminimum essentialprotein family medium member alpha 1 medium(PLEKHM1) (α-MEM) (A,B). withRaw 50264.7 ng/ mLcells receptor were cultured activator in ofminimum nuclear essentialfactor-κB medium ligand (RANKL) alpha medium in the absence(α-MEM) or with presence 50 ng/mL of 1–10 receptorµM aesculetin activator for of 5 nuclear days. Whole-cell factor-κB ligandlysates (RANKL) were subject in the to sodiumabsence dodecyl or presence sulphate-polyacrylamide of 1–10 μM aesculetin gel for electrophoresis 5 days. Whole-cell (SDS-PAGE) lysates were and subjectWestern to blot sodium with adodecyl specific antibodysulphate-polyacrylamide against SytVII, LAMP2, gel electrophoresis LIS1, and PLEKHM1. (SDS-PAGE)β-Actin and wasWestern used blotas internal with a control. specific The antibody bar graphs agains (meant SytVII,SEM, LAMP2,n = 3) representLIS1, and quantitative PLEKHM1. results β-Actin of blotswas obtainedused as ± internalfrom a densitometer. control. The bar Respective graphs values(mean in± SEM, double-bar n = 3) graphsrepresent not quantitative sharing a small results letter of are blots significantly obtained fromdifferent a densitometer. at p < 0.05. Respective values in double-bar graphs not sharing a small letter are significantly different at p < 0.05. The process of microtubule organization and lysosomal secretion requires the interaction between microtubuleThe process regulator of microtubule LIS1 and lysosome-associated organization and PLEKHM1 lysosomal [22 secretion,37]. This requires study found the thatinteraction cellular betweeninduction microtubule of LIS1 and regulator PLEKHM1 LIS1 was and enhanced lysosome by-associated an exposure PLEKHM1 of RANKL [22,37]. (Figure This6B). study However, found thatadministering cellular induction5 µM aesculetinof LIS1 and to PLEKHM1 RANKL-challenged was enhanced macrophages by an exposure inhibited of theRANKL induction (Figure of LIS16B). ≥ However,and PLEKHM1. administering Thus, aesculetin ≥ 5 μM may aesculetin lead to disorganized to RANKL-challenged microtubules macrophages and abrogate theinhibited peripheral the inductiondistribution of ofLIS1 lysosomes and PLEKHM1. in osteoclasts. Thus, aesculetin may lead to disorganized microtubules and abrogate the peripheral distribution of lysosomes in osteoclasts.

2.6. Disruption of Trafficking of Lysosomes into Ruffled Border by Aesculetin This study conducted a double-staining with green FITC-conjugated anti-α-tubulin and red lysoTracker dye to detect colocalization of microtubules and lysosomes. RANKL created multinucleated giant cells with a solid microtubule cytoskeleton and heavy reddish LysoTracker fluorescence (Figure 7). However, the red staining with lysoTracker was reduced in RANKL-exposed and 10 μM aesculetin-treated cells, indicating a decrease in lysosomal flux of osteoclasts with mature cytoskeleton.

Int. J. Mol. Sci. 2020, 21, 8581 9 of 17

2.6. Disruption of Trafficking of Lysosomes into Ruffled Border by Aesculetin This study conducted a double-staining with green FITC-conjugated anti-α-tubulin and red lysoTracker dye to detect colocalization of microtubules and lysosomes. RANKL created multinucleated giant cells with a solid microtubule cytoskeleton and heavy reddish LysoTracker fluorescence (Figure7). However, the red staining with lysoTracker was reduced in RANKL-exposed and 10 µM Int.aesculetin-treated J. Mol. Sci. 2020, 21, cells,x indicating a decrease in lysosomal flux of osteoclasts with mature cytoskeleton.10 of 19

FigureFigure 7. 7. BlockadeBlockade of ofpositioning positioning of oflysosomes lysosomes and and microtubules microtubules in actin in actin ring-bearing ring-bearing osteoclasts. osteoclasts. Raw 264.7Raw cells 264.7 were cells cultured were cultured for 5 days for with 5 days 50 withng/mL 50 receptor ng/mL receptoractivator activatorof nuclear of factor- nuclearκB ligand factor- (RANKL)κB ligand in(RANKL) the absence in the or absence presence or presenceof 1–10 ofμM 1–10 aesculetin.µM aesculetin. RANKL-differentiated RANKL-differentiated cells were cells werefixed fixedin 4% in paraformaldehyde4% paraformaldehyde for 10 for min, 10 min,followed followed by immunocytochemical by immunocytochemical analys analysisis with a with primary a primary antibody antibody of α- tubulinof α-tubulin and andanti-mouse anti-mouse IgG IgGconjugated conjugated with with fluorescein fluorescein isothiocyanate isothiocyanate (FITC). (FITC). Subsequently, Subsequently, commerciallycommercially available available red red fluorescent fluorescent lysoTracker lysoTracker Red Red DND-99 DND-99 dye dye was was employed employed for forthe the staining staining of lysosomes.of lysosomes. Finally, Finally, cells cellswere werecounterstained counterstained with 4 with′ ,6-diamidino-2-phenylindole 40,6-diamidino-2-phenylindole (DAPI) (DAPI)for nuclei. for Representativenuclei. Representative fluorescent fluorescent images were images obtained were with obtained a fluorescence with microscope a fluorescence (n = 3). microscope Scale bar = (20n μ=m.3). Scale bar = 20 µm. In osteoclasts, the formation of the ruffled border involves vesicular trafficking pathways that In osteoclasts, the formation of the ruffled border involves vesicular trafficking pathways that are regulated by the Rab family of GTPase, in particular lysosomal Rab7 [34]. The Rab7 induction by are regulated by the Rab family of GTPase, in particular lysosomal Rab7 [34]. The Rab7 induction by RANKL was diminished in ≥ 5 μM aesculetin-exposed mature osteoclasts (Figure 8A). The RANKL was diminished in 5 µM aesculetin-exposed mature osteoclasts (Figure8A). The conjugation conjugation and lipidation of≥ LC3 protein are responsible for fusion of secretory lysosomes into bone- and lipidation of LC3 protein are responsible for fusion of secretory lysosomes into bone-apposed apposed plasma membrane in order to form dense folds of the ruffled border [24]. The conversion of plasma membrane in order to form dense folds of the ruffled border [24]. The conversion of soluble soluble LC3I to lipid-bound LC3II was enhanced in osteoclasts differentiated by RANKL (Figure 8B). LC3I to lipid-bound LC3II was enhanced in osteoclasts differentiated by RANKL (Figure8B). In contrast, In contrast, the LC3II induction was reduced by administering 10 μM aesculetin. The LC3II the LC3II induction was reduced by administering 10 µM aesculetin. The LC3II localization onto localization onto the ruffled border involves aegis of the Atg12–Atg5 conjugate and Atg7 [24]. the ruffled border involves aegis of the Atg12–Atg5 conjugate and Atg7 [24]. RANKL promoted RANKL promoted the induction of Atg5 and Atg7, which was blunted by adding 10 μM aesculetin the induction of Atg5 and Atg7, which was blunted by adding 10 µM aesculetin during osteoclast during osteoclast differentiation (Figure 8C). Thus, aesculetin may block the formation of the putative ruffled border by impairing lysosomal trafficking and protease exocytosis.

Int. J. Mol. Sci. 2020, 21, 8581 10 of 17 differentiation (Figure8C). Thus, aesculetin may block the formation of the putative ru ffled border by Int. J. Mol. Sci. 2020, 21, x 11 of 19 impairing lysosomal trafficking and protease exocytosis.

Figure 8. Western blot data showing inhibition of induction of Rab7 (A), microtubule-associated protein lightFigure chain 8. Western 3 (LC3) (Bblot), Atg5, data and showing Atg7 ( Cinhibition) by aesculetin. of induction Raw 264.7 of cellsRab7 were(A), culturedmicrotubule-associated for 5 days with 50protein ng/mL light RANKL chain in3 (LC3) the absence (B), Atg5, or presence and Atg7 of ( 1–10C) byµ aesculetin.M aesculetin. Raw Whole-cell 264.7 cells lysates were cultured were subject for 5 todays SDS-PAGE with 50 andng/mL Western RANKL blot in with the aabsence primary or antibody presence against of 1–10 Rab7, μM LC3,aesculetin. Atg5, andWhole-cell Atg7. β lysates-Actin waswere used subject as an to internalSDS-PAGE control. and TheWestern bar graphs blot with (mean a primarySEM, antibodyn = 3) represent against quantitativeRab7, LC3, Atg5, results and of ± blotsAtg7. obtained β-Actin fromwas aused densitometer. as an internal Respective control. values The bar in double-bar graphs (mean graphs ± SEM, not sharing n = 3) a represent letter are significantlyquantitative diresultsfferent of at blotsp < 0.05. obtained from a densitometer. Respective values in double-bar graphs not sharing a letter are significantly different at p < 0.05. 3. Materials and Methods 3. Materials and Methods 3.1. Materials 3.1. MaterialsMinimum essential medium alpha medium (α-MEM), Dulbecco’s modified eagle’s media (DMEM), aesculetin,Minimum and aesculinessential were medium purchased alpha from medium Sigma-Aldrich (α-MEM), Chemicals Dulbecco’s (St. Louis,modified MO, eagle’s USA), as media were all(DMEM), other reagents, aesculetin, unless and otherwise aesculin were specified. purchased Fetal bovinefrom Sigma-Aldrich serum (FBS) and Chemicals penicillin–streptomycin (St. Louis, MO, wereUSA), obtained as were from all other BioWhittaker reagents, (Sanunless Diego, otherwise CA, USA). specified. 3-(4,5-Dimetylthiazol-yl)-diphenyl Fetal bovine serum (FBS) and tetrazolium penicillin– bromidestreptomycin (MTT) were was obtained supplied from from BioWhittaker DUCHEFABiochemie (San Diego, (Haarlem, CA, USA). Netherlands). 3-(4,5-Dimetylthiazol-yl)- RANKL was obtaineddiphenyl fromtetrazolium PeproTech bromide (Rocky Hill,(MTT) NJ, was USA). su Antibodiespplied from of mouseDUCHEFA V-ATPase Biochemie (cat. no. (Haarlem, sc-69108), mouseNetherlands). cathepsin RANKL K (cat. was no. obtained sc-48353), from mouse PeproTech LIS1 (cat. (Rocky no. sc-374586),Hill, NJ, USA). mouse Antibodies integrin ofαV mouse (cat. no.V- sc-6617-R),ATPase (cat. mouse no. sc-69108), Cdc42 (cat. mouse no. sc-8401), cathepsin and K mouse (cat. no.α-tubulin sc-48353), (cat. mouse no. sc-8035) LIS1 (cat. were no. obtained sc-374586), from Santamouse Cruz integrin Biotechnology αV (cat. no. (Dallas, sc-6617-R), TX, USA). mouse Antibodies Cdc42 (cat. of mouseno. sc-8401), CAII (cat. and no.mouse MAB2184) α-tubulin and (cat. mouse no. MMP-9sc-8035) (cat.were no. obtained MAB936) from were Santa provided Cruz Biotechnolo by R&D Systemsgy (Dallas, (Minneapolis, TX, USA). An MN,tibodies USA). of Antibodies mouse CAII of mouse(cat. no. Rab7 MAB2184) (cat. no. ab137029),and mouse mouse MMP-9 ClC-7 (cat. (cat. no. ab86196),MAB936) and were mouse provided Syt VII (cat.by R&D no. ab174633) Systems were(Minneapolis, provided MN, by Abcam USA). (Cambridge, Antibodies UK).of mouse Antibodies Rab7 (cat. of mouse no. ab137029), Atg7 (cat. no.mouse 4445), ClC-7 mouse (cat. Atg5 no. (cat.ab86196), no. 4445), and mousemouse CD44 Syt VII (cat. (cat. no. 37259),no. ab174633) and mouse were integrin providedβ3 (cat. by no.Abcam 4702) (Cambridge, were supplied UK). by CellAntibodies Signaling of Technologymouse Atg7 (Beverly, (cat. no. MA,4445), USA). mouse Human Atg5 LC3(cat. antibodyno. 4445), (cat. mouse no. M152-3)CD44 (cat. was no. obtained 37259), fromand mouse MBL International integrin β3 (Woburn,(cat. no. MA),4702) andwere mouse supplied Ae2 (cat.by Cell no. NBP1-59858)Signaling Technology antibody was(Beverly, purchased MA, fromUSA). Novus Human Biologicals LC3 antibody (Littleton, (cat. CO,no. M152-3) USA). Antibodies was obtained of mouse from MBL PLEKHM1 International (cat. no. (Woburn, bs-8062R) MA), and LAMP2and mouse (cat. Ae2 no. (cat. bs-2379R) no. NBP1-59858) were provided antibody by Bioss was Antibodiespurchased (Woburn,from Novus MA, Biologicals USA). Horseradish (Littleton, peroxidaseCO, USA). (HRP)-conjugatedAntibodies of mouse goat PLEKHM1 anti-rabbit, (cat. goat no anti-mouse,. bs-8062R) and and donkey LAMP2 anti-goat (cat. no. immunoglobulin bs-2379R) were Gprovided (IgG) were by Bioss provided Antibodies by Jackson (Woburn, Immuno MA, Research USA). Horseradish Laboratories peroxidase (West Grove, (HRP)-conjugated PA, USA). goat anti-rabbit, goat anti-mouse, and donkey anti-goat immunoglobulin G (IgG) were provided by Jackson Immuno Research Laboratories (West Grove, PA, USA). Aesculetin and aesculin were dissolved in dimethyl sulfoxide (DMSO) for live culture with cells; the final culture concentration of DMSO was < 0.5%.

Int. J. Mol. Sci. 2020, 21, 8581 11 of 17

Aesculetin and aesculin were dissolved in dimethyl sulfoxide (DMSO) for live culture with cells; the ﬁnal culture concentration of DMSO was < 0.5%.

3.2. Osteoclast Differentiation of Raw 264.7 Cells Murine macrophage Raw 264.7 cells (American Type Culture Collection, Manassas, VA, USA) were cultured in DMEM containing 10% FBS, 2 mM glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 ◦C in a humidified atmosphere of 5% CO2 in air. For osteoclast differentiation, Raw 264.7 cells were plated on 24-well plates at the density of 1 104 cells/mL and cultured for × 5 days in α-MEM containing 10% FBS and 50 ng/mL RANKL in the absence and presence of 1–15 µM aesculetin or 1–40 µM aesculin. Cell culture medium was newly changed every 2 days. The cytotoxicity of aesculetin and aesculin was determined using an MTT assay. Cells were treated with 1 mg/mL MTT solution and incubated at 37 ◦C for 3 h, resulting in the formation of an insoluble purple formazan product that was dissolved in 250 µL of isopropanol. Optical density was measured using a microplate reader at λ = 570 nm. Raw 264.7 cell viability was not influenced by administering 10 µM aesculetin and 40 µM aesculin in the absence (Figures1B and2B) and presence of 50 ng /mL ≤ ≤ RANKL (Figures1C and2C).

3.3. Measurement of Tartrate-Resistant Acid Phosphatase (TRAP) Activity Cells were fixed with 4% formaldehyde for 10 min and then were incubated in 50 mM citrate buffer (50 mM citric acid and 50 mM sodium citrate (pH 4.5)) containing 5 mM 4-nitrophenylphosphate, and 10 mM sodium tartrate for 1 h. The reaction was terminated by adding 0.1 N NaOH. Absorption intensity was measured using a microplate reader at λ = 405 nm. Furthermore, cells were fixed with 4% formaldehyde and stained for 30 min with a commercially available TRAP kit (Sigma-Aldrich Chemical, St. Louis, MO, USA). TRAP-positive multinucleated osteoclasts were visualized under light microscopy (ECLIPSE Ni-U, Nikon, Tokyo, Japan).

3.4. Western Blot Analysis Cell lysates were prepared from Raw 264.7 cells differentiated with RANKL. Equal amounts of lysate proteins or equal volumes of culture media were electrophoresed on 6–20% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and transferred onto a nitrocellulose membrane. Nonspecific binding was blocked by soaking membranes in a Tris-buffered saline/Tween-20 (TBS-T) buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 0.1% Tween-20) containing 3% bovine serum albumin or 5% nonfat milk for 3 h. The membranes were incubated with a primary antibody against CAII, V-ATPase, ClC-7, cathepsin K, Rab7, Atg5, Atg7, Syt VII, Ae2, CD44, integrin αV, integrin β3, MMP-9, LC3, PLEKHM1, LAMP2, Cdc42, or LIS1. The membranes were then incubated with goat anti-rabbit, goat anti-mouse, or donkey anti-goat IgG conjugated to HRP as a secondary antibody. The protein levels on gels were measured by using ECL chemiluminescent detection reagents (Millipore, Billerica, MA, USA) and Konica X-ray film (Konica, Tokyo, Japan). Incubation with β-actin antibody was conducted for comparative control. The bone resorption assay was performed using a resorption assay kit (CSR-BRA-48X2KIT, Cosmo Bio Co., Tokyo, Japan). Raw 264.7 cells were seeded on calcium phosphate-coated well plates at a density of 1 104 cells/mL and cultured for 5 days in phenol red-free α-MEM containing 10% FBS × and 50 ng/mL RANKL in the absence and presence of 1–10 µM aesculetin. To measure resorbed pit areas, the cells were washed in 5% NaOCl. The resorption pits developed on the plate were visualized under light microscopy (ECLIPSE TS100, Nikon, Tokyo, Japan).

3.5. Actin Ring Staining The formation of actin rings was determined by staining actin filaments with rhodamine phalloidin (Life Technologies, Carlsbad, CA, USA). RANKL-differentiated Raw 264.7 cells were fixed with 4% formaldehyde for 10 min and washed with prewarmed phosphate-buffered saline (PBS). Cells were Int. J. Mol. Sci. 2020, 21, 8581 12 of 17 incubated for 30 min with 10 units of rhodamine phalloidin. Fluorescent images were taken with a fluorescence microscope (Carl Zeiss, Oberkochen, Germany).

3.6. Immunochemical Staining Raw 264.7 cells grown on glass coverslips were differentiated for 5 days in α-MEM containing 10% FBS and 50 ng/mL RANKL in the absence and presence of 1–10 µM aesculetin. After permeabilizing on ice with 0.1% Triton X-100 and 0.1% sodium citrate for 1 min, α-tubulin as a primary antibody was added and incubated overnight at 4 ◦C. After several washes with PBS/Tween-20, cells were incubated with anti-mouse IgG conjugated with a green-fluorescent dye fluorescein isothiocyanate (FITC). Following washes, nuclear counterstaining was conducted using 40,6-diamidino-2-phenylindole (DAPI) for 15 min. Another experiment was performed for the detection of acidic lysosomes. Cells were incubated for 1 h with 50 nM acidotropic probe LysoTracker Red DND-99 (Life Technologies, Carlsbad, CA, USA). Subsequently, cells were fixed with 4% ice-cold formaldehyde and blocked nonspecific binding with 20% FBS in PBS for 1 h. Immunocytochemical analysis was further conducted for the staining of FITC-conjugated α-tubulin. Each slide was mounted in a mounting medium and images of each slide were taken using an optical fluorescence microscope (Zeiss, Oberkochen, Germany).

3.7. Statistical Analysis The results were presented as means SEM for each treatment group. Statistical analyses were ± performed using Statistical Analysis Systems statistical software package (SAS Institute Inc., Cary, NC, USA). Significance was determined by one-way ANOVA, followed by Duncan’s range test for multiple comparisons. Differences were considered significant at p < 0.05.

4. Discussion Seven major findings were extracted from this study. (1) Nontoxic aesculetin but not aesculin at 10 µM showed a marked inhibition of the formation of TRAP-positive osteoclasts. (2) RANKL ≤ promoted the induction of CAII, Ae2, ClC-7, and V-ATPase in Raw 264.7 macrophages, which was deterred by aesculetin. (3) The presence of aesculetin attenuated the cathepsin K expression and MMP-9 secretion of osteoclasts in parallel with its reduction of the resorption pits and the actin ring formation up-regulated by RANKL. (4) When Raw 264.7 cells were treated with 50 ng/mL RANKL, aesculetin was effective in reducing the induction of αv integrin and β3 integrin but not core-linking CD44. (5) The induction of Cdc42, Atg5, Atg7, SytVII, and LAMP-2 was diminished by administering 10 µM aesculetin during osteoclast differentiation, indicating that aesculetin may suppress formation of lysosomes and ruffled border crucial for bone resorption. (6) Aesculetin inhibited the induction of LIS1 and PLEKHM1 in RANKL-exposed osteoclasts, suggesting that aesculetin blocked lysosome positioning to microtubules involved in the lysosomal transport within osteoclasts. (7) The induction of LC3II and Rab7 by RANKL was diminished in 5 µM aesculetin-exposed mature osteoclasts. ≥ Thus, aesculetin may block the formation of the putative ruffled border through impairing lysosomal trafficking and protease exocytosis (Figure9). Osteoclasts are giant multinucleated hematopoietic cells specialized for bone resorption [7,8]. Bone remodeling comprises finely coordinated steps of activation, resorption, reversal, formation, and termination via activation of tightly coupled signaling of RANK/RANKL/OPG [4,5,9]. OPG is secreted by osteoblasts and osteogenic stromal stem cells and protects the skeleton from excessive bone resorption through binding to RANKL and hampering interaction with RANK [38]. On the other hand, the remodeling is regulated by signaling of various calcitropic hormones, proinflammatory cytokines, and growth factors [1,5,39]. Signaling of transforming growth factor-β and parathyroid hormone is known to control bone remodeling and maintenance [40]. In addition, the Wnt signaling pathway manipulates bone formation and remodeling [41,42]. Mechanistic signaling pathways involved in bone remodeling can be targets for pharmacological interventions for antiresorptive and bone-forming Int. J. Mol. Sci. 2020, 21, 8581 13 of 17 therapies [10–12]. In particular, the main classes of antiresorptives currently in use are calcium, bisphosphonates, estrogen, selective estrogen receptor modulators, and calcitonin [43]. However, theInt. J. mechanistic Mol. Sci. 2020, e21ffi, xcacy and safety of such therapeutic interventions remain to be confirmed. 14 of 19

FigureFigure 9.9. Schematic diagram showing the inhibitory eeffectffectss of aesculetin on ruruffledffled borderborder formationformation andand lysosomallysosomal tratraffickingfficking responsibleresponsible forfor bonebone resorption.resorption. AsAs depicted, aesculetin blocksblocks thethe RANKL signalingsignaling processprocess responsible responsible for bonefor bone resorption. resorption. The symbol The symbol I− indicates Ͱ indicates sites of inhibition sites of manifestedinhibition bymanifested aesculetin. by aesculetin. The ruffled border of osteoclasts in contact with the bone surface is formed by massive trafficking Osteoclasts are giant multinucleated hematopoietic cells specialized for bone resorption [7,8]. of secretory lysosomes [14,15]. Bone resorption by osteoclasts highly depends on exocytosis of Bone remodeling comprises finely coordinated steps of activation, resorption, reversal, formation, lysosomes full of proteases, acid hydrolases, and acid phosphatases [15]. Considerable efforts have and termination via activation of tightly coupled signaling of RANK/RANKL/OPG [4,5,9]. OPG is been made in seeking naturally occurring compounds as novel therapeutic agents targeting against secreted by osteoblasts and osteogenic stromal stem cells and protects the skeleton from excessive bone loss [25,26,44]. One study shows that aesculetin inhibits osteoclast differentiation through bone resorption through binding to RANKL and hampering interaction with RANK [38]. On the suppressing expression of c-Fos and nuclear factor of activated T cell c1 [45]. This study found that other hand, the remodeling is regulated by signaling of various calcitropic hormones, aesculetin abrogated the intracellular induction of TRAP, cathepsin K, and MMP-9 from RANKL-loaded proinflammatory cytokines, and growth factors [1,5,39]. Signaling of transforming growth factor-β macrophages. Osteoclastic activity resorbing the mineralized bone matrix entails acidification of and parathyroid hormone is known to control bone remodeling and maintenance [40]. In addition, resorption lacuna generated by contact of osteoclasts with the bone surface [15,16]. Previous studies the Wnt signaling pathway manipulates bone formation and remodeling [41,42]. Mechanistic showed that phloretin and fisetin blunted bone resorption through inhibition of acidification of signaling pathways involved in bone remodeling can be targets for pharmacological interventions resorption lacuna of TRAP-positive osteoclasts [46,47]. Nontoxic aesculetin but not aesculin inhibited for antiresorptive and bone-forming therapies [10–12]. In particular, the main classes of RANKL induction of CAII, Ae2, ClC-7, and V-ATPase necessary for the extracellular acidification in antiresorptives currently in use are calcium, bisphosphonates, estrogen, selective estrogen receptor osteoclasts. It should be noted that the glucose moiety of aesculin interfered with the inhibitory effects modulators, and calcitonin [43]. However, the mechanistic efficacy and safety of such therapeutic on RANKL-induced osteoclast differentiation. Accordingly, the defect of bone-resorbing proteases and interventions remain to be confirmed. proton providers by aesculetin failed to digest organic components of bone matrices. Furthermore, The ruffled border of osteoclasts in contact with the bone surface is formed by massive the bone-resorbing activity of osteoclasts counts on a tight integrin-based adhesion to the bone surface, trafficking of secretory lysosomes [14,15]. Bone resorption by osteoclasts highly depends on forming an actin-rich sealing zone [48]. During the bone resorption of RANKL-loaded macrophages, exocytosis of lysosomes full of proteases, acid hydrolases, and acid phosphatases [15]. Considerable aesculetin suppressed the formation of podosome belts (actin rings), through deterring the induction efforts have been made in seeking naturally occurring compounds as novel therapeutic agents of αv integrin and β3 integrin but not podosome core-linking CD44. Small GTPases of Rac and Cdc42 targeting against bone loss [25,26,44]. One study shows that aesculetin inhibits osteoclast differentiation are crucial for actin polymerization, podosome organization, and the assembly of the sealing zone in through suppressing expression of c-Fos and nuclear factor of activated T cell c1 [45]. This study found that aesculetin abrogated the intracellular induction of TRAP, cathepsin K, and MMP-9 from RANKL- loaded macrophages. Osteoclastic activity resorbing the mineralized bone matrix entails acidification of resorption lacuna generated by contact of osteoclasts with the bone surface [15,16]. Previous studies showed that phloretin and fisetin blunted bone resorption through inhibition of acidification of resorption lacuna of TRAP-positive osteoclasts [46,47]. Nontoxic aesculetin but not aesculin inhibited RANKL induction of CAII, Ae2, ClC-7, and V-ATPase necessary for the extracellular acidification in

Int. J. Mol. Sci. 2020, 21, 8581 14 of 17 coordination with integrin signaling [34]. Aesculetin retarded the Cdc42 induction in RANKL-loaded macrophages. Osteoclasts generated from Cdc42-knockout mice reduce actin ring formation and bone resorption [49]. It should be noted that aesculetin inhibited the formation of the microtubule cytoskeleton within the actin ring but not the organization of the microtubule architecture. Emerging evidence suggests that several membrane trafficking pathways including endocytic, secretory, transcytotic, and autophagic pathways are involved in operating the subdomains of ruffled border in osteoclasts [13,15]. The means by which secretory lysosomes directly fuse with the ruffled border are elusive. The autophagic proteins of the Atg12–Atg5 conjugate, Atg7, Atg4B, and LC3 are involved in fusion of secretory lysosomes with the ruffled border, promoting bone resorption activity [23,24,50]. Atg5-dependent LC3II localization onto the osteoclast ruffled border promotes fusion with secretory lysosomes [24]. The deletion of Atg5 or Atg7 suppresses trafficking of cathepsin K-containing lysosomes at the ruffled border and inhibits secretion of lysosomal contents into the resorption lacuna [20,24]. This study found that aesculetin inhibited the induction of LC3, Atg5, and Atg7 involved in LC3 lipidation of osteoclasts, indicating that this compound inhibited fusion of LC3II-decorated membrane with secretory lysosomes. Similarly, gossypetin diminished the ATG induction and lysosomal cathepsin K activity in actin ring-bearing osteoclasts [27]. Furthermore, it was shown that LC3 associates with microtubules in proximity of the actin ring and regulates Cdc42 [23]. The inhibition of actin ring formation by aesculetin might be ascribed to failure of the regulatory link among the LC3, microtubules, and Cdc42 [23]. In addition, another small GTPase Rab7 is crucial for vesicular trafficking to the ruffled border in an Atg5-dependent manner [24]. The Rab7 induction by RANKL was diminished in aesculetin-bearing mature osteoclasts. It can be assumed that aesculetin impaired fusion of secretory lysosomes with bone-apposed plasma membrane, leading to decreased bone resorption of osteoclasts. There is recent evidence demonstrating that molecular machinery governs the positioning and peripheral distribution of lysosomes in osteoclasts [20,37]. Distinct vesicular transport in osteoclasts may be exploited for the development of pharmacological therapies for bone disorders [13]. The lysosomal exocytosis is mediated by SytVII colocalized with LAMP2 and cathepsin K on lysosomes and promotes the fusion of lysosomes with the ruffled border [21,51]. In this study, aesculetin inhibited cellular induction of SytVII and LAMP2 by the presence of RANKL, indicating defect of formation and exocytosis of secretary vesicles during bone resorption. Furthermore, loss of connection of lysosomes to microtubules impairs lysosomal distribution in osteoclasts leading to failure of bone resorption [20]. LIS1 regulates osteoclastic function through interactions with PLEKHM1 and dynein moving along microtubules [22]. Moreover, the Rab7-associated molecular machinery with LIS1 and PLEKHM1 mediates lysosome peripheral positioning in osteoclasts [22,37]. This study showed that aesculetin hampered the interaction between microtubules and lysosomes by virtue of its ability to inhibit the induction of Rab7, LIS1, and PLEKHM1. These results suggest that Rab7/LIS1/PLEKHM1-mediated positioning to microtubules and movement of lysosomes along with microtubules may be a new therapeutic target for bone diseases with increased resorption. Since this study focused on exploring lysosomal mechanisms involved in anti-resorptive effects of aesculetin, the potential antiosteoclastogenetic mechanism(s) of aesculetin should be elucidated. In summary, the current study demonstrated that aesculetin abrogated the RANK signaling-triggered multinucleated osteoclast formation and bone resorption in macrophages. Aesculetin hampered the actin ring formation along with deterring cellular induction of αvβ3 integrin and Cdc42. This compound failed to create the acidic resorption lacuna at the osteoclast–bone matrix interface through dampening the induction of proton pump and chloride channel on the ruffled border and anion exchanger on the contra-lacunar membrane domain. Furthermore, aesculetin inhibited the fusion of lysosomes with the ruffled border through disturbing induction of Atg and LC3 proteins and blunted SytVII- and LAMP2-mediated lysosomal exocytosis. Finally, aesculetin deranged lysosome positioning to microtubules and lysosomal movement mediated by the Rab7/LIS1/PLEKHM1-associated Int. J. Mol. Sci. 2020, 21, 8581 15 of 17 trafficking machinery. Although aesculetin may serve as a modulator against osteoclastogenesis under in vitro conditions, its dietary role as an osteoprotective agent remains elusive.

Author Contributions: W.N., E.-J.L., and Y.-H.K. (Young-Hee Kang) designed the research; W.N., E.-J.L., D.Y.K., H.O., S.-I.K., and S.Y.O. conducted the research; Y.-H.K. (Yun-Ho Kim) and M.-K.K. analyzed the data; W.N. and Y.-H.K. (Young-Hee Kang) wrote the paper; Y.-H.K. (Young-Hee Kang) had primary responsibility for the final content. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by a National Research Foundation of Korea grant funded by the Korea government (2019R1A2C1003218). Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

Ae2 anion exchange protein 2 Atg autophagy-related proteins CAII carbonic anhydrase II ClC-7 chloride channel 7 LAMP2 lysosome-associated membrane protein 2 LC3 light chain 3 LIS1 lissencephaly-1 OPG osteoprotegerin PLEKHM1 Pleckstrin homology domain-containing protein family member 1 MMP matrix metalloproteinase RANK receptor activator of nuclear factor-κB RANKL RANK ligand Syt VII synaptotagmin VII TRAP tartrate-resistance acid phosphatase V-ATPase vacuolar-type H(+)-ATPase

References

1. Kenkre, J.S.; Bassett, J.D. The bone remodelling cycle. Ann. Clin. Biochem. 2018, 55, 308–327. [CrossRef] [PubMed] 2. Feng, X.; McDonald, J.M. Disorders of bone remodeling. Annu. Rev. Pathol. 2011, 6, 121–145. [CrossRef] [PubMed] 3. Hadjidakis, D.J.; Androulakis, I.I. Bone remodeling. Ann. N. Y. Acad. Sci. 2006, 1092, 385–396. [CrossRef] [PubMed] 4. Raggatt, L.J.; Partridge, N.C. Cellular and molecular mechanisms of bone remodeling. J. Biol. Chem. 2010, 285, 25103–25108. [CrossRef][PubMed] 5. Xiao, W.; Wang, Y.; Pacios, S.; Li, S.; Graves, D.T. Cellular and molecular aspects of bone remodeling. Front. Oral Biol. 2015, 18, 9–16. [CrossRef] 6. Asagiri, M.; Takayanagi, H. The molecular understanding of osteoclast diﬀerentiation. Bone 2007, 40, 251–264. [CrossRef] 7. Boyle, W.J.; Simonet, W.S.; Lacey, D.L. Osteoclast diﬀerentiation and activation. Nat. Cell Biol. 2003, 423, 337–342. [CrossRef] 8. Wada, T.; Nakashima, T.; Hiroshi, N.; Penninger, J.M. RANKL–RANK signaling in osteoclastogenesis and bone disease. Trends Mol. Med. 2006, 12, 17–25. [CrossRef] 9. Kenny, A.M.; Raisz, L.G. Mechanisms of bone remodeling: Implications for clinical practice. J. Reprod. Med. 2002, 47, 63–70. 10. Cosman, F. Anabolic and antiresorptive therapy for osteoporosis: Combination and sequential approaches. Curr. Osteoporos. Rep. 2014, 12, 385–395. [CrossRef] 11. Anagnostis, P.; Gkekas, N.K.; Potoupnis, M.; Kenanidis, E.; Tsiridis, E.; Goulis, D.G. New therapeutic targets for osteoporosis. Maturitas 2019, 120, 1–6. [CrossRef][PubMed] 12. Papapoulos, S.; Makras, P. Selection of antiresorptive or anabolic treatments for postmenopausal osteoporosis. Nat. Clin. Pract. Endocrinol. Metab. 2008, 4, 514–523. [CrossRef][PubMed] Int. J. Mol. Sci. 2020, 21, 8581 16 of 17

13. Ng, P.Y.; Ribet, A.B.P.; Pavlos, N.J. Membrane trafficking in osteoclasts and implications for osteoporosis. Biochem. Soc. Trans. 2019, 47, 639–650. [CrossRef][PubMed] 14. Väänänen, H.K.; Zhao, H.; Mulari, M.; Halleen, J.M. The cell biology of osteoclast function. J. Cell Sci. 2000, 113, 377–381. [PubMed] 15. Stenbeck, G. Formation and function of the ruffled border in osteoclasts. Semin. Cell Dev. Biol. 2002, 13, 285–292. [CrossRef] 16. Rousselle, A.-V.; Heymann, D. Osteoclastic acidification pathways during bone resorption. Bone 2002, 30, 533–540. [CrossRef] 17. Henriksen, K.; Sørensen, M.G.; Nielsen, R.H.; Gram, J.; Schaller, S.; Dziegiel, M.H.; Everts, V.; Bollerslev, J.; A Karsdal, M. Degradation of the organic phase of bone by osteoclasts: A secondary role for lysosomal scidification. J. Bone Miner. Res. 2005, 21, 58–66. [CrossRef] 18. Josephsen, K.; Praetorius, J.; Frische, S.; Gawenis, L.R.; Kwon, T.-H.; Agre, P.; Nielsen, S.; Fejerskov, O. Targeted disruption of the Cl-/HCO3- exchanger Ae2 results in osteopetrosis in mice. Proc. Natl. Acad. Sci. USA 2009, 106, 1638–1641. [CrossRef] 19. Kornak, U.; Kasper, D.; Bösl, M.R.; Kaiser, E.; Schweizer, M.; Schulz, A.; Friedrich, W.; Delling, G.; Jentsch, T.J. Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 2001, 104, 205–215. [CrossRef] 20. Lacombe, J.; Karsenty, G.; Ferron, M. Regulation of lysosome biogenesis and functions in osteoclasts. Cell Cycle 2013, 12, 2744–2752. [CrossRef] 21. Zhao, H.; Ito, Y.; Chappel, J.; Andrews, N.; Ross, F.P.; Teitelbaum, S.L. How do bone cells secrete proteins? Adv. Exp. Med. Biol. 2010, 658, 105–109. [CrossRef][PubMed] 22. Ye, S.; Fowler, T.W.; Pavlos, N.J.; Ng, P.Y.; Liang, K.; Feng, Y.; Zheng, M.; Kurten, R.; Manolagas, S.C.; Zhao, H. LIS1 regulates osteoclast formation and function through its interactions with Dynein/Dynactin and Plekhm1. PLoS ONE 2011, 6, e27285. [CrossRef][PubMed] 23. Chung, Y.-H.; Yoon, S.-Y.; Choi, B.; Sohn, N.H.; Yoon, K.-H.; Kim, W.-J.; Kim, D.-H.; Chang, E.-J. Microtubule-associated protein light chain 3 regulates Cdc42-dependent actin ring formation in osteoclast. Int. J. Biochem. Cell Biol. 2012, 44, 989–997. [CrossRef][PubMed] 24. DeSelm, C.J.; Miller, B.C.; Zou, W.; Beatty, W.L.; van Meel, E.; Takahata, Y.; Klumperman, J.; Tooze, S.A.; Teitelbaum, S.L.; Virgin, H.W. Autophagy proteins regulate the secretory component of osteoclastic bone resorption. Dev. Cell 2011, 21, 966–974. [CrossRef] 25. Ðudarić,L.; Fužinac-Smojver, A.; Muhvić,D.; Giacometti, J. The role of polyphenols on bone metabolism in osteoporosis. Food Res. Int. 2015, 77, 290–298. [CrossRef] 26. Horcajada, M.N.; Offord, E. Naturally plant-derived compounds: Role in bone anabolism. Curr. Mol. Pharmacol. 2012, 5, 205–218. [CrossRef] 27. Antika, L.D.; Kim, Y.-H.; Kang, M.-K.; Park, S.-H.; Lee, E.-J.; Choi, Y.-J.; Kang, Y.-H. Dietary compound gossypetin inhibits bone resorption through down-regulating lysosomal cathepsin K activity and autophagy-related protein induction in actin ring-bearing osteoclasts. J. Funct. Foods 2016, 24, 390–402. [CrossRef] 28. Turk, V.; Stoka, V.; Vasiljeva, O.; Renko, M.; Sun, T.; Turk, B.; Turk, D. Cysteine cathepsins: From structure, function and regulation to new frontiers. Biochim. Biophys. Acta 2012, 1824, 68–88. [CrossRef] 29. Sundaram, K.; Nishimura, R.; Senn, J.; Youssef, R.F.; London, S.D.; Reddy, S.V. RANK ligand signaling modulates the matrix metalloproteinase-9 gene expression during osteoclast differentiation. Exp. Cell Res. 2007, 313, 168–178. [CrossRef] 30. Wang, Q.; Xie, Y.; Du, Q.-S.; Wu, X.-J.; Feng, X.; Mei, L.; McDonald, J.M.; Xiong, W.-C. Regulation of the formation of osteoclastic actin rings by proline-rich tyrosine kinase 2 interacting with gelsolin. J. Cell Biol. 2003, 160, 565–575. [CrossRef] 31. Nakamura, I.; Pilkington, M.F.; Lakkarkorpi, P.T.; Lipfert, L.; Sims, S.M.; Dixon, S.J.; Rodan, G.A.; Duong, L.T. Role of αvβ3 integrin in osteoclast migration and formation of sealing zone. J. Cell. Sci. 1999, 112, 3985–3993. [PubMed] 32. Jurdic, P.; Saltel, F.; Chabadel, A.; Destaing, O. Podosome and sealing zone: Specificity of the osteoclast model. Eur. J. Cell Biol. 2006, 85, 195–202. [CrossRef][PubMed] 33. Chabadel, A.; Bañon-Rodríguez, I.; Cluet, D.; Rudkin, B.B.; Wehrle-Haller, B.; Genot, E.; Jurdic, P.; Anton, I.M.; Saltel, F. CD44 and beta3 integrin organize two functionally distinct actin-based domains in osteoclasts. Mol. Biol. Cell. 2007, 18, 4899–4910. [CrossRef][PubMed] Int. J. Mol. Sci. 2020, 21, 8581 17 of 17

34. Itzstein, C.; Coxon, F.P.; Rogers, M.J. The regulation of osteoclast function and bone resorption by small GTPases. Small GTPases 2011, 2, 117–130. [CrossRef][PubMed] 35. Okumura, S.; Mizoguchi, T.; Sato, N.; Yamaki, M.; Kobayashi, Y.; Yamauchi, H.; Ozawa, H.; Udagawa, N.; Takahashi, N. Coordination of microtubules and the actin cytoskeleton is important in osteoclast function, but calcitonin disrupts sealing zones without affecting microtubule networks. Bone 2006, 39, 684–693. [CrossRef] 36. Zhao, H.; Ito, Y.; Chappel, J.; Andrews, N.W.; Teitelbaum, S.L.; Ross, F.P. Synaptotagmin VII regulates bone remodeling by modulating osteoclast and osteoblast secretion. Dev. Cell 2008, 14, 914–925. [CrossRef] 37. Fujiwara, T.; Ye, S.; Castro-Gomes, T.; Winchell, C.G.; Andrews, N.W.; Voth, D.E.; Varughese, K.I.; Mackintosh, S.G.; Feng, Y.; Pavlos, N.; et al. PLEKHM1/DEF8/RAB7 complex regulates lysosome positioning and bone homeostasis. JCI Insight 2016, 1, e86330. [CrossRef] 38. Wang, X.; Diao, L.; Sun, D.; Wang, D.; Zhu, J.; He, Y.; Liu, Y.; Xu, H.; Zhang, Y.; Liu, J.; et al. OsteoporosAtlas: A human osteoporosis-related gene database. PeerJ 2019, 7, e6778. [CrossRef] 39. Eriksen, E.F. Cellular mechanisms of bone remodeling. Rev. Endocr. Metab. Disord. 2010, 11, 219–227. [CrossRef] 40. Crane, J.L.; Cao, X. Bone marrow mesenchymal stem cells and TGF-β signaling in bone remodeling. J. Clin. Investig. 2014, 124, 466–472. [CrossRef] 41. Houschyar, K.S.; Tapking, C.; Borrelli, M.R.; Popp, D.; Duscher, D.; Maan, Z.N.; Chelliah, M.P.; Li, J.; Harati, K.; Wallner, C.; et al. Wnt pathway in bone repair and regeneration—What do we know so far. Front. Cell. Dev. Biol. 2019, 6, 170. [CrossRef][PubMed] 42. Glass, D.A.; Karsenty, G. Molecular bases of the regulation of bone remodeling by the canonical Wnt signaling pathway. Curr. Top. Dev. Biol. 2006, 73, 43–84. [CrossRef][PubMed] 43. Reid, I.R. Anti-resorptive therapies for osteoporosis. Semin. Cell Dev. Biol. 2008, 19, 473–478. [CrossRef] [PubMed] 44. Suvarna, V.; Sarkar, M.; Chaubey, P.; Khan, T.; Sherje, A.; Patel, K.; Dravyakar, B. Bone health and natural products- An insight. Front. Pharmacol. 2018, 9, 981. [CrossRef] 45. Baek, J.M.; Park, S.-H.; Cheon, Y.-H.; Ahn, S.-J.; Lee, M.S.; Oh, J.; Kim, J.-Y. Esculetin attenuates receptor activator of nuclear factor kappa-B ligand-mediated osteoclast differentiation through c-Fos/nuclear factor of activated T-cells c1 signaling pathway. Biochem. Biophys. Res. Commun. 2015, 461, 334–341. [CrossRef] 46. Kim, J.-L.; Kang, M.-K.; Gong, J.-H.; Park, S.-H.; Han, S.-Y.; Kang, Y.-H. Novel antiosteoclastogenic activity of phloretin antagonizing RANKL-induced osteoclast differentiation of murine macrophages. Mol. Nutr. Food Res. 2012, 56, 1223–1233. [CrossRef] 47. Kim, Y.-H.; Kim, J.-L.; Lee, E.-J.; Park, S.-H.; Han, S.-Y.; Kang, S.A.; Kang, Y.-H. Fisetin antagonizes cell fusion, cytoskeletal organization and bone resorption in RANKL-differentiated murine macrophages. J. Nutr. Biochem. 2014, 25, 295–303. [CrossRef] 48. Anderegg, F.; Geblinger, D.; Horvath, P.; Charnley, M.; Textor, M.; Addadi, L.; Geiger, B. Substrate adhesion regulates sealing zone architecture and dynamics in cultured osteoclasts. PLoS ONE 2011, 6, e28583. [CrossRef] 49. Ito, Y.; Teitelbaum, S.L.; Zou, W.; Zheng, Y.; Johnson, J.F.; Chappel, J.; Ross, F.P.; Zhao, H. Cdc42 regulates bone modeling and remodeling in mice by modulating RANKL/M-CSF signaling and osteoclast polarization. J. Clin. Investig. 2010, 120, 1981–1993. [CrossRef] 50. Bestebroer, J.; V’Kovski, P.; Mauthe, M.; Reggiori, F. Hidden behind autophagy: The unconventional roles of ATG proteins. Traffic 2013, 14, 1029–1041. [CrossRef] 51. Zhao, H. Membrane trafficking in osteoblasts and osteoclasts: New avenues for understanding and treating skeletal diseases. Traffic 2012, 13, 1307–1314. [CrossRef][PubMed]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).