1 CHARACTERIZATION OF ENOXACIN: A NOVEL VACUOLAR H+ ATPase DIRECTED OSTEOCLAST INHIBITOR By EDGARDO J. TORO QUI ONES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Edgardo J Toro Qui ones
3 To God, for allowing me to take on this challenge and leading me every step of the way To my wife, Ileana for her unconditional love and support which have been crucial in the fulfillment of my professional goals To my children, Marco and Marsela for patiently waiti ng for me to come home every night you are both incredible and make me very proud every day.
4 ACKNOWLEDGMENTS My fondest gratitude goes to my mentor Dr. Holliday for his guidance, patience and support He provided excellent mentorship and always encouraged me to pursue my career goals. I want to thank the members of my committee : Dr. Culp, Dr. Wronski and Dr. Neubert for all their useful comments, suggestions and generous support. I am particularly thankful to Dr. Wallet for her unconditional advice and support during critical times. I would not have made it this far in my career without the support and encouragement of Dr. Robert Burne. His helping hand has been pivotal in obtaining the many research experiences I have had along my professional career. I want to thank Dr Wheeler and Dr. Dolce for the opportunity and also for their ge nerous support and mentorship during my residency Throughout my professional career I have met a lot of people that I consider my friends and to whom I owe my gratitude for their support: my classmates and future colleagues for always sharing their exper iences in and outside the lab, the many administrators, lab technicians, and other support personnel that h ave always had words of wisdom, provided me with good advice and encouragement Most importantly, I want to thank my family for their support and understanding. They have lived this process every step of the way by my side and I will never be able to thank them enough for their love.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF FIGURES .......................................................................................................... 8 LIST OF ABBREVIATIONS ........................................................................................... 10 ABSTRACT ................................................................................................................... 12 CHAPTER 1 INTRODUCTION .................................................................................................... 14 Osteoclast Activation .............................................................................................. 14 RANKL/RANK/OPG Signaling Mechanism ............................................................. 15 Signaling Pathways of Osteoclast Formation and Activation .................................. 15 Osteoclast Attachment to the Bone Surface is an Essential Step prior to Bone Resorption ........................................................................................................... 17 Cytoskeletal Rearrangement during Osteoclast Activation ..................................... 18 Vacuolar H+ATPases (V ATPases) ....................................................................... 19 Intracellular Functi on and Distribution of V ATPases .............................................. 20 Subcellular Localization of V ATPase during Osteoclast Activation ........................ 21 V ATPases Bind Actin Microfilaments and Other Cytoskeletal Elements ............... 22 Characterization of V ATPase Binding to Microfilaments through the B Subunit .... 25 Genetic Analys is of the Physiologic Role of the Actin Binding Activity of B Subunits ............................................................................................................... 26 Identification of Enoxacin as an Inhibitor of the Binding Interaction between V ATPase and Microfilaments ................................................................................. 28 Orthodontic Tooth Movement ................................................................................. 29 General Purpose and Rationale of the Research ................................................... 31 2 CHARACTERIZATION OF THE INHIBITION OF OSTEOCLASTS BY ENOXACIN ............................................................................................................. 37 Introductory Remarks .............................................................................................. 37 Materials and Methods ............................................................................................ 38 Reagents and Antibodies ................................................................................. 38 Osteoclast Differentiation ................................................................................. 38 Tartrate Resistant Acid Phosphatase 5b (TRAP5b) Activity Assay .................. 39 TUNEL Assay and Nuclei Counting .................................................................. 39 Caspase 3 Assay for Apoptosis ........................................................................ 39 Qu antitative Immunoblotting ............................................................................. 40 Statistics ........................................................................................................... 40 Results .................................................................................................................... 41
6 Enoxacin Inhibits Differentiation of RAW 264.7 Cells into Osteoclasts without Inducing Excess Apoptosis ........................................................................... 41 Enoxacin Does Not Alter the Expression of Several Osteoclast and V ATPase Selective Genes .............................................................................. 42 Enoxacin Decreases the Amount of V ATPase Associated with the Cytoskeleton ................................................................................................. 43 Enoxacin Alters Proteolytic Processing of TRAP .............................................. 43 Effects of Enoxacin on LPlastin ....................................................................... 44 Summary of Results ................................................................................................ 44 Discussion .............................................................................................................. 45 3 RATIONAL DEVELOPMENT OF A DELIVERY METHOD FOR THE LOCALIZED SUSTAINED RELEASE OF ENOXACIN ........................................... 55 Introductory Remarks .............................................................................................. 55 Materials and Methods ............................................................................................ 56 Reagents and Antibodies ................................................................................. 56 Microfilament Binding Assay ............................................................................ 57 Osteoclast Differentiation ................................................................................. 57 Tartrate Resistant Acid Phosphatase 5b (TRAP5b) Activity Assay .................. 58 Bone Targeting Assay ...................................................................................... 58 Resorption A ssay ............................................................................................. 58 Caspase 3 Assay for Apoptosis ........................................................................ 59 Quantitative Immunoblotting ............................................................................. 59 Statistics ........................................................................................................... 60 Results .................................................................................................................... 60 BENX Inhibits the Interaction between the Yeast B Subunit and Microfilaments ............................................................................................... 60 BENX Reduced TRAP+ Cells like Enoxacin ..................................................... 60 BENX Does Not Alter Protein Expression of V ATPase and Osteoclast Related Proteins ............................................................................................ 61 BENX Triggers Higher Levels of Apoptosis in RANKLStimulated Cultures ..... 61 BENX Effectively Decrease Osteoclast Mediated Bone Resorption ................. 61 BENX Inhibits Bone Resorption with an IC50=1 M ........................................... 62 Summary of Results ................................................................................................ 63 Discussion .............................................................................................................. 63 4 EFFECTS OF BENX ON O RTHODONTIC TOOTH MOVEMENT .......................... 75 Introductory Remarks .............................................................................................. 75 Materials and Methods ............................................................................................ 76 Orthodontic Tooth Movement Model as Described by GJ King ........................ 76 Orthodontic Tooth Movement Model Modified by JK Neubert .......................... 77 Statistics ........................................................................................................... 78 Results .................................................................................................................... 78 BENX Significantly Inhibits Orthodontic Tooth Movement ................................ 78 Summary of Results ................................................................................................ 80
7 Discussion .............................................................................................................. 80 5 SUMMARY, CONCLUSIONS AND FUTURE DIRECTIONS .................................. 88 Summary and Conclusions ..................................................................................... 88 Future Directions .................................................................................................... 93 Examine the Ef fects of Enoxacin on MicroRNA Activity and Gene Expression in Osteoclasts ................................................................................................ 93 Determine Whether the Inhibition of the Interaction between V ATPase and Microfilaments by Enoxacin is Important for Its Anti Tumorigenic Activity ..... 94 Determine the Effect of BENX on Osteonecrosis of the Jaw ............................ 94 Determine Alternative Mechanisms for Local Delivery of Enoxacin or BENX ... 95 APPENDIX: E NOXACIN ALTERS THE SURFACE EXPRESSION OF DCSTAMP, A PROTEIN LINKED TO CELL FUSION.................................................................... 98 Materials and Methods ............................................................................................ 98 Real Time PCR ................................................................................................ 98 Flow Cytometry ................................................................................................ 99 LIST OF REFERENCES ............................................................................................. 102 BIOGRAPHICAL SKETCH .......................................................................................... 115
8 LIST OF FIGURES Figure page 1 1 Schematic drawing of osteoclast organization during bone resorption. ............. 33 1 2 Vacuolar H+ ATPase subunit composition and existing isoforms. ...................... 34 1 3 Selective inactivation of osteoclast by inhibition of V ATPase/microfilament interaction.. ......................................................................................................... 35 1 4 Molecular structure of enoxacin.. ....................................................................... 36 2 1 Enoxacin dosedependently reduces the number of TRAP+ multinuclear osteoclast like. ................................................................................................. 48 2 2 Enoxacin does not alter number of viable nuclei or levels of apoptosis ............ 49 2 3 Quantitative analysis of caspase3 activity in RAW 264.7 cells plus or minus RANKL, and treated with vehicle or 50 M enoxacin. ......................................... 50 2 4 Enoxacin does not alter the expression of several osteoclast and V ATPase selective genes. .................................................................................................. 51 2 5 Enoxacin decreases the amount of V ATPase associated with the detergent insoluble cytoskeletal fraction. ............................................................................ 52 2 6 Enoxacin alters proteolytic processing of TRA P5b. ........................................... 53 2 7 Enoxacin triggered conversion of cellular Lplastin from a 67 kD to 57 kD form.. .................................................................................................................. 54 3 1 Molecular structure of bisphosphonateconjugated enoxacin. ........................... 66 3 2 BENX disrupts the interaction between B subunit of V ATPase and microfilaments.. .................................................................................................. 67 3 3 BENX reduces the number of TRAP+ giant cells with IC50 similar to enoxacin. ........................................................................................................... 68 3 5 BENX triggers higher levels of apoptosis in R ANKL stimulated cultures ........... 70 3 6 BENX inhibits osteoclast bone resorption in vitro with an IC50 in the low micromolar range. ............................................................................................... 71 3 7 BENX binds to bone at a saturation level of 0.8 mg/ml/cm1.. ............................ 72 3 8 BENX inhibits bone resorption more effectively than ENX by predictably binding to the bone surface.. .............................................................................. 73
9 3 9 BENX inhibits bone resorption by a different mechanism than ALN. ................. 74 4 1 Orthodontic tooth movement in a rat model.. ..................................................... 83 4 2 Modified rat orthodontic tooth movement model ............................................... 84 4 3 Weight monitoring of rats in the control group with original OTM model.. .......... 85 4 4 Weight monitoring of rats treated with the modified OTM model. ...................... 86 4 5 BENX reduces OTM in rats after 28day activation period. ............................... 87 5 1 ENX can be incorporated into a biologically inert polymer. ................................ 97 A 1 Enoxacin has only minor effec ts on mRNA levels of many osteoclast genes.. 100 A 2 Enoxacin alters the distribution of DC STAMP. ............................................... 101
10 LIST OF ABBREVIATION S A ATPase Archaea V ATPase ADAM A Disintegrin and Metalloproteinase Domain ALN Alendronate AP1 Activator Factor 1 Arf 6 ADP Ribosylating F actor 6 ARNO ADP Ribosylation F act or Nucleotide Site Opener Arp2/3 Actin Related Protein 2/3 ATF Activating Transcription F actor BENX Enoxacin Bisphosphonate Hydrobromide ( BisphosphonateConjugated Enoxacin) BIJON BisphosphonateInduced Oral Osteonecrosis Caspase Cyst eineAspartic Acid Protease cfos Cellular Nuclear Protein Fos DCSTAMP Dendrit ic CellS pecific Transmembrane Protein ENX Enoxacin F actin Filamentous Actin F ATPase Mitochondrial F0F1 ATPase FPPS F arnesyl Diphosphate Synthase M CSF Macrophage Colony Stimulating Factor MITF M icrophthalmiaAssociated Transcription Factor MITF Microphthalmia Transcription Factor NFATc1 Nucle ar Factor of Activated T cells Cytoplasmic 1 NF B Nuclear Factor Kappa of Activated B Cells N WASP Neu ronal Wis cott Aldrich Syndrome P rotein ONJ Osteonecrosis of the Jaw
11 OPG O steoproteg erin OTM Orthodontic Tooth Movement PI3K P h osphatidylinositol 3K inase PU.1 PU Box Binding 1Transcription Factor RANK Receptor Activator of NF B RANKL Receptor A ctivator of NF B L igand SERM Selective Estrogen Receptor Modulator TARP 2 TAR RNA Binding Protein 2 TRAF TNF Receptor A ssoci ated Factor TRAP Tartrate Resistant Acid Phosphatase TUNEL Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling V ATPase V acuolar H+ ATPases WASH WASP and SCAR Homologue
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF ENOXACIN: A NOVEL VACUOLAR H+ ATPase DIRECTED OSTEOCLAST INHIBITOR By Edgardo J. ToroQuiones August 2012 Chair: L. Shannon Holliday M ajor: Medical Sciences Molecular Cell Biology E noxacin was recently identified as an inhibitor of a binding interaction between the B2 subunit of vacuolar H+ATPase (VATPase) a nd microfilaments and of osteoclast formation and bone resorption in mouse marrow cultures containing osteoclasts and osteoblasts. I demonstrated that enoxacin blocked osteoclast formation in Raw 264.7 cells indicating that it acts directly on osteoclasts Treatment of mouse marrow cultures and Raw 264.7 osteoclast like cells wit h enoxacin reduced the amount of B2 subunit and a3 subunit associated with the detergent insoluble cytoskeleton in differential pelleting assays. The relative levels of a variety of osteoclast marker proteins were not altered by treatment with enoxacin. Enoxacin blocked the proteolytic activation of TRAP5b and triggered a reduction in the size of the actinbundling protein LPlastin. B one resorption by mature osteoclasts wa s blocked by enoxacin i n vitro To target enoxacin to bone, it was incorporated into a bisphosphonate backbone (bis enoxacin; BENX). BENX retained its capacity to block binding between recombinant B2subunit and pure mi crofilaments Osteoclast formation in either calcitriol stimulated mouse marrow cultures or receptor activator of nucl ear factor kappa B ligand (RANKL) stimulated Raw 264.7 cells was reduced by BENX with an IC50 of 10 M identical to enoxacin. However,
13 u nl ike enoxacin, BENX increased caspase 3mediated apoptosis B ENX was significantly more effective than enoxacin at red uc ing bone resorption when bone was present (IC50=1 M). A novel orthodontic tooth movement model in rats was used to study the effect of BENX in vivo After a 28day period of orthodontic activation, BENX significantly inhibited OTM in vivo when compared to vehicle control In summary, we have utilized a rational approach to identify an anti osteoclastic agent which selectively inhibits bone resorption by inhibiting the interaction between vacuolar H+ATPases and microfilaments. This novel m echanism of action may confer advantages over existing therapeutic agents in the management of osteoclast rel ated disease.
14 CHAPTER 1 INTRODUCTIO N B one remodeling is a physiologic process that results from the controlled rate of bone deposition and resorption via the recrui tment of specialized cells called osteoblasts and osteoclasts. Osteoblasts arise from multipotent mesenchymal stem cells (1) whereas osteoclasts are derived from the fusion of plur ipotent mononuclear cells of the hematopoietic lineage in the bone marrow (2). These osteogenic cells are responsible for the formation of bone during development and for the remodeling of bone throughout life. The i nteraction between these cells is dynamic, highly regulated, and has a central role in the coup ling of bone formation and resorption. A change in the coordina tion of the bone remodeli ng process result s in the development of pathologic processes underlying certain hum an conditions such as: infantile malignant osteopetrosis (3), osteoporosis (4) and chronic inflammatory diseases such as rheumatoid arthritis (5,6) and periodontal disease (7). Osteoclast Activation Osteoclasts are specialized multinuclear cells that dissolve bone mineral and degrade organic matrix in a highly regulated manner (8). Formation of osteoclasts is governed primarily by the expression of receptor activator of NFkB ligand (RANKL) and macrophage colony stimulating factor (M CSF) (9 12) Osteoblasts express RANKL on their cell surf ace thus ; a cell cell interaction between these cel ls and osteoclast precursors is required f or osteoclastogenesis (13,14) Upon activation of the osteoclast precursors, a number of signaling mechanisms are triggered in the differentiating osteoclast which will be described i n detail.
15 RANKL/RANK/OPG Signaling M echanism T he description of the RANK/RANKL/OPG system is considered by many to be the most important finding in the past 15 years in the field of bone biology. In 1997 Simonet and colleagues from Amgen identified osteoprotegerin (OPG), a member of the tumor necrosis receptor family, as a novel soluble factor capable of preventing osteoclast differentiation from precursor cells in vitro (15) Shortly after, Lacey and colleagues identified an OPG ligand (OPGL, RANKL) by screening a complimentary DNA expression library of mouse bone marrow cells (ST2) treated with 1,25(OH)2D3 and using OPG as a probe (16) The protein was f ound to induce osteoclast formation from osteoclast progenitors in vitro and later termed RANKL (17) Thi s finding was quickly followed by the identification of RANK as a receptor on osteoclast precursor cells that mediate RANKLinduced osteocl ast differentiation and activation (10) Studies of transgenic mice with mutations leading to the deficiency of RANK, RANKL, or M CSF demonstrate severe osteopetrosis excessive accumulation of bone, and lack of bone modeling along with a significant decrease in osteoclast number and increased trabecular bone density. Osteoprotegerindeficient mice suffer from severe osteoporosis reduced bone strength and bone mineral density associated with increased osteoclast numbers and vascular calcification. The data from these murine models show the importance of these molecules in the process of osteoclastogenesis and bone resorption (9,12,15,18) Signaling Pathways of Osteoclast Formation and Activation The M CSF and RANKL activation signals trigger an intracellular cascade of signaling molecules which activate transcription factors such as : PU Box binding 1 ( PU 1 ) microphthalmia transcription factor ( MITF ) cellular nuclear protein Fos ( cfos )
16 and nuclear factor kappa of activated B cells ( NFkB ), and nuclear factor of activated T cells cytoplasmic 1( NFATc1 ) that result in precursor commitment and formation of osteoc lasts The PU.1, MITF and cfos transcription factors are involved in the stages of commitment and differentiation of hematopoietic precursors into osteoclasts. The phenotype that results from deletion of the PU 1 gene is not compatible with life; however, it can be rescued by bone marrow transplant of cells from a normal animal indicating that PU.1 is an important transcription factor during the early commitment process of hematopoietic precursors (19,20) Mice deficient in PU. 1 show a reduction in the expression of M CSF receptor (c fms) and are devoid of macrophages and osteoclasts leading osteopetrosis (19) A similar osteopetrotic phenotype i s observed in mice deficient of MITF or c fos. This phenotype is due mostly to a reduction in the number of osteoclasts and can also be rescued by bone marrow transplantation. However, no alterations in the number of macrophages were observed indicating th eir osteoclastic regulatory function is downstream from PU.1 (21,22) Another transcription factor that is expressed early during activation of osteoclasts is the act ivator protein 1 ( AP1 ) It is a heterodimeric protein complex formed by the association of either Fos or activating transcription factor (ATF) pro teins with Jun proteins. AP1 is activated via RANKL induced expression of c fos which subsequently promotes the nuclear translocati on of c Jun (23) Activation of NF B occurs mainly via the recruitment of TNF receptor associated factor (TRAF) proteins such as TRAF2, TRAF5, and TRAF 6 to the RANKL RANK receptor complex TRAF6 activate s I B kinase (IKK) which in turn will phosphorylate I B protein, an inhibitor of NF B and
17 targets it for proteosomal degradation (24) NF B is then free to migrate into the nucleus and induce transcription of genes that characterize the osteoclast phenoty pe. In addition to NF B and AP1, the Nuclear Factor of Activated T cells c1 ( NFATc1 ) is strongly induced by RANKL during osteoclast activation (25) NFATc1 is a member of the NFAT family of proteins ( NFATc1 5) which are activated by the serine/threonine phosphatase called calci neurin. During osteoclastogenesis, i ntracellular calcium activates calcineurin which then removes the phosphates from the serine residues on the inactive cytoplasmic NFAT exposing the nuclear localization and translocation signal. NFATc1 is then recruited to its own promoter and selectively autoamplifies its gene (26) Both NFATc1 and NF B are involved in the regulation of the gene expression of tartrate resistant acid phosphatase ( TRAP ) cathepsin K, 3 integrin, and calcitonin receptor (25,2729) Ost eoclast Attachment to the Bon e Surface is an Essential Step p rior to Bone Resorption Osteoclast binding to bone surface occurs via cell surface receptors know n as integrins (30) These are heterodimeric surface proteins composed of an and subunit that me diate cell cell and cell matrix interacti ons. Osteoclasts express several integrins that are involved in osteoclast attachment to bone surface such as: 21 (collagen/laminin) v1 (fibronectin) and v3 (vitronectin) From these integrins the vitronectin receptor ( v3) is the most abundant integrin in osteoclasts (31 34) Ligand binding to vitronectin receptor s is mediated via an argi nineglycine aspartic acid (RGD) pepti de sequence present within the ligands (35) It has been shown that RGD containing peptides and antibodies against the vitronectin receptor can inhibit osteoclast
18 bone resorption in vitro (36,37) In addition, v3deficient mice display progressive osteosclerosis whe re osteoclas ts develop but have migratory difficulty and lack actin rings (30) Taken together these data demonstrate an important role of v3 in the adhesion and polarization of osteoclasts. Cytoskeletal Rearrangement d uring Osteoclast Activation Osteoclasts become activated when they bind to the bone and the plasma membrane becomes compartmentalized into specialized areas such as : the basolateral and apical membranes which are not in direct contact with bone, the sealing zone which lies directly o ver the bone surface, and the ruffled membrane which is a highly convoluted membrane present within the sealing zone ( Figure 1 1 ) (8,3841) The sealing zone is delineated by an actin ring composed of a core of filamentous actin structures called podosomes The actin ring is also surrounded by matrix binding proteins such as: integrins vinculin and talin (42,43) Other proteins including cortactin, the neuronal Wiscott Aldrich Syndrome protein ( N WASP ) a nd the Arp2/3 complex are also associated with the sealing zone (44 46) The highly convoluted ruffled membrane within the actin ring is packed with vacuolar H+ATPases (VATPases) These enzymes pump protons unidirectionally across the cell mem brane into the bone interface to create an acidic microenvironment necessary for the dissolution of bone minerals and the ac tivation of secreted cathepsin K, an enzyme involved in the degradation of the organic fraction of bone (47,48) Formation of both the act in ring and ruffled membrane is essential for osteoclast mediated bone resorption and are characteristic marker s of osteoclast activity (49) Osteotropic regulators of osteoclast activation such as parathyroid hormone (PTH) and
19 1,25dihydroxyvit amin D3 can alter the size of actin ring s and ruffled membranes (50) The absence of th ese structures seen in osteopetrotic rats and mice cor relate with the inability of osteoclast s to resorb bone (51) Vacuolar H+ATPases (V ATPases) V ATPases are lar ge multisubunit enzymes expressed at low levels in most eukaryotic cells ( Figure 1 2) They localize to a number of intracellular membranous organelles of the endocytic, exocytic and phagocytic pathways to carry out cellular housekeeping functions. V ATPases are forbidden to enter the plasma membrane. However, V ATPase s containing cell type specific isoforms of certain subunits are found in the plasma membranes of cells with specialized functions such as: renal intercalated cells (52) osteoclasts (53) epi didymal clear cells (54) and metastatic cancer cells (55) V ATPases pump protons across cellular membranes and are critical for the regulation of pH inside intracellular organelles (56) .The a cidification of intracellular compartments is required for a variety of cellular processes such as receptor mediated endocytosis, protein degradation, and the processing of signaling molecules (56 58) V ATPases mitocho ndrial F0F1 ATPase (F ATPase, ATP synthase) and Archaea V ATPase (A ATPase) share a close structural and enzymatic relationship (59,60) Significant amount of information regarding the overall structure of V ATPases has been inferred from the still accumulating collection of crystal structures of F ATPases (61 66) H owever, the three also diverge in crucial structural and enzymatic features (67) V ATPases are organized into two domains, V1 and V0, composed of eight and six subunits respectively that operate by a rotary mechanism ( Figure 1 2 ) (68) The V1 domain is a peripheral complex on the cytoplasmic side of the membrane composed of
20 the A and B subunits These form a hexameric ring that is structurally similar to other rotary enzymes. T he A subuni t contains the catalytic site for ATP hydrolysis which is coupled to structural changes in the AB hexameric ring and powers turning of a central stalk. In addition, V ATPases have three nonidentical stator arms and each of them has an EG dimer that interacts with the V1 head ( Figure 1 2 ) The V0 domain of V ATPase is embedded in the cell ular membrane and mediates proton transport across the membrane (69) ATP hydrolysis in the V1 domain is coupled to the active transport of protons across the membrane via a ring o f c subunits present in the V0 domain (70) In mammals, the V0 domain also contains one of four isoforms of the asubunit, a large integral protein that is thought to contribute to the proton channel The a subunit isoforms are also linked to the sorting of vesicles to different subcellular compartments, although the mechanisms involved in differential sorting are not yet known (71 77) Intracellular Function and Distribution of V ATP ases V ATPases with the same basic overall structure and enzymatic activity are segregated in cells so that they can perform a variety of housekeeping functions. However, some cells have additional subsets of V ATPases that perform specialized roles. For exampl e, o steoclast s contain housekeeping V ATPases that acidify various cellular compartments (74) but, they also express a specialized subs et of V ATPases that are targeted to the plasma mem brane during bone resorption (75) This targeting is particularly interesting because V ATPases are normally excluded from the plasma memb rane. Some of t he subunits that make up V ATPase s exist as multiple isoforms ( Figure 1 2) (68) Some of these are ubiquitous elements of the housekeeping V ATPases
21 whereas others like subunit B2 in osteoclasts, are also components of specialized V ATPases (78) Specialized V ATPases are typically associated with one or more cell and tissue specific isoforms These isoforms are involved in the cells speci alized function, and are often targeted to cellular domains other than those required for the housekeeping functions. For example, four isoforms of subunit a (a1, a2, a3 and a 4) have been identified. T he a1 and a2 isofor ms are ubiquitously expressed T he a3 isoform is highly expressed by osteoclasts (75) microglia (79) and pancreatic beta cells (80) and the a4 isoform is highly expressed by renal inter calated cells (76,77) and epididymal clear cells (54) In humans, m utations of the a3or a4 isoform s lead to infantile autosomal malignant osteopetrosis (81,82) or a kidney disease called autosomal distal renal tubular acidosis respectively (76) In eukaryotic cells, very precise targeting of V ATPases is required for cell function and survival. However, d espite the fundamental importance of this regulation the process of V ATPase targeting is not mechanistically understood The precision by which V ATPases must be delivered in space and time to specific cellu lar compartments suggest that either direct o r indirect associations exist between V ATPase and cytoskeletal elements. Subcellular Localization of V ATP ase d uring Osteoclast Activation Bone resorption by osteoclasts requires degradation of the organic and inorganic components of bone. When o s t eoclasts bind to bone profound cytoskeletal rearrangement occurs that lead to the formation of a specialized resorptive structure called the ruffled plasma membrane (ruffled membrane, ruffled border), which is encircled by an equally unusual cytoskeletal structur e called the actin ring which is formed as a coalescence of distinct structures called podosomes (42,43,83)
22 In osteoclasts, the ruffled membrane is highly enriched with V ATPases which acidify the extracellular space between the ruffled border and bone surface (84) The excess protons create an environment in which the bone mineral dissolv es and the acid cyste ine proteinase, cathepsin K, digest s the organic components of the bone (47,48) During osteoclast differentiation, the mRNA and protein levels of most V ATPase subunits increase and cell specific isoforms are expressed ( i.e. a3, d2) leading to increased levels of a specialized subset of V ATPases stored within intracellular vesicles (83,85) The V ATPase containing vesi cles are directed to the newly forming ruffled plasma membrane in coordination with actin filaments, and then fuse wi th the plasma m e mbrane (86) Once bone resorption at a specific site is completed, the V ATPases are reinternalized into the cytoplasm and the osteoclast migrates to anothe r site of resorption where actin rings and ruffled membranes are formed again for another round of resorption (87,88) V ATPases Bind Actin Microfilaments a nd Other Cytoskeletal Elements V ATPases bind directly to microfilaments (actin filaments, F actin) (86) Immunoprecipitation studies using an antibody against subunit E (E11) demonstrated that F actin and myosin II are immunoprecipitated along with V ATPase subunits. This binding interaction was competitively inhibited by the target peptide of E11. Under actindepoly merizing conditions, onl y V ATPase subunits but no actin or myosin II were found in the precipitates suggest ing that V ATPases bind actin filaments directly, and myosin II indirectly by way of microfilaments. This was confirmed by demonstrating that V ATPases isolated from either mouse osteoclasts or porcine kidneys bound microfilaments composed of actin that had been isolated from rabbit muscles, and which did not contain myosin II (86)
23 V ATPase s bound microfilaments with a s toichiometry of 1 V ATPase per every 67 actin monomers in a filament, and an affinity of 55 nM (86) T ransmission electron microscopy of V ATPases bound to microfilaments showed that they interacted with F actin through the top of the V1 domain suggesting that intact V ATPases associated with membranes could potentially bind microfilaments. Thus, in principle, the interaction could have a role in sorting V ATPases between vesicle or membrane populations. The interaction between V ATPase s and microfilament in osteoclasts varied in response to physiologic stimuli. In inactive osteoclasts V ATPases co localize d with mic rofilaments whereas in actively resorbing osteoclasts, the amount of co localization is sha rply reduced. A transition state was detected during activation of osteoclasts where both V ATPases and m icrofilaments concentrate d in a patch together at the site of the nascent ruffled membrane (86) Moreover, a similar transition state was observed as V ATPase s were internalized from the ruffled membrane of activated osteoclasts in response to treatment with phosphatidylinositol 3kinase (PI3K ) in hibitors (89) T hese data suggested that binding between V ATPases and microfilaments might be involved in regulating V ATPase distribution and/or activity in osteoclasts (86) Actin binding sites have been identified in both the B1a nd B2 isoforms of the B subunit (88,90) and in the C subunit (91 93) The sequence in the actin binding site of the V ATPase B subunit is conser ved in organisms such as: Manduca sexta, Saccharomyces cerevisiae and Arabid opsis columbia (92,94,95) The C subunit from Tobacco hornworm ( Manducca sexta) also binds F actin, but the actinbinding activity of the C subunit has not yet been characterized in other organisms (92,93)
24 V ATPases also interact with the small GTP binding pr otein ADP ribosylating factor 6 (Arf6) and its activator, ADP ribosylation factor nucleotide site opener (ARNO) both of which are involved in vesicular trafficking and cytoskeletal reorganization (96 100) Studies by Vlad Mar shansky in Dennis Browns group showed that the recruitment of ARNO to early endosomes was dependent upon the intravesicular pH of the V ATPase containing vesicles (96) They propose that V ATPases regulate vesicular trafficking and cytoskeletal remodeling in response to pH (96) V ATPases also interact with several glycolytic enzymes that include aldolase and phosphofructokin ase 1 (101105) which are known to also bind microfilaments (106 108) A recent description of the interactions between V AT Pases, fructose bis phosphate aldolase and ARNO suggest a mechanism by which the spatial localizatio n and activity of V ATPases might be coupled to the metabolic state of the cell (97) Carnell and colleagues showed that the WASP and SCAR homologue (WASH) complex regulates the exocytic process in Dictyostelium discoideum by inducing the removal of V ATPases from the lysosomal surface to neutralize the vesicle before the contents of the vesicle are exocytosed (109) The WASH complex is an evolutionarily conserved regulator of actin filaments that coats intracellular vesicles and is involved in vesic ular trafficking. WASH was shown to recruit Arp2/3, an actin nucleator, which leads to the formation of patches of microfilaments on the WASH coated vesicles This WASH Arp 2/3microfilament interaction was required for the removal of V ATPases from the vesicle (109) Interestingly, Arp2/3 is also involved in the formation of actin rings and is upregulated during osteoclast activation (46) suggesting that the V ATPase
25 interaction with cytoskeletal elements might be important for the precise intracellular trafficking of V ATPa se coated vesicles (110) Recent unpublished data from our group (Jian Zuo, Edgardo J. Toro and L. Shannon Holliday, unpublished ) shows that all seven components of the WASH complex (WASH1, FAM21, strumpellin, SWIP, CDC53, CapZalpha and CapZbeta) are upregulated during osteoclast formation and that the WASH complex is associated with actin rings and vesicle that are located near ruffled membranes of resorbing osteoclasts. Efforts are underway to test whether knockdown of the WASH complex disrupts osteoclasts, and whether this is tied to the ac tivity of enoxacin. Characterization o f V ATP ase Bin ding to Microfilaments t hrough t he B Subunit In 2000 the B subunit of V ATPase was identified as a potential F actin binding protein by conducting blot overlay studies in which F actin was used to probe V ATPases that had been isolated, separated by SDS PAGE an d transferred to nitrocellulose (90) This was confirmed in experiments in which V ATPases were disassembled and the B subunit remained isolated bound to F actin. Both of the known isoforms of the B subunit (B 1, B2) w ere shown to bind F actin as demonstrated by experiments using bacterially expressed recombinant fusion proteins containing various fra gments of the B1 and B2 isoforms. S mall fragment s of approximately 44 amino acids in length of the B1 ( amino acids 2367 ) and B2 subunits (amino acids 2973) bou nd microfilaments as tightly as full length subunits. However, smaller fragments did not bind F actin (90) The fusion proteins bound to F actin with a stoichiometry of one B subunit per monomer actin molecule within a filament and with a KD of 100 to 200 nM similar to that found in the B subunit from mammals and other evoluti onarily distinct organisms like Manducca sexta (93) and yeast (94) These data suggest that the
26 capacity of B subunits of V ATPase to bi nd microfilaments developed early in evolution and has been conserved. The actin binding region of the B subunit contains a 13 amino acid sequence similar to a portion of the actinbinding site of a mammalian protein called profilin (Dr. Michael R. Bubb) (88) Studies o f synthesized peptides confirm ed that this profili n like sequence can bind actin and also that p eptides derived from the actinbinding site on the B subunit compete with profilin for binding to actin. Moreover, point mutations known to disrupt the actinbinding activity of profilin (111) also decrease d the actin binding activity of the B subunit derived peptides (88) These data were consistent with the profilin like regio n being a vital element of the actin binding domain of B subunits The enzymatic activity of V ATPases wa s not affected by the actinbinding activity of the B subunit. When the sequence of recombinant B1 and B2, was altered by replacing the profilinlike domain with an identical length spacer composed of the sequence of the B subunit from the Archaean Pyrococcus horikoshii the actin binding activity was eliminated (88) This spacer was selected because it would probably not alter the overall st ructure of the subunit, and because Archaeans do not have a mi crofilament based cytoskeleton thus, it would be unlikely for them t o have actin binding activity. The sub stitution of the profilin like domain with the Archaeaderived spacer in a yeast model also confirms that the enzymatic activity of V ATPase is unaffected, even though it eliminated the ability of the pump to bind microfilaments (94) Genetic Analysis of the Physiologic Role of t he Actin Binding Activity o f B Subunits The structure of V ATPases in Saccaromyces cerveciae is similar to those found in mammals (112) U nl ike most other eukaryotic cells, it do es not require V ATPase
27 enzymatic activity to survive U nder acidic culture conditions Saccaromyces cerveciae can transport sufficient external protons across the membrane for house keeping functions H owever, w hen challenged under external alkaline pH it requires V ATPase activity for survival allowing for knockin experiments of mutant subunits of V ATPase to be readily performed (113116) The knock in of the mutant B subunit sequence containing the Archaea derived spacer into a wild type B subunit background eliminated t he actin binding activity of B subunit s without affect ing the ability of the yeast to survive under normal culture conditions in alkaline pH. However, when yeast expressing the mutant construct was cultured in the presence of sub lethal doses of wortmannin and cyclohexamide, it grew two orders of magnitude slower when compared with yeast expressing wild type B subunits (94) This suggests that the actinbinding activi ty in B subunit may have initially evolved as an element of a response to environmental toxins. To test the role of the actin binding activity of B subunit in osteoclasts either wildtype B1 or B1 lacking the actinbindin g site (B1 mut) were transduc ed usi ng a viral vector (70) Both B1 and B1mut assembled with other V ATPase subu n its in osteoclasts and localized to vesicular compartments in inactive osteoclasts. Upon activation, t he B1 subunit was efficiently targeted to the ruffled plasma membrane of resorbing osteoclasts in the same manner as the endogenous B2 subunit. However, B1mut was never detected in ruffled plasma membranes (70) These results s uggested that the actin binding activity of B subunit is necessary to traffic V ATPases to vesicles that later fuse with the plasma membrane to form the ruffled plasma membrane. Therefore, an inhibitor of the interaction between the B2subunit and microfi laments would be
28 expected to prevent proper targeting of the V ATPases to the ruffled plasma membrane and inhibit bone resorption ( Figure 1 3 ) Identification of Enoxacin as an Inhibitor of t he Binding Interaction between V ATPase and Microfilaments To screen for small molecule inhibitors of the B2microfilaments interaction, a virtual highconfidence atomic level model of the actinbinding domain of B2 was constructed (117) T his virtua l atomic model was made considering the close sequence homology that exists bet subunit of F ATPase as well as the many crystal structures available subunit of F ATPase as the primary guide in modelin g the B2 structure. F1F0 ATPase (ATP synthase) is an energy converting multisubunit mechanoenzyme present in mitochondria and bacteria. It consists of a globular domain (F1) and a membranebound domain (F0) and uses the flow of protons across the membrane to synthesize of ATP (61) T he virtual model was also informed by the relationship between B2 and profilin since profilin has been crystallized in complex with actin, and this allowed better insight into the actinbinding surface of B2 (118) A computer based virtual screening of over 300,000 small molecules in a library from the National Cancer Institute was c arried out to identify drug like molecules that bind to the actinbinding site on the B subunit by virtual molecular docking. This kind of screening at the very best reduces the number of candidate molecules from 100,000s to 100s (119 121) A simple microfilament pelleting assay was then used to determine whether 100 M of each of the test molecules affected interaction between recombinant B2 and rabbit muscle F actin (117) From the top ranked molecules identified a total of four were found to strongly inhibit the interaction between B2 and F actin in the pelleting
29 assay. Of these, two were found to inhibit osteoclast formation a nd activity with an IC50 around 10 M without affecting the viability of the c ells, while the other two were lethal to osteoclasts and other cells. The lethal molecules had been identified in previous screens and were known to be cytotoxic for re asons unr elated to the V ATPase. To date we have focused our attention primarily on enoxacin ( Figure 1 4 ) which is a second generation fluoroquinolone antibiotic (122) A practical advantage to studying enoxacin is that it can be obtained in large quantities for modest prices making detailed in vitro and i n vivo studies possible. Enoxacin was used in the United States for about ten years for the treatment of urinary tract infections and gonorrhea, and was voluntarily taken from the market in the US because of adverse effects including insomnia dizziness, and photosensitivity (123) However, i t is still used in much of the rest of the world as an antibiotic It has been reported that V ATPase microfilamen t interactions occur in microglia thus, i t is plausible that some of the side effects linked to enoxacin might be the result of disrupting the B2microfilament interaction in other c ell types (79) More studies will be required to identify other cells that might be affected by inhibitors of B2microfilament interactions and subsequent effects. Tendon ruptures have also been associa ted with enoxacin. T hese are likely class action effects (124) since they have also been linked to other quinolones that do not affect the microfilament B2 interaction (117) O rthodontic Tooth Movement In the mouth, osteoclast medi ated bone r esorption is critical to physiologic processes such as tooth eruption and exfoliation of deciduous teeth. It is important for the orthodontic correction of malocclusions and is the underlying cause of
30 pathophysiologic processes such as generation of bone defects and tooth loss which may arise from periodontal disease and/or traumatic injury to the teeth (125,126) D uring orthodontic tooth movement, mechanical stress is applied directly on the tooth resulting in t he alveolar bone and periodontal ligament being compressed on one side and stretched on the opposite side of the tooth socket This mechanical stimulus is sensed by the bone cells and triggers bone remodeling (mechanotransduction) which results in displacement of the tooth. Many orthodontic treatment plans require the movement of one tooth while keeping a nearby tooth stable. Currently, a variety of elaborate appliances are used for these procedures but they require significant patient compliance, and increase the level of biomechanics complexity making them far from ideal for orthodonti c treatment. Ideally, the clinician could pharmacologically block th e bone remodeling process to prevent tooth movement without the use of these appliances. Recent studies performed in rats have shown that it is possible to pharmacologically manipulate the orthodontic tooth movement process (127130) A number of groups have reported either accelerating or reducing orthodontic tooth mo vement using agents such as RANKL, matrix metalloproteinase inhibitors and, integrin inhibitors (129,130) Cert ain bisphosphonates have also been shown to effective ly inhibit tooth movement (127) Bisphosphonates are a well established class of drugs used for the treatment of diseases which involve excess osteoclast mediated bone resorption (131) They are thought to act mainly by induction of cell ular apoptosi s; h owever, an increasing number of bisphosphonatein duced osteonecrosis of the jaw (BIJON) cases (132) have made these drugs less than ideal for the treatment of excess
31 bone resorption in oral tissues. I t is apparent that the development of new therapeutic approaches is necessary to target excess osteoclast activity more effectively. General Purpose and Rationale of t he Research V ATPase s have both housekeeping roles and specialized roles in specific cell types that are related to important human pathologies including cancer and bone disease (56) Significant ef forts have been put forth attempting to identify inhibitors of V ATPase enzymatic activity that are selective for subsets of V ATPases (133,134) T o date this work has not yielded therapeutic agents that are clinically useful. The purpose of the present study is to help determine whether enoxacin or other inhibitors of the interaction between B2and microfilaments prove useful for treating difficult orthodontic cas es or even against bonerelated disease in general. More generally, our approach to the study of enoxacin involve s targeting the specific and specialized interactions of a housekeeping enzyme. The goal for the use of enoxacin is not to kill the cells, but rather to regulate cell type selective machinery so that there is less of an undesired activity. A better understanding of how proteinprotein interactions define cell type specific functions and of the structural ba sis for these interactions at the atomic level would enhance the use of virtual screens as an alternative to high throughput screens and as a practical shortcut in the drug discovery process (135 140) This study represent one out of a number of examples that have emerged during the past few years showing the feasibility of knowledgebased virtual screening as a rational approach to drug discovery (117,141144) The long term goal of this project is to determine whether enoxacin holds promise as a drug for use in various medical and dental applications in which inhibition of osteoclast mediated bone resorption would be beneficial. The main objective is to
32 exa mine the mechanisms by which enoxacin inhibits osteoclast formation and bone resorption in greater detai l and to test the effects of enoxacin on bone resorption in a well characterized rat orthodontic tooth movement model. The hypothesis is that enoxacin is an example of a novel class of ther apeutic agents for the treatment of osteoclast mediated disease that may selectively reduce osteoclast formation and bone resorption.
33 Figure 1 1. Schematic drawing of osteoclast organization during bone resorption (side view). Osteoclasts are multin ucleated (N) cells that bind to bone surfaces via vitronectin receptors. Upon activation their cellular membrane polarizes into an apical domain (A), basolateral domains ( B) and a ruffled membrane (RM) within the sealing zone (SZ).
34 Figure 1 2. Vacuolar H+ ATPase subunit composition and existing isoforms. V ATPases are multisubunit mec h anoenzymes organized into two domains, V1 and V0 of 8 and 6 subunits respectively. Various of the subunits exist as isoforms which are tissuespecific and involved in s pecialized cellular functions (i.e. o steoclasts, renal intercalated cells) (145) Adapted from ( Toro EJ Ostrov DA Wronski TJ Holliday LS Ration al identification of enoxacin as a novel V ATPase directed osteoclast inhibitor. Curr Protein Pept Sci. 2012 Mar 1;13(2):18091. Bentham Direct Open Ac cess P ublisher ) with permission.
35 Figure 1 3 Selective inactivation of osteoclast by inhibition of V ATPase/microfilament interaction. A) Binding of microfilament to actin binding site on the B2subunit during osteoclast activation. B) Small molecule binds to actin binding site on B2 subunit and prevents V ATPase/F actin interaction.
36 Figure 1 4 Molecular structure of enoxacin. Copied from ( Herczegh P, Buxton TB, McPherson JC 3rd, Kovcs Kulyassa A, Brewer PD, Sztaricskai F, Stroebel GG, Plowman KM, Farcasiu D, Hartmann JF. Osteoadsorptive bisphosphonate derivatives of fluoroquinolone antibacterials. J Med Chem 2002 May 23;45(11):23384. ACS Publications ) with permission.
37 CHAPTER 2 CHARACTERIZATION OF THE INHIBITION OF OS TEOCLASTS BY ENOXACI N Introductory Remarks Osteoclasts express very high levels of a subpopulation of V ATPases (a3and d2containing V ATPases) that are targeted to the ruffled plasma membrane when they encounter bone (75,83,146) The targeting of V ATPases to the ruffled plasma membrane is absolutely r equired for bone resorption (75) In osteoclasts, a high proportion of V ATPases are b ound to microfilaments (86) This V ATPase/micro filament interaction correlates with the activation state of osteoclasts and is mediated by a specific region of the B subunit capable of binding the actin filament (70,88,90,147) Small changes in the profilin like domain of the B2subuni t disrupt the actin binding activity without interfering with the capacity of the altered B2subunit to contribute to the enzymatic activity of the multisubunit V ATPase (94) More importantly, V ATPase s containing the altered B su bunits were not targeted to the ruffled pl asma membrane of osteoclasts (70) T he mechanism by which this subpopulation of V ATPases bind to microfilaments and are transported to the ruffled membrane during osteoclast activation represent a novel potential therapeutic target that might be selective for osteoclast mediated bone resorption. By using c omputational chemistry techniques we identified enoxacin which blocked the interaction between recombinant B2subunit and microfilaments, and also blocked osteoclast differentiation and bone resorption in tissue culture (117) The transport of V ATPases to the ruffled plasma membranes of osteoclasts on bone slices was also disrupted. Moreover, e noxacin did not prevent growth or mineralization by osteoblasts at conc entrations where osteoclast activity was inhibited. In this study, we
38 examine in greater detail the mechanism by which enoxacin affects osteoclasts, making use of RANKL stimulated RAW 264.7 cells and RANKL and CSF 1 stimulated primary osteoclasts. Materials and Methods Reagents and Antibodies The polyclonal anti E, a3, and B2 subunit antibodies were described previously (88,148) The anti TRAP antibody was from Biolegend (cat.no. 648402). Anti L plastin was obtained from Abcam (cat # ab83496). Anti cortactin and anti DCSTAMP anti bodies were from Santa Cruz (sc 25577, and sc 25579). Unless otherwise noted, other antibodies and reagents were obtained from the SigmaAldrich Chemical CO. (St. Louis, MO). Recombinant sRANKL was produced in E.Coli as described previously (46) Enoxacin was obtained from SigmaAldrich and dissolved in DMSO (117) or in 0.1 M sodium hydroxide (149) BENX was obtained from SynQuest Laborator ies (Alachua, FL, cat # 8H77B 06). Osteoclast Differentiation RAW 264.7 cells w ere differentiated into osteoclast like cells by stimulating them with recombinant RANKL as described previously (46) RAW 264.7 were seeded on 24well plate at a density of 1.25 104 cells per well, or 6well plates at 1.8 X 105 cells per well. These cells were cultured for 5 days with 5 ng/ml of RANKL and were fed on day 3 in culture. Mouse marrow osteoclasts were generated as described (150) Briefly, 8 20 g m Swiss Webster mice were euthanized by cervical dislocation and femorae and tibiae dissected. The bone marrow was flushed f rom the dissected bones by cutting both ends of the bones and inserting a syringe with a 25 gauge needle and 0% fetal
39 bovine serum The marrow was washed twic days at a density of 1X106 cells/cm2 on tissue culture plates in 8 M 1, 25 dihydroxyvitamin D3. Osteoclasts in culture were detected as giant cells which sta ined positive for tartrate resistant acid phosphatase (TRAP) activity All mouse protocols were approved by t he University of Florida Institution al Animal Care and Usage Committee (IACUC) Tartrate Resistant Acid Phosphatase 5b (TRAP 5b ) Activity Assay TRAP activity was detected using the Leukocyte Acid Phosphatase kit (Sigma Aldrich; cat. # 387A KT) following the instructions from the manufacturer Osteoclasts were detected as staining positive for TRAP 5b activity. TRAP+ cells were counted and classified as mono, multi nuclear (210 nuclei), or giant cells (more than 10 nuclei) according to the number of nuclei present. TUNEL Assay a nd Nuclei Counting After 5 days of cell culture with 50 M enoxacin, the cells were rinsed twice with 1X PBS solution and fixed for 2 min in 4% paraformaldehyde at room temperature. Apoptotic cells were detected by using an in situ apoptosis detection kit (Promega, G7132) according to the manufacturer's instructions. Measurement of apoptosis was calculated as percentage of apoptotic nuclei (dark brown nuclei) versus total nuclei of multinucleated TRAP positive cells, evaluated in three independent measurements. Casp ase 3 Assay f or Apoptosis Cells were plated in 24 well plates at a density of 0.5 104 cells/cm2 and treated with 5 ng/ ml M enoxacin for 24, 48, and, 72 h. Caspase3 assays were performed following the manufacturer's instructions (catalog No. APT131, Millipore, Temecula, CA). At each time point, the plate was centrifuged at 1200 rpm for
40 centrifuged at 1200 rpm for 10 min. T he colorimetric reaction was quantified using a BioTek KC4 spectrophotometer (Winooski, VT) at 405 nm. Quantitative Immunoblotting Immunoblots were performed by standard procedures using the SuperSignal WestDura chemiluminescence detection system (Pierce). To determine the association of V ATPase with the detergent insoluble cytoskeletal fraction, blots of supernatants and pellets were obtained after extraction of osteoclasts with 20 mM Tris HCl, pH 7.4, 100 mM NaCl, 5 mM MgCl2, 0.5 mM ATP, 0.2 mM CaCl2, 0.2 mM dithiothrei tol (Buffer F) plus 1% Triton X 100 and protease inhibitors, and ultracentrifugation (100,000 x g for 45 min). The samples were heated at 85 C for 10 min, cooled to room temperature, and centrifuged at 10,000 x g for 1 min, and the supernatants were appli ed to SDS polyacrylamide gels, transferred to nitrocellulose, and probed with antibodies as described in the figure legends. Cell sample dilutions were applied to SDS polyacrylamide gel and stained with coomasie blue stain (BioRad, cat # 1610436) to identify and put target proteins into the linear range of detection by immunoblot. We then assayed samples from enoxacin doseresponse experiments to correlate changes in protein levels with doses of enoxacin. Statistics Count e rs were precal ibrated for their ability to identify TRAP+ multinuclear cells and then blinded to treatment groups. Results are expressed as mean S.E. Samples were compared by oneway ANOVA followed by Bonferroni post test or S tudents t test using the program GraphPad Prism 5 (GraphPad Software, La Jolla, CA). p values < 0.05 were considered significant.
41 Results Enoxacin Inhibits Differentiation of RAW 264.7 Cells i nto Osteoclasts without Inducing Excess Apoptosis RAW 264.7 cells were stimulated with recombinant RANKL in 24well plates and treated with either vehicle, 5, 10, 25, 50 or, 100 M enoxacin. After 5 days cells were fixed and examined for TRAP activity and fusion into multinuclear cells (both are characteristics of osteoclasts). A significant dose dependent reduction in the number of TRAP+ and multinucleated cells was detected ( Figure 2 1 A ). To determine whether enoxacin disrupted cell growth or survival, we counted the number of nuclei after 5 days in cultures of RAW 264.7 cells treated with RANKL plus various concentrations of enoxacin. Although the phenotype of the cells was very different when the cells wer e treated with enoxacin ( Figure 2 1 C ), there were no significant differences in the number of nuclei ( Figure 2 1 B ). To determine if enoxacin induced apoptosis during the maturation process, RAW 264.7 cells were treated with either vehicle or 50 M enoxacin and stimulated with RANKL. Colorimetric TUNEL assays were performed and the number of TUNELpositive cells quantified. No significant difference was observed in the number of apoptotic cells that was attributable to enoxacin ( Figure 2 2 ). Activation o f the cysteineaspartic acid protease (Caspase) family initiates apoptosis in mammalian cells. Caspase 3 plays an integral role during the sequential activation of other caspases and can be used as a marker of cell apoptosis (151) To confirm the results from the TUNEL assay, we tested caspase3 activity at different time points (24,48, or 72 hours) after stimulation with RANKL in Raw 264.7 cells treated with either vehicle or 50 M enoxacin ( Figure 2 3 ) No statistically significant differences in
42 caspase3 activity were observed in unstimulated cells tr eated with either vehicle or enoxacin. After 48 hours, a significant increase in caspase3 activity is observed in both treatments upon stimulation with RANKL. More importantly, enoxacin significantly decreased caspase3 activity at 72hours compared with control. Enoxacin Does Not Alter t he Expression of Several Osteoclast a nd V ATP aseSelective Genes During osteoclast differentiation, the cellular content of the different V ATPase subunits is increased in addition to other proteins associated with osteoclast fusion and activation. A study by Lee et al showed that V ATPase d2deficient mice had reduced number of TRAP+ multinuclear cells and decreased bone resorption similar to our preliminary results (152) Their data suggest that the phenotype observed in these mice may result from a reduction in the expression of factors necessary for fusion of osteoclast precursors. To determine the effect of enoxacin on th e protein expression level s of V ATPase and related proteins in differentiating osteoclasts total protein was extracted from RAW 264.7 cells cultured for 5 days with RANKL plus or minus 50 M enoxacin and analyzed by quantitative immunoblotting using act in as an internal standard. No significant differences in expression levels of subunits a3, B2, or E were detected ( Figure 2 4 ). Similarly, the protein levels of NFATc1, A Disintegrin and metalloproteinase ( ADAM12 ) cortactin, Dendritic cellspecific Transmembrane protein ( DCSTAMP) and cathepsin K were not altered by enoxacin ( Figure 2 4 ). Consistent with this result, real time PCR performed in the Holliday lab by Jian Zuo showed little change in the mRNA levels of osteoclast genes in response to enoxacin ( Figure S 1 Appendix A ).
43 Enoxacin Decreases the Amount o f V ATPase Associated with t he Cytoskeleton V ATPases are associated with the detergent insoluble cytoskeleton during osteoclast activation (69) and this interaction is important for osteoclasts to form ruffled membranes (153) Lee et al demonstrated that V ATPase binds directly to actin filaments (86) and subsequent studies have shown that the actin binding activity of V ATPase B subunit is necessary for proper targeting of the V ATPase to the ruffled membrane (70) To determine whether enoxacin reduced the amount of V ATPase associated with the cytoskeleton in osteoclasts we examined the level of B2subunit associated with the detergent insoluble cytoskeleton plus or minus 50 M enoxacin for 5 days in both RANKL stimulated RAW 264.7 osteoclast like cells and mouse marrow osteoclasts. There was no difference in the total level of B2subunit, but a s hift was detected in the B2 subunit from the detergent insoluble cytoskeleto n t o the soluble fraction ( Figure 2 5 ). Enoxacin Alters Proteolytic Processing o f TRAP When osteoclasts are treated with enoxacin, a d ose dependent reduction in the amount of T RAP activity is observed; however, mRNA levels of TRAP5b are only slightly reduced (154) TRAP5b is expressed as a 38 kD latent proenzyme and is proteolytically cleaved to the active form, which is characterized in blots by a 16 kD band (155) Anti TRAP immunoblots of whole cell extracts from cells treated wi th vehicle showed the expecte d 16 kD band but a series of bands st arting at 38 kD, and a reduction in the 16 kD band was detected in osteoclasts tr ea ted with 50 M enoxacin ( Figure 2 6 ).
44 Effects of Enoxacin on LP lastin Enoxacin was previously shown to inhibit actin ring formation (117) The actin binding proteins cortactin and L plastin are two proteins involved in actin ring formation (156158) During active bone resorption by osteoclasts, the level s of cortactin and Lplastin were reported to increase and decrease respectively (158) To st udy the changes in ac tin ring formation associated with enoxacin, the protein levels of cortactin and Lplastin were assayed by W estern blot The results showed that t he protein expression level of cortactin is not significantly al tered by enoxacin ( Figur e 2 4 ). In contrast, anti L plastin antibody detected a dosedependent shift from the expected 67 kD band to a 57 kD band ( Figure 2 7 ). The amount of the 67 kD band was reduced 88 % (determined by densitometry). Summary of R esults In this chapter we sh ow that enoxacin affects RAW 264.7 osteoclast like cells and primary osteoclasts directly. Enoxacin reduces the expression of TRAP activity and cell fusion with an IC50 the number of nuclei present or inducing apoptosis. The protein levels of several V ATPase subunits were not affected by enoxacin however t he mRNA levels of several osteoclast selective genes were reduced slightly in enoxacintreated cells. A s predicted by its capacity to block the B2subunit/microfilament interaction, enoxacin reduced the amount of B2subunit associated with the detergent insoluble cytoskeleton. Finally we show that enoxacin triggered changes in osteoclasts including reduction in proteol ytic activation of TRAP5b, and a shift in Lplastin from 67 kD to 57 kD
45 Discussion In this study we show for the first time that enoxacin affects osteoclasts directly. In 2009 enoxacin was identified as an inhibitor of an interaction between the B2subunit of V ATPase and microfilaments (117) The treatment of osteoclasts in mixed calcitriol stimulated mouse marrow cultures with enoxacin resulted in a dramatic reduction in TRAP+ m ultinuclear cells and a reduction in the ability of those cultures to resorb bone (IC50=10 M) (117) The effects of enoxacin in these cultures were selective for osteoclasts since growth and mineralization of osteoblasts was not aff enoxacin. Similar results were obtained from our study where we tested enoxacin on osteoclast like cells (Raw 264.7) stimulated with RANKL ( Figure 2 1). Other groups have reported that fibroblast cell lines and primary lymphomononuclear cells are u (159) Because enoxacin was identified as an inhibitor of interaction between B2 and microfilaments, it was not surprising that in enoxacintreated osteoclasts, less B2 was associ ated with the detergent insol uble cytoskeletal fraction. When osteoclasts are treated with enoxacin, a significant reduction in TRAP activity and cel l fusion is observed. Mice that are devoid of DCSTAMP show osteoclasts that are unable to fuse into mult inuclear/giant cells (160) Data from FACS analysis of cells treated with enoxacin showed alterations in the sorting of DC STAMP ( Figure S 1) This represents direct evidence that enoxacin may be perturbing a vesicle sorting pathway in oste oclasts. Because DC STAMP is known to be involved in the fusion of osteoclast precursors (160 163) the change in cell surface expression could
46 be linked functionally to the reduction in multinuclear and giant cells detected in enoxacintreated cultures. The red uction in TRAP activity could also be explained if the interaction between V ATPase and microfilaments plays a role in directing vesicular trafficking. T artrate resistant acid phosphatase is expressed as an inactive proenzyme of approximately 38kD that requires activation by the proteolytic cleavage of its fulllength form into a smaller and highly active 16kD form (164) F ailure of V ATPases to be sorted to the same pathway as TRAP and the consequent failure to acidify vesicles might disrupt the a ction of activating acid proteinases leading to decreased levels of TRAP in its active form Also consistent with the hypothesis that enoxacin was perturbing vesicle trafficking in osteoclast were the results of flow cytometry studies in collaboration with Catalfamo and Wallet. These studies showed that enoxacin altered the amount of the DCSTAMP on the plasma membrane surface ( Figure S 2, Appendix A) A change in Lplastin is consistent with reduction in the formation of actin rings detected previously in response to enoxacin (117) L plastin is an actin bundling p rotein that is present in the core of podosomes and actin rings (165) The formation of actin rings as well as the ability of osteoclasts to resorb bone has been linked directly to the r egulation o f L plastin levels (158) The red uction observed in the 67KDa Lp lastin and corresponding smaller fragment of the molecule (57KDa) suggest that L plastin is likely proteolytically cleaved so that the aminoterminal EF hand domain is removed in response to enoxacin. The EF hand binds calcium and mediates inhibition of binding of L plastin to microfilaments in the presence of calcium (166168)
47 In summary, enoxacin directly inhibits osteoclasts. It reduced the amount of the B2 subunit bound to the detergent insoluble cytoskeletal fraction, altered the transport of DC STAMP to the plasma membrane, and blocked proteolytic activation of TRAP5b from its latent pro enzyme to the active form. Enoxacin triggered a reduction in the size of L plastin, likely due to proteolytic cleavage of the amino terminal calcium binding EF hand domain. We propose that enoxacin selectively affects elements of the osteoclast differentiation and activation program that are downstream of the binding interaction between a3containing V ATPases and microfilaments, which is mediated by the actinbinding site on the B2subunit.
48 Figure 2 1 Enoxacin dosedependently reduces the number of TRAP+ multinuclear osteoclast like. A) TRAP activity assay of osteoclasts treated with vehicle, 5, 10, 25, 50, or 100 M enoxacin (ENX). Cells were stained for TRAP activity and the number of mononuclear, multinuclear (210 nuclei), and giant cells (>10 nuclei) was counted. The average number of TRAP+ cells in each category in vehicletreated cultures was defined as 100%. The values in enoxacintreated cultures are depicted as percentage of the vehicle controls. B) Quantification of the total number of nuclei present in cultures of RAW 264.7 cells treated with RANKL plus various concentrations of enoxacin. The values are depicted as average number of nuclei present in cultures per well of each condition. C) Representative images from the experiment documented in (A) Significance was determined by onew ay ANOVA. p < 0.05 was considered significant. Each condition was compared to vehicle control cells. N= 3 for each condition. Panel s A and B were copied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permissi on.
49 Figure 2 2. Enoxacin does not alter number of viable nuclei or levels of apoptosis A) Qu a n titative analysis of the number of TUNEL positive cells per well after treatment with vehicle or 50 M enoxacin. B and C, Representative images of Raw 264.7 cell cultures stimulated with RANKL and treated with vehicle (B) or 50 M enoxacin (C) for 5 days Apoptotic nuclei are stained dark. Arrows point the edges of g iant cells ( > 10 nuclei ) Bar = 27 M. Students t test was used to determine statistical significance. p > 0.05. The data represent mean SE. N=3 for each condition Panel A was adapted from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894 904. Epub 2012 Apr 2 ) with permission.
50 Figure 23. Quantitative analysis of caspase3 activity in RAW 264.7 cells plus or minus RANKL, and treated with vehic le or 50 M enoxacin. Caspase3 activity was measured at 24, 48, or 72 hours. Units are defined as the amount of enzyme that cleaves 1 nM colorimetric substrate per hour. At 48 and 72 hours a signi ficant increase in caspase3 activity is observed in both treatments upon stimulation with RANKL. Enoxacin significantly decreased caspase3 activity at 72 hours compared with control. OneWay ANOVA; p<0.05. The data represent mean S.E. N=3 for each condition. Copied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894 904. Epub 2012 Apr 2 ) with permission.
51 Figure 2 4 Enoxacin does not alter the expression of several osteoclast and V ATPase selective genes Whole cell extracts from RAW 264.7 cells stimulated with RANKL and treated with SDS PAGE, blotted to nitrocellulose membrane and probed with antibodies directed against the proteins indicated. No significant differences were detected in the amount of any of the probed proteins in cells treated with enoxacin. Preliminary experiments were performed to ensure that the target proteins were at levels within the linear detection range of the antibodies.
52 Figure 2 5 Enoxacin decreases the amou nt of V ATPase associated with the detergent insoluble cytoskeletal fraction. Raw 264.7 cells stimulated wi th RANKL and treated with enoxacin were extracted using buffer F containing 1% Triton X 100 and centrifuged at 100,000 x g for 30 mi n; the supernatants and pellets were suspended in equivalent volumes and the proteins separated by SDS PAGE and analyzed by Western blot V ATPase subunits were found primarily in the detergent insoluble fraction (P) in the vehicle treated cells but shifted toward the soluble fraction (S) in the enoxacintreated osteoclasts. C opied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permission.
53 Figure 2 6 Enoxacin alters proteolyti c processing of TRAP 5b Whole cell extracts from RAW 264.7 cells were treated with RANKL in the presence of vehicle or 50 PAGE, blotted and probed with anti TRAP antibody Cells treated with control express mainly the active form of TRAP ( 16kD ) Treatment with enoxacin led to a reduction in the 16kD band and a large increase in the higher molecular weight bands up to the 38kD band, which is likely the fulllength of TRAP Adapted from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Che m 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permission.
54 Figure 2 7 Enoxacin triggered conversion of cellular L p lastin from a 67 kD to 57 kD form RAW 264.7 cells were stimulated with RANKL and treated with vehicle, 1, 10, 25, or Proteins were extracted, 100 loaded onto SDS PAGE gels, and the proteins were separated, blotted and probed with anti L Plastin In the vehicle the expected 67kD band was detected but at higher concentrations of enoxacin, the level of 67kD detected was reduced and a 57kD band appeared. Actin was used as loading control. C opied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permission. 0 1 10 25 50
55 CHAPTER 3 RATIONAL DEVELOP M E NT OF A DELIVERY METHOD FOR THE LOCAL IZED SUSTAINED RELEASE OF ENOXACIN Introductory Remarks In dentistry, e xcessive bone resorption by osteoclasts is the underlyi ng cause of clinically relevant conditions such as tooth loss, inflammatory root resorption, and replacement resorption as result from bacterial infection or damage to the attachment apparatus. In orthodontics, mechanical stimulation of bone resorption on the pressure side is necessary for tooth movement to occur. In most cases selective control of tooth movement is essential to achieve treatment objectives C omplex treatment mechanics become necessary to prevent unwanted tooth movement thus increasing the overall com plexity of the case and length of treatment If undesirable tooth movement can be prevented with local administration of anti resorptive agents, treatment would be less complex and more predictable. To date, a number of molecular approaches have been studied for the pharmacologic manipulation of orthodontic tooth movement (OTM) and other dental relat ed conditions. Several pharmacologic agents such as: RANKL, osteoprotegerin, integrin inhibitors, inhibitors of matrix metalloproteinases and relaxin can either speedup or slow down OTM in animal models (128 130,169171) However, these agents also have an increased likelihood of collater al effects making them unsuitable for use in the standard practice of orthodontics. Because our data support the idea of a selective inhibition of osteoclast activity by blocking the interaction between V ATPase and microfilaments we believe that enoxacin represents a better candidate for the regulation of OTM and possibly other bonerelated diseases
56 Enoxacin was previously identified as a novel anti osteocl astic and anti resorptive agent (117) Because enoxacin is an antibiotic and has a relatively short half lif e in the body, a number of potential side effects can occur such as: 1) t he development of antibiotic resistance, 2) destruction of intestinal flora and 3) lo w concentration levels insufficient to obtain anti resorptive activity T o overcome t hese limitations the development of methods for the delivery of enoxacin to bone becomes extremely necessary Bisphosphonates have a strong affinity for the calcium phosphate in bone and accumulate rapidly in bone tissues (131) In our efforts to find viable delivery methods, we decided to use an approach based on bisphosphonate conjugates This approach has been used for the delivery of small molecules such as diclofenac (172,173) prostaglandins (174) and steroids (175) I n 2002, Herczegh and his group showed that it is possible to produce fluoroquinolones conjugated to bi sphosphonate backbones that target the molecule to the bone mineral (176,177) These conjugates were envisioned as being useful as antibiotics for the treatment of chronic bone infections By using the free amino group on enoxacin as a tether they were able to bind a bisphosphonate backbone to the molecule (176) ( Figure 3 1). In this chapter, we examine in detail whether the bisphosphonateconjugated enoxacin ( BENX ) retains the capacity to block the B2actin interaction and inactivate osteoclasts like enoxacin. Materials a nd Methods Reagents and Antibodies The polyclonal anti E, anti a3 and anti B2 subunit antibodies were described previously (88,148) The anti TRAP antibody was from Biolegend (cat.no. 648402). Anti cortactin and anti DCSTAMP anti bodies were from Santa Cruz (sc 2 5577, and sc -
57 25579). Unless otherwise noted, other antibodies and reagents were obtained from the Sigma Aldrich Chemical CO. (St. Louis, MO). Recombinant sRANKL was produced in E.Coli as described previously (46) BENX was obtained from SynQuest Laborator ies (Alachua, FL, cat # 8H77B 06). Micr ofilament Binding Assay BENX was tested for its ability to inhibit interaction between rabbit muscle actin and the Vma2pMBP in a pelleting assay (178) Briefly Vma2pVma2pMBP microfila ments ( 1 or 25 BENX or vehicle (ethanol) maximize filament polymerization. Samples were subjected to ultracentrifugation using a Beckman Airfuge (Beckma n Coulter, Fullerton, CA) and pellets and supernatants were collected, subjected to SDS PAGE, and stained with Coo massie Brilliant Blue. Osteoclast Differentiation RAW 264.7 cells were differentiated into osteoclast like cells by stimulating them with recombinant RANKL as described previously (46) RAW 264.7 were seeded on 24well plate at a density of 1.25 104 cells per well, or 6well plates at 1.8 X 105 cells per well. These cells were cultured for 5 days with 5 ng/ml of RANKL and were fed on day 3 in culture. Mouse marrow osteoclasts were generated as described (150) Briefly, 8 20 gm Swiss Webster mice were killed by cervical dislocation, femora and tibia were dissected from adherent tissue, and marrow was removed by cutting both bone ends, inserting a plated at a density of 1X106 cells/cm2 on tissue c
58 plus 108 M 1, 25dihydroxyvitamin D3. Cultures were fed on day 3 by replacing half the media per plate and adding fresh 1, 25dihydroxyvitamin D3. After 5 days in culture, osteoclasts appeared. These were detected as giant cells which sta ined positive for TRAP activity ; a marker for mouse osteoclasts All mouse protocols were approved by t he University of Florida Institutional Animal Care and Usage Committee (IACUC) Tartrate Resistant Acid Phosphatase 5b (TRAP 5b ) Ac tivity Assay TRAP activity was detected using the Leukocyte Acid Phosphatase kit (Sigma Aldrich; cat. # 387A KT) following the instructions from the manufacturer. Osteoclasts were detected as staining positive for TRAP 5b activity. TRAP+ cells were counted and classified as mono, multi nuclear (210 nuclei), or giant cells (more than 10 nuclei) according to the number of nuclei present. Bone Targeting Assay T he affinity of BE NX to bone mineral was determined by making a serial dilution of BENX in 1ml diH2O (1uM to 10mM). A 1 cm2 bone slice was then inserted into each dilution and incubated for 24 hr at 37oC. The absorbance after each incubation period was measured by spectrophotometry (260nm) and plotted into a saturation curve of BENX Resorption Assay R esorpt ion assays were performed by scanning electron micros copy . The area resorbed was determined by taking random microscopic photos, then determining the area on Adobe Photoshop by overlaying a grid and counting grid intersections over pits vs. tota l grid intersections. Micrographs were taken at 200X. Pits were defined as continuous resorbed areas. Bone slices were from bovine cortical bone obtained from
59 Publix supermarket. The bone had marrow removed, then was dried, rough cut and sliced using a l ow speed diamond saw (Buehler, Rockville, IN). Caspase 3 Assay f or Apoptosis Cells were plated in 24 well plates at a density of 0.5 104 cells/cm2 and treated with 5 ng/ ml M enoxacin for 24, 48, and, 72 h. Caspase3 assays were performed following the manufacturer's instructions (catalog No. APT131, Millipore, Temecula, CA). At each time point, the plate was centrifuged at 1200 rpm for cent rifuged at 1200 rpm for 10 min. T he colorimetric reaction was quantified using a BioTek KC4 spectrophotometer (Winooski, VT) at 405 nm. Quantitative Immunoblotting Immunoblots were performed by standard procedures using the SuperSignal WestDura chemiluminescence detection system (Pierce). To determine the association of V ATPase with the detergent insoluble cytoskeletal fraction, blots of supernatants and pellets were obtained after extraction of osteoclasts with 20 mM Tris HCl, pH 7.4, 100 mM N aCl, 5 mM MgCl2, 0.5 mM ATP, 0.2 mM CaCl2, 0.2 mM dithiothreitol (Buffer F) plus 1% Triton X 100 and protease inhibitors, and ultracentrifugation (100,000 x g for 45 min). The samples were heated at 85 C for 10 min, cooled to room temperature, and centrif uged at 10,000 x g for 1 min, and the supernatants were applied to SDS polyacrylamide gels, transferred to nitrocellulose, and probed with antibodies as described in the figure legends. Cell sample dilutions were applied to SDS polyacrylamide gel and stai ned with coomasie blue stain (BioRad, cat # 1610436) to identify and put target proteins into the linear range of detection by immunoblot. We
60 then assayed samples from enoxacin doseresponse experiments to correlate changes in protein levels with doses o f enoxacin. Statistics Microscopists were trained to identify TRAP+ multinuclear cells or resorption pits and then blinded to treatment groups. Results are expressed as mean S.E. Samples we re compared by one or two way ANOVA followed by Bonferroni post test or Student's t test using the program GraphPad Prism 5 (GraphPad Software, La Jolla, CA). p values < 0.05 were considered significant. Results BENX Inhibits the Interaction between t he Y east B Subunit a nd Microfilaments In the previous chapter we showed that enoxacin reduces the amount of V ATPase associated with the cytoskeleton in osteoclasts without affecting the total level of B2 subunit ( Figure 2 5 ). When we tested BENX for its ability to inhibit interaction between rabbit muscle actin and th e Vma2pMBP in a pelleting assay (117) we found that it inhibited the interaction of B2subunit and microfilaments with an IC50 of approximately 10 M ( Figure 3 2 ). BENX Reduced T R AP + Cells like Enoxacin To test whether BENX inhibits osteoclast differentiation like enoxacin we cultured Raw 264.7 cells in DMEM plus 10% fetal bovine serum and RANKL to induce their differentiation into giant multinucleated cells. The cells were treated with diff erent concentrations of BENX After 5 days, the cells were fixed and stained for TRAP activity. BENX dosedepende ntly reduced the number of TRAP + cells and of multinucleated cells in culture. The number of giant cells, a marker for osteoclast
61 fo rmation, was dosedependently reduced by BENX with an IC50 in the low micromolar range similar to enoxacin ( Figure 3 3 ) B ENX Doe s Not Alter Protein Expression of V ATPase and Osteoclast Related Proteins T o determine the effect of BENX on the expression of V ATPase and related proteins we treated RANKLstimulated osteoclasts with vehicle or 50 M BENX for 5 days. Cells were then harvested and whole cell extracts were analyzed by gel electrophoresis and western blotting. Similar to enoxacin, we found that BENX does not alter the expression of V ATPase or osteoclast related markers ( Figure 3 4 A). Importantly, the expression of active form of TRAP5b was reduced indicating that, like ENX, BENX also blocks the proteolytic activation of TRAP 5b ( Figure 3 4 B). BENX Trigger s Higher Levels of Apoptosis i n RANKL Stimulated Cultures To test whether BENX triggers apoptosis RAW 264.7 cells were stimulated with performed at 24, 48 and 72 H to test for apoptosis. We found that BENX modestly promoted unstimulated RAW 264.7 cells to undergo apoptosis, and triggered higher levels of apoptosis in RANKLstimulated cultures after 48 and 72H (Jian Zuo, unpublished data; Figure 3 5 ). BENX Effectively Decrease Osteoclast Mediated Bone Resorption To analyze the effect of B ENX on bone resorpt ion, mature osteoclasts were derived from MBM cells cultured on tissue culture plates with MEM plus 10% FBS and stimulated with 1,25 dihydroxyvitamin D3 (179) On day 6, the cultures were scraped free loaded atop bone slices and treated with either vehicle or various concentrations of B ENX for an additional 3 days The cells were then removed from the bone slices and
62 resorption pits analyzed by scanning electron microscopy (SEM). ENX inhibited bone resorption by calcitriol stimulated mouse marrow osteoclasts containing mature osteoclasts with an IC50 in the low micromolar range ( Figure 3 6 ). BENX Inhibits Bone Resorption with an IC50=1 M BENX binds to b one at a saturati on level of 0.8mg/ ml/ cm1 determined by bone targeting assay ( Figure 3 7 ) When mature osteoclasts were loaded atop bone slices, treated with different concentrations of BENX and allowed to resorb for 3 days, bone resorption was inhibited by 50% at 1 M BENX when all of the BENX was predictably bound to bone ( Figure 3 8 ). We hypothesized that this increase in inhibitory activity was the result of BENX being concentrated to the bone surface and being mobilized as the osteoclasts began resorbing. To analy ze the anti resorptive effect of BENX in greater detail, we compared BENX to ALN Alendronate (ALN; Fosamax) is an active and highly successful anti osteoporotic drug. It binds to bone and inhibits osteoclastic bone resorption by blocking the mevalonate pathway in osteoclasts (180,181) Importantly, ALN inhibits osteoclasts at the same IC50 as ENX (182) For the resor ption assay, bone slices were pre coated in a 1mM BENX or ALN solution for 24 hr at 37oC. After inc ubation, the bone slices were washed 3 times in diH2O to remove the excess unbound drug (measured by spectrophotometry) prior to loading the osteoclasts. The cells were allowed to resorb for additional 3 days and resor ption pits analyzed by SEM. Our results showed BENX inhibited bone resorption by more than 80% whereas the osteoclasts were completely unable to resorb on alendronatecoated bone slices ( Figure 3 9 ).
63 Summary of R esults In this chapter, we show ed that BENX retained its capacity to block interaction between recombinant B2subunit and pure microfilaments in pelleting assays, and reduced formation of TRAP+ osteoclasts in either calcitriol stimulated mouse marrow cultures or RANKL stimulated Raw 264.7 cells wit h an IC50 of 10 M which was identical to enoxacin. Unlike enoxacin, BENX induced higher level of apoptosis in RANKLstimulated Raw 264.7 cells. BENX effectively binds to bone and inhibits osteoclast mediated bone resorption with an IC50 = 1 M. W hen bone slices were pretreated with BENX bone resorption was reduced by approximately 80%. In summary, we have utilized a rational approach to identify an anti oste oclastic agent with a novel mechanism of action. B ENX inhibits bone resorption in tissue c ulture at similar concentrations as alendronate and is selecti vely active against osteoclasts Discussion To reduce the likelihood for unwanted secondary effects related to enoxacin, the compound B ENX was synthesized by adding a bisphosphonate backbone to enoxacin. Based on the hig h affinity of bisphosphonates for bone mineral we expect BENX to be rapidly sequestered and accumulate in bone. Here for the first time we show that BENX is a selective inhibitor of osteoclastogenesis and bone resorption. It bl ocks the interaction between recombinant B subunit and microfilaments in pelleting assays and has identical anti osteoclastic pr operties as enoxacin. However; when bone is present, it inhibits bone resorption more efficiently than enoxacin. We hypothesize that BENX binds to bone and dissociates from bone mineral during active resorption by the osteoclast. BENX is then mobilized into the cell to disrupt the V ATPase actin interaction.
64 Drugs like the bisphosphonates are effective inhibitors of tooth moveme nt (127) They act as potent suppressors of osteoclast s and are currently the major class of drugs used for the treatment of bone diseases such as ost eoporosis and bone cancer (131) Currently, bisphosphonates fall into two classes described by their mechanism of action: 1) the non nitrogen and 2) nitrogencontaining molecules T he n onnitrogen containing bisphosphonates (i.e. clodronate, etidronate) are taken up by the cells and conv erted into a nonhydrolyzable form of ATP that accumulate within the cells and lead to apoptosis (183) Similarly, nitrogencontaining bisphosphonates (i.e. alendronat e, zolendronate, and risedronate) also lead to cellular apoptosis They act by preventing the prenylation of small GTPases via in activation of the enzyme farnesyl diphosphate synthase (FPPS) of the mevaloni c acid pathway (184) Unlike enoxacin, BENX enhances osteoclast apoptosis. This may be a necessary result of the inclusion of the bisphosphonate backbone to target an active moiety to bone. Although we do not fully understand the underlying mechanism, we suspect that like simple bisphosphonates, BENX may function as a nonhydrolysable ATP analog. The fact that BENX induces apoptosis raises concerns tha t, like other bisphosphonates, it may increase the risk of oral osteonecrosis. However, because oral osteonecrosis is linked to bacterial infection, the antibiotic activity of enoxacin may reduce or eliminate the risk or oral osteonecrosis. Our report showed that BENX inhibits the interaction between recombinant B2subunit and microfilaments. It reduced the formation o f TRAP+ giant cells with an IC50 of 10 M which was identical to enoxacin. Also, similar to the parent molecule, i t reduced the amount of the B2subunit bound to the detergent insoluble cytoskeletal fraction and
65 blocked proteolytic activation of TRAP5b. U nlike enoxacin, BENX induced higher level of apoptosis in RANKL stimulated osteoclasts BENX effectively bind s to bone and inhibits osteoclast mediated bone resorption at concentrations 10fold lower than enoxacin. In summary, we have utilized a rational approach to identify an anti osteoclastic agent with a novel mechanism of action that may prove useful in the future for the treatment of osteoclast related disease.
66 Figure 3 1. Molecular structure of bisphosphonateconjugated enoxacin. Enoxacin bisphosphonate hydrobromide (BENX; SynQuest Labs, cat # 8H77 B 06) was synthesized by adding a bisphosphonate backbone to enoxacin (MW : 670.16). Copied from ( Herczegh P, Buxton TB, McPherson JC 3rd, Kovcs Kulyassa A, Brewer PD, Sztaricskai F, St roebel GG, Plowman KM, Farcasiu D, Hartmann JF. Osteoadsorptive bisphosphonate derivatives of fluoroquinolone antibacterials. J Med Chem 2002 May 23;45(11):23384. ACS Publications ) with permission
67 Figure 3 2. BENX disrupts the interaction between B subunit of V ATPase and microfilaments Vma2p Vma2pMBP microfilaments ( 1 sence of vehicle, 100 or 25 BENX in actin polymerizing buffer plus 10 polymerization. Samples were subj ected to ultracentrifugation. The pellet (P) and supernatant (S) fractions were collected, subjected to SDS PAGE, and stained with Coo massie Brilliant Blue BENX induced a shift of B subunit from the pellet to the soluble fraction. Densitometry was performed and the amount (in absorbance units) of Vma2pMBP pelleting in the presence of inhibitor was determined.
68 Figure 3 3 BENX reduces the number of TRAP+ giant cells with IC50 similar to enoxacin. A) Bar graph. R aw 264.7 cells were stimulated with RANKL (50ng/mL) and treated with vehicle, 5, 10, 25, 50, or 100 M BENX. T he number of TRAP+ cells was quantified and categorized as mono, multi nuclear (2 10 nuclei) or giant cells (> 10 nuclei) B) Representative images of osteoclasts cultured with vehicle, 25 M or 100 M BENX. The average number of TRAP+ cells in the vehicle treated cultures was defined a s 100%. Significance was determined by onew ay ANOVA followed by Bonferroni post test *p < 0.05 was c onsidered significant. The number of TRAP+ cells in BENX treated cultures was compared to vehicletreated cells and depicted as a perc entage of the vehicle controls. N= 3 for each condition.
69 Figure 3 4 BENX alters proteolytic cleavage of TRAP5b but not the protein expression of V ATPase or osteoclast related proteins A) RAW 264.7 cells were cultured for 5 days with 5 ng/ml of RANKL and treated with vehicle or 50 M BENX. The proteins from whole cell extracts were c ollected in equal volumes of SDS PAGE sample buffer, loaded equivalently on gels, separated by gel electrophoresis and an alyzed by western blot using antibodies against the indicated proteins. B) Whole cell extracts from RAW 264.7 cells treated with RANKL in the presen prepared, separated by SDS PAGE, blotted and probed with anti TRAP antibody. Most of t he active form of TRAP (16kD) is mainly expressed in the cells treated with control vehicle Treatment with BENX led to a decrease in the 16kD band and a simultaneous increase in a series of higher molecular weight bands up to the fulllength of TRAP ( 38kD band) N=3 by de nsitometry of blots
70 Figure 3 5 BENX triggers higher levels of apoptosis in RANKLstimulated cultures Quantitative analysis of caspase3 activity in Raw 264.7 cells plus or minus RANKL were treated with vehicle or 50 M BENX for 24, 48, and, 72 h. Caspase3 assays were performed following the manu facturer's instructions (cat # APT131, Millipore, Te mecula, CA). Caspase3 activity was measured by a colorimetric reaction and quantified using a BioTek KC4 spectrophotometer (Winooski, VT) at 405 nm. Units are defined as the amount of enzyme that cleaves 1 nM colorimetric substrate per hour. BENX induce d unstimulated RAW 264.7 cells to undergo apoptosis, and triggered higher levels of apoptosis in RANKLstimul ated cultures after 48 and 72 hours compared with control One way ANOVA; *p<0.05 was considered significant. N=3 for each condition.
71 Figure 3 6 BENX inhibits osteoclast bone resorption in vitro with an IC50 in the low micromolar range. A) Representative s canning electron micrographs of bone slices resorbed by osteoclasts in the presence of vehicle, 1, 10, or 3 0 M BENX B) T abulation of the percent of the control amount of pit numbers, area per pit, and percent of total area resorbed. The average number of all values ( percentage resorbed/total area, number of pits, and are a/pit) in the vehicle treated cultures was defined as 100%. The average numbers for the vehicle control cultures were: 3 for percent resorbed/total area, 18 for number of pits, and 440 for area/pit. Cultures treated with BENX were compared to vehicletreated cells and depicted as a perc entage of the vehicle controls. Bone sur face area=1cm2. Bar = 200 m. Significance was determined by oneway ANOVA, *p < 0.05 was considered significant. N=4 for each condition.
72 Figure 3 7 BENX binds to b one at a saturation level of 0.8 mg /ml /cm1. BENX was diluted in a volume = 1ml diH2O at different concentrations (1 M 10mM). A 1 cm2 bone slice was incubated for 24 hours at 37oC in each solution. The amount of BENX present in solution after the 24hour incubation was measured by spectrophotometry (260nm) and plotted into a saturation cur ve of BENX
73 Figure 3 8 BENX inhibits bone resorption more effectively than ENX by predictably binding to the bone surface. A ) T abulation of the percent of the control amount of pit numbers, area per pit, and percent of total area resorbed. Mature osteoclasts derived from mixed cultures were loaded atop bone slices and treated with vehicle or various concentrations of B ENX The resorption pits on the bone slices were analyzed by scanning electron microscopy (SEM ). The average number for each value in the vehicletreated cultures was defined as 100%. BENX treated cultures were compared to vehicletreated cells and depicted as a perc entage of the vehicle controls. The amount of BENX bound to bone or in solution (unbound) for each concentration was predicted based on the saturation graph (Figure 37). B) Representative sections from the scanning electron micrographs of bone slices resorbed by osteoclasts in the presence of vehicle, 1, 10, or 100 M BENX. Pits were defined as resorbed areas with c ontinuous scalloped outline Bone surface area=1cm2. Bar = 200 m. One way ANOVA was used to determine differences among groups. p < 0.05 was considered significant N=4 for each condition.
74 Figure 3 9 BENX inhibits bone resorption by a different mechanism than ALN A ) T abulation of the percent of the control amount of pit numbers, area per pit, and percent of total area resorbed. Mature osteoclasts derived from mixed cultures were loaded atop bone slices that were precoated with either alendronate (ALN) or BENX. The resorption pits on the bone slices were analyzed by scanning electron microscopy (SEM). The average number for each value in the vehicletreated cultures was defined as 100%. The values from BENXor ALN treated cultures were compared to vehicletreated cells and depicted as a perc entage of the vehicle controls. B) Representative sections from the scanning electron micrographs of bone slices pre coated in solutions with vehicle, 1m M ALN, or 1mM BENX and resorbed by osteoc lasts Pits were defined as resorbed areas with a continuous scalloped outline. Bone surface area=1cm2. Bar = 200 m. One way ANOVA was used to determine differences among groups. *p < 0.05 was considered significant N=4 for each condition.
75 CHAPTER 4 EFFECTS OF BE N X ON ORTHODONTIC TOOTH MOVEMENT Introductory Remarks Orthodontic tooth movement (OTM) requires alterations in bone remodeling that are temporarily and spatially regulated to facilitate specific displacement of teeth through alveolar bone (185) When mechanical stress is applied to a tooth dur ing OTM the alveolar bone and periodontal ligament are compressed on one side and st retched on the opposite side. This mechanical stimulus is sensed by the bone cells and triggers bone remodeling which results in tooth displacement. Orthodontic tooth movement occurs in three stages: 1) cell activation (tipping phase) 2) cell recruitment and initiation of bone resorption ( lag phase) and 3) tooth movement across alveolar bone ( post lag phase) (186) Recent studies performed in rats have shown that it is possible to pharmacologically manipulate the OTM process (127130) A number of pharmacologic agents such as virally introduced RANKL, recombinant oste oprotegerin, matrix metalloprotease inhibitors, and integrin inhibitors have been used to either accelerate or reduce OTM (128130,169) Although in vivo studies suggest the efficacy of these molecules during OTM, they are too likely to have unknown side effects to be considered practical for standard orthod ontic treatment. Anti osteoclastic drugs like alendronate and zolendronate are effective inhibitors of tooth movement in vivo (127,187) They act as potent suppressors of osteoclast s and are currently the major class of drugs used for the treatment of bone diseases such as osteoporosis and bone cancer (131) However, an increasing number of
76 bisphosphonateinduced oral osteonecrosis (BIJON) cases have made these drugs problematical for the treatment of excess bone resor ption in oral tissues (188 190) Thus, i t becomes apparent that the development of new therapeutic approaches is necessary to target osteoclas t activity mo re effectively. In the previous chapters we have shown that both enoxacin and BENX inhibit the V ATPase/microfilament interaction and selectively inhibit osteoclast formation and bone resorption. We also showed that BENX maintains many of the anti osteoge nic effects of enoxacin but targeted to bone and is more potent as an anti resorptive agent We believe that our molecular approach may serve as a convenient and less intrusive means for controlling tooth movement during orthodontic treatment and even trea ting bonemediated disease in general In this chapter we used an in vivo model for OTM and followed tooth movement kinetics in the presence of BENX. Materials and Methods Orthodontic Tooth Movement Model as D escribed by GJ King The protocol require d two preparatory sessions as described by King et al (191) I n the first session, modified cleats w e re bonded onto the maxillary molars and all four incisors pinned to prevent further eruption ( Figure 4 1) It also required the extraction of both m andibular first molars T he animals were then allowed to reco ver for a period of 2 weeks During the second session, a coiled spring w as ligated to the molar cleat and bonded to the anter ior incisor with 40g of force ( Figure 4 1 A) The orthodontic forces were applied for 14 days Cephalometric x rays were taken pri or to activation and on days 0, 1, 3, 5, 7, 10, and 14 using a headholding device. The cephalograms were digitized and analyzed using Image J software ( Figure 4 1 B) Tooth movement was
77 quantified by measuring the distance from the cleat to the distal end of the pin on the incisor. O rthodontic T ooth M ovement Model M odified by JK Neubert Our rat OTM model is a modification from a previously described model by King et al (191) Young male SpragueDawley rats, 40 50 days old (Charles River Laboratories) were purchased and allowed to acclimatize to our facilities for two weeks using a normal day/night cycle and given food and water ad libitum. Each rat was weighed prior to all procedures and anesthetized using isofluorane. Their body temperature was stabilized using an isothermal pad during the orthodontic activation procedure. All animals were closely monitored post operatively for weight gain. The orthodontic appliance activation required a 26 mm section of a round (0.014inch diameter) Australian orthodontic archw ire was cut and a 90degree bend was incorporated at one end of the segment. The resulting archwire was Lshaped with a short arm of approximately 5mm and a long arm of 21mm. The wire was then bonded to the buccal surface of the incisors. The long arm was resting passively on the palatal side of the r ight first molar. To activate the orthodontic wire, the long arm was moved over the occlusal surface of the right first molar and bonded onto the buccal side of the same tooth ( Figure 4 2 ). Activation of the wi re generated approximately 16g of force. The force vector generated from the wire was expected to tip the molar in a palatal direction. Polyvinyl siloxane dental impressions were taken at 7 day interval for a period of 28 days The impressions where then poured in epoxy (Cat No 40200029, Struers, Cleveland, OH) and an electronic image of each model was taken. The distances between pre set landmarks on the teeth where then analyzed using Image J software.
78 Statistics Count e rs were precalibrated for their abi lity to identify the maxillary first molar and measure the distance between the lingual surfaces of the mesial cusp of contralateral first molars. Counters were blinded to the treatment groups. Based on the results of a power calculation, a sample size of 12 rats per group (N=12 per time point) is adequate to achieve statistical significance for epoxy cast measurements. The parameters used for the power calculation include significance p value <0.05, and a desir ed statistical power of 80%. D igit al images of the casts were taken and analyzed using Image J software. The data for tooth movement and drift was expressed as means SEM. Samples were compared by oneway ANOVA followed by Bonferroni post test using the program GraphPad Prism 5 (GraphPad Software, La Jolla, CA). Results BENX Significantly Inhibits Orthodontic Tooth Movement To determine if BENX is an effective inhibitor of osteoclast resorption, we initially studied the effect of BENX on a rat OTM model as originally described by King et al (191) However, a number of difficulties were encountered during the application of this model. The animals showed significant weight loss and moderate pain soon after activation of the orthodontic closed coil. To stabilize the animals supportive care was provided including soft ground chow and medication for pain management. Most animals returned to baseline weights 7 days after activation but did not continue to gain weight as expected ( Figure 4 3). We reasoned that the pain and discomfort experienced by the animals was the result of inflammation from the surgically implanted pins. The fact that animals were not gaining weight after surgery could possibly be due to the interference of the active coil while chewing or to the discomfort from the tipping molar.
79 The potential effects that the pain and inflammation could have on osteoclasts and subsequently in tooth movement prompted us to develop a modified method to test OTM. We then made a new OTM model developed by Dr. John K. Neubert which in turn was developed from a mouse model developed by the King group. Unlike the previous model, rats in which the modified model was used displayed no loss of weight (Fig ure 4 4) or signs of discomfort. Preliminary studies suggest that ample too th movement could be achieved to test BENX (Neubert et al manuscript i n Prep.). To test whether BENX affected OTM, we injected the animals with either BENX (50mg/Kg/day) ALN as a positive control, ( 1.5mg/Kg/day ) (127) or PBS ( ml/Kg ) and followed the kinetics of OTM using the modified Rat OTM model ( Figure 4 5 ) D aily subcutaneous (SC) injections of BENX were given for two weeks prior to activation o f the orthodontic appliance (loading phase) and continued throughout the 28day activation period. The positive and negative control groups received daily SC injections of alendronate and 1X PBS solution respectively A continuous gain in weight was obser ved during the experimental procedure indicating that the animals tolerated the modifications that we made to the OTM model very well ( Figure 4 4) All three groups exhi bited the initial tipping and lag phases. No significant differences in either the tipping phase or lag phase were noted among the study groups The tooth movement in the control group at day 28 was significantly greater compared to the other groups (p<0.05). Although the ALN group displayed less tooth movement (0.1mm) than the BENX group ( 0.2mm), no significant difference in tooth movement was detected between the two groups ( Figure 4 5 ).
80 Summary of Results The kinetics of orthodontic tooth movement (activation, lag and post lag phases) was observed in both controls and experimental gr oups. The PBS group displayed a significant increas e in tooth movement after day 28 when compared to the other two groups. Both ALN and BENX inhibited OTM significantly (p<0.05) during the post lag phase. Discussion In this s tudy we show for the first time that a bisphosphonatederivative of enoxacin which has anti resorptive activity in vitro also inhibits OTM, an osteoclast dependent process, in a rat model. These data confirm that in addition to inhibiting osteoclast activi ty in cell culture, BENX also has biological effects in vivo that are consistent with the ability to inhibit osteoclast ic bone resorption. Enoxacin was identified by conducting a virtual screen to identify small molecules predicted to bind the actin binding surface of the B2subu nit, and thus block its interaction with microfilaments. Our studies have shown that enoxacin blocks osteoclast differentiation and bone resorption in tissue culture (chapter 2) In order to find a delivery method that would target enoxacin specifically t o bone, we identified a bisphosphonatelinked derivative of enoxacin (BENX). To date, our data shows that BENX retains the same anti osteoclastic activity as enoxacin and it effectively inhibits bone resorption in vitro afte r it is bound to bone (chapter 3). Current pharmacologic approaches against bone resorption involve the use of selective estrogen receptor modulators (SERMs), calcitonin, bisphosphonates, and more recently the use of a monoclonal antibody against RANKL (Denosumab). However, patients t aking these medications fall at a greater risk of experiencing adverse
81 eff ects such as thromboembolism (192) or osteonecrosis of the jaw (193) We suggest that the parent molecule ENX or BENX represent a novel class of an anti osteoclast selective agent with well characterized molecular mechanism (117,154) that may prove useful for controlled tooth movement during orthodontic therapy. The m odified OTM model presents numerous advantages over the original model. First, the orthodontic appliance is significantly simpler and can be activated in one session as opposed to the two sessions required for the original model. This shortens the surgical and handling time, and reduces the stress to which the animals are exposed to during the activation session. Second, the active appliance is bonded to the buccal surface (side surface) of the molar thus extraction of the mandibular molars and pinning of the anterior teeth become unnecessary. Third, the measurement of tooth movement does not require radiographic analysis thus surgically implanted pins on the palate can be avoided and also the exposure of the animal and investigator to radiation is decreased. Our data showed that BENX significantly inhibits OTM in a rat model when compared to vehicle after 28 days o f activation ( Figure 4 5 ) All experimental groups showed the three phases of OTM. However, the lag phase observed in our model lasted approximately twice as long (14 days) as ex pected (7 days). In orthodontics, effective tooth movement can be achieved with light continuous forces of approximately 40g; however, the orthodontic forces applied by the wire i n our OTM model are approximately 15 grams. We concluded that such light forc es in our model although continuous, may have not been sufficient for cell recruitment to occur effic iently The post lag phase was significantly reduced by both BENX and ALN on day 28. The fact
82 that BENX reduced the post lag phase as effectively as ALN during OTM suggests that the selective inhibition of osteoclast activity by blocking the interaction between V ATPase and microfilaments may serve as a potential therapeutic target for orthodontic applications We suggest that BENX is a novel bone targeted anti osteoclastic agent that functions by a different mechanism from anti resorptives currently used in the clinic and may prove to be suitable for dental applications.
83 Figure 4 1. Orthodontic tooth movement in a rat model. The rat OTM m odel as described by King et al (191) involves bondi ng modified cleats onto the maxillary molars and all four incisors pinned to alveolar bone to prevent further eruption. Mandibular first molars are extracted and the animals allowed recovering for 3 wks. A) A NickelTitanium closed coiled spring w as ligated to the molar cleat and bonded to the anter ior incis or with 40g of force. The orthodontic appliance remained active for 14 days B) Cephalometric x rays were taken at different time points using a headholding device. The cephalograms were digitized and analyzed using Image J software. Tooth movement was quantified by measuring the distance from the cleat to the distal end of the pin on the incisor. Copied from ( King GJ Keeling SD McCoy EA, Ward TH Measuring dental drift and orthodontic tooth movement in response to various initial forces in adult rats. Am J Orthod Dentofacial Orthop. 1991 May;99(5):45665. Elsevier Publications ) with permission.
84 Figure 4 2 M odified rat orthodontic tooth movement model. A ) schematic description of the modified rat orthodontic tooth movement model (adapted from Chu et al (194) ; B) active orthodontic appliance. Briefly, a 26mm section of a round (0.014inch diameter) Australian orthodontic archwire was cut and a 90degree bend incorporated at one end of the segme nt. The resulting archwire was Lshaped with a short arm of approximately 5mm and a long arm of 21mm. The short arm of the wire was then bonded to the buccal surface of the incisors. The long arm rests passively on the palatal side of the right first molar During activation, the long arm was moved over the occlusal surface of the right first molar and bonded onto the buccal side of the same tooth. The orthodontic appliance remained active for 28 days. Maxillary dental impressions were taken at different ti me points using polyvinyl siloxane (PVS) material and epoxy casts of the impressions made. Digital images of the casts were taken and tooth movement measured using Image J software. Panel A adapted from ( Chu S Ishikawa H Kim T Yoshida S Analysis of scar tissue distribution on rat palates: a laser Doppler flowmetric study. Cleft Palate Craniofac J. 2000 Sep;37(5):48896) with permission.
85 Figure 4 3. Weight monitoring of rats in the control group with original OTM model. Rats were weighted on days 0, 7, and 14 after activation. Each color represents a single animal. All animals lost 10% total body weight 1 day after appliance activation. Most rats returned to baseline weight by day 7 but did not cont inue to gain weight by day 14.
86 Figure 4 4. Weight monitoring of rats treated with the modified OTM model. The animals were divided into three treatment groups. Daily subcutaneous (SC) injections of BENX (50mg/Kg/day), alendronate (ALN) (1.5mg/Kg/day) (127) or P BS (ml/Kg) were given throughout the activation period. A bsolute weight values were recorded from all animals on days 0, 7, and 14. An average of the weight values was calculated for each group at every timepoint. N=12 animals per group.
87 Figure 4 5 BENX reduces OTM in rats after 28day activation period. Young male SpragueDawley rats were distributed into three groups according to treatment. Daily subcutaneous (SC) injections of BENX ( 50mg/Kg/day ) alendr onate ( 1.5mg/Kg/day ) or weight /vol 1x PBS were given for two weeks prior to activation of the orthodontic appliance and continued throughout a 28day activation period. Significant differences in tooth movement were observed at day 28 between control and the ALN and BENX groups (p<0.05). N o significant di fference in tooth movement was detected between the ALN (0.1mm) and BENX ( 0.2mm) groups S i gnificance was determined by oneway ANOVA. N=12 for each group. p < 0.05 was considered significant. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0 activation 7 14 21 28 tooth movement (mm) days PBS ALN BENX
88 CHAPTER 5 SUMMARY, CONCLUSIONS AND FUTURE DIRECTIONS Summary and C onclusions Bone remodeling is a physiologic process important for the removal of mechanical stressre lated microfractures, and regulation of the systemic mineral homeostasis. It requires precise control between bone formation by osteoblasts and bone resorption by osteoclasts (195) Bone resorption is an important element of many physiologic processes in the mouth such as tooth eruption and a significant complication in clinical dentistry. During bone resorption osteoclasts becom e highly polarized and form the resorptive organ known as ruffled membrane surrounded by a ring of filamentous actin and associated proteins (87,196,197) In actively resorbing osteoclasts, cellular V ATPases are targeted to the ruffled membrane (83) for acidification of the extracellular resorptive compartment. A direct interaction between V ATPase and actin is necessary for proper tar geting of the V ATPase to the ruffled membrane. Earlier studies have shown that a binding interaction between subunit B2 of the V ATPase and F actin is important for bone resorption (70,86,88,90,94) By using computational che mistry analys is, we identified enoxacin, a member of the fluoroquinolone antibiotics, as a candidate small molecule that may interact with the actin binding site on subunit B2. In v itro assays demonstrate that enoxacin blocks the interaction between V ATPa se and microfilaments and inhibits osteoclast formation and activity while osteoblast formation remained unaffected (117) To date, our data supports the idea of a selective inhibi tion of osteoclast activity by blocking the interaction between V ATPase and microfilaments.
89 The scientific work presented in this dissertation was designed and executed to examine the mechanisms by which enoxacin inhibits osteoclast formation and bone re sorption in greater detail and to determine whether enoxacin holds promise as a drug for use in various medical and dental applications in which inhibition of osteoclast mediated bone resorption would be beneficial. To study the proposed molecular mechanisms by which inhibition of the V ATPase B2 actin interaction may be affecting osteoclast differentiation and activation in vitro we used RAW 264.7 an osteoclast like cell line, and treated them with enoxacin. We foun d that e noxacin reduces the expression of TRAP activity and cell fusion with an IC50 of similar to the previously reported observations done on mousemarrow co cultures demonst rating that it acts directly on osteoclasts (117) Vacuolar H+ ATPases are normally expressed in all eukaryotic cells at very low levels however; osteoclasts express high levels of this enzym e to carry out their specializ ed functions. In our study we showed that enoxacin did not affect t he protein levels of several V ATPase subunits This may be indicative that the reduction in osteoclast maturation and bone resorption is the effect o f enoxacin on post transla tional mechanisms A reduction in the amount of the active form of TRAP and LPlastin was observed after treatment of osteoclasts with enoxacin. Both LPlastin and TRAP are involved in many biological processes during osteoclast activity (158,198) T he mechanism s by which enoxacin r educes TRAP activity and actin ring formation are not clearly understood. However, w e reasoned that enoxacin is affecting the activation of TRAP by
90 interfering with the trafficking of V ATPases to TRAP/cathepsincontaining vesicles thus preventing acidific ation and subsequent activation of these enzymes. L Plastin is present in the core of podosomes and actin rings serving as an actin bundling protein (165) The formation of actin rings as well as the ability of osteoclasts to resorb bone has been linked directly to the r egulation o f L plastin levels (158) A change in Lplastin is consistent with reduction in the formation of actin rings detected previously in response to enoxacin (117) The reduction observed in the 67KDa LPlastin and corresponding smaller fragment of the molecule (57KDa) suggest that L plastin is likely proteolytically cleave d so that the EF hand domain on the amino terminal is removed in response to enoxacin. The EF hand binds ca lcium and prevents binding of L plastin to microfilaments in the presence of calcium (166 168) Cells treated with enoxacin show alterations in the sorting of DC STAMP ( Appendix A ) DCSTAMP is preferentially expressed in osteoclasts and an essential cellcell fusion regulator during os teoclastogenesis (160) Because DC STAMP is known to be involved in the fusion of osteoclast precursors (160163) the change in cell surface expression could be lin ked functionally to the reduction in multinuclear and giant cells detected in enoxacintreated cultures. This represents direct evidence that enoxacin may be perturbing a vesicle sorting pathway in osteoclasts In support of our proposed mechanism a rece nt study by Brown et al suggest that V ATPases may serve as vesicle coating proteins, like clathrin, to regulate vesicular trafficking (199) The V ATPases are able to carry out this function by interacting with the small GTPase ARF6 and its activator ARNO via the actinbinding activity of B2 to modulate vesicular trafficking and cytoskeletal reorganiz ation (96) This hypothesis
91 implies that the V ATPase microfilament i nteraction could be a vital for the integration of mult iple regulatory pathways and could explain the diverse, but selective, effects enoxacin has on osteoclasts. Another piece of evidence comes from studies done in Dicytostelium where they report that V ATPases work in coordination with a complex compose d of ARP 2/3 and WASH proteins They show that the vesicular sorting in Dicytostelium is dependent upon the direct interaction of V ATPase with the ARP2/3WASH complex (109) Because WASH and the actinbinding site in the B subunit are both evolut ionarily conserved, a mammalian adaptation of this sorting mechanism could possibly allow cells like osteoclasts, to carry out thei r specialized functions. Our lead molecule enoxacin is an antibiotic; therefore, systemic administration would seem impractical due to the potential adverse effects it may pose and the possibility of not achieving high enough levels to have an effect on osteoclasts at the target tissues. To overcome these limitations we modified the pharmacokinetic properties of enoxacin by adding a bisphosphonat e moiety to its structure Our study showed that BENX binds to bone and inhibits osteoclast ogenesis and bone resorption in a way that resembles the novel mechanism by which enoxacin functions. However, when bone is present BENX is 10 fold more effective at inhibiting osteoclast formation and resorption than the parent molecule. Importantly, its anti resorptive act ivity is achieved by a different mechanism from current anti resorptives. It will be of interest to determine whether the unique mec hanism of action of BENX confers advantages over existing therapeutic agents in the treatment of osteoclast mediated disease.
92 To study the effect of BENX on osteoclasts in vivo, we used an orthodontic tooth movement (OTM) model in rats modified by Dr. JK Neubert from a model originally described for mice. The us e of this modified rat OTM model represents a novel way to test po tential anti resorptive agents. A number of advantages were derived from this model such as shortening surgical time, eliminates the need for potentially painful procedures, and is better tolerated by the animals. It also is a novel method to quantitativel y assess tooth movement in response to a defined force. Our data shows for the first time that BENX which has anti resorptive activity in vitro also inhibits OTM in a rat model We believe that our molecular approach may serve as a convenient and less intrusive means for controlling tooth movement during orthodontic treatment and even treating bone mediated disease in general. In summary, enoxacin directly inhibits osteoclasts. It reduced the amount of the B2 subunit bound to the detergent insoluble c ytoskeletal fraction, altered the transport of DC STAMP to the plasma membrane, and blocked proteolytic activation of TRAP5b from its latent pro enzyme to the active form. Enoxacin triggered a reduction in the size of L plastin, likely due to proteolytic cleavage of the amino terminal calcium binding EF hand domain. We propose that enoxacin selectively affects elements of the osteoclast differentiation and activation program that are downstream of the binding i nteraction between V ATPases and microfilaments, which is mediated by the actinbinding site on the B2subunit.
93 Future Directions Examine the Effects of Enoxacin on MicroRNA Activity and Gene Expression in Osteoclasts It has been shown that enoxacin is a stimulator of RNAi and microRNA activity (200) R ecent findings by Melo and colleagues confirm that enoxacin enhances the production of miRNAs. Moreover, they show that enoxacin inhibit tumor growth and metastasis by binding to the miRNA biosynthesis protein TAR RNA binding protein 2 (TARP2) (159) However, the mechanism by which binding of enoxacin to TARP2 enhances microRNA processing is not known. Based on these studies it will be important to determine whether enhancement of microRNAs has a role in the regulation of osteoclasts. This should be possible by studying in greater detail the effects of pefloxacin and Binhib16, molecules that inhibit V ATPase/microfilament interaction wit hout stimulating microRNA s. Con sistent with the previous data, enoxacin increased GW bodies (an indicator of increases in overall microRNA activity) at concentrations of enoxacin starting at 100 M (Edward K.L. Chan and L. Shannon Holliday, unp ublished d ata ). However, pefloxacin (117) and another nonfluoroquinolone (Binhib 16; unpublished data) have similar inhibitory effects in osteoclasts but they do not hav e microRNA stimulating activity To date, our data does not seem to support the idea that the inhibition of osteoclasts by enoxacin is derived from effects on microRNAs However, because of the exciting recent findings showing enoxacin to be an inhibitor of tumor growth and m etastasis, it will be vital to examine the effects of enoxacin, pefloxacin and Binhib16 on microRNA activity and gene expression in osteoclasts in greater detail.
94 Determine Whether the Inhibition of t he I nteraction between V ATPase and Microfilaments by Enoxacin i s Important f or Its Anti Tumorigenic Activity In the study by Melo and colleagues (159) it was not reported whether inhibition of the B2microfilament interaction played any role in the response of cancer cells to enoxacin. This should be examined in the future given that plasma membrane V ATPases have been reported in metastatic tumor cells (201) It is therefore plausible that disruption of the interaction between B2 and microfilaments by enoxacin plays a role in the inhibition of cancer by blocking V ATPase microfilament binding and the transport of V ATPases to the plasma membrane. Within the context of cancer, it is also of interest that enoxacin disrupted actin rings in osteoclasts (117) As described earlier, actin rings are composed of structures called podosomes, which are similar or identical to invadopodia in metastatic cancer cells that are important for tissue invasion (202) It will be of interest to determine whether enoxacin inhibits the formation of podosomes in cancer cells. Determine the Effect o f BENX on Osteonecrosis of t he Jaw Unlike enoxacin, BENX enhances osteoclast apoptosis. This may be a necessary result of the inclusion of the bisphosphonate backbone to target an active moi e ty to bone. Although we do not fully understand the underlying mechanism, we suspect that like simple bisphosphonates, BENX may function as a nonhydrolyzable ATP analog. The fact that it induces apoptosis raises concerns that, like other bisphosphonates and Denosumab, could also lead to increased risk of oral osteonecrosis. However, because oral osteonecrosis is linked to bacterial infection, the antibiotic activity of enoxacin may reduce or eliminate the risk or oral osteonecrosis. The recent identification of the rice rat
95 as a model for studying ONJ (203) provides a potential mean for studying whether BENX is associated with ONJ. Determine Alternative Mechanisms for Local Delivery of Enoxacin o r BENX Although the bisphosphonate backbone linkage provides a convenient method for delivering enoxacin to bone, other approaches for local administration of enoxacin, which do not trigger osteoclast apoptosis, may prove better suited for oral therapeutic use. For example, local delivery of enoxacin using inert or biodegradable sustained release devices may allow the local delivery of enoxacin (or BENX ) required for orthodontic purposes, without the potential risks associated with the bisphosphonate effects. The ethylenevinyl acetate copolymer (ELVAX; DuPont, Wilmington, DE) is a biologically inactive polymer which has been used in the past by our group to deliver biologically active substances over time to the gingival tissue (129,130) ELVAX can be used as a platform for the sustainedrelease of enoxacin into the local microenvironment. Preliminary studies were done to determine if enoxacin can become incorporated into ELVAX. The preparation of the polymer was made in accordance with the protocol published by Langer and Folkman (204) An important advantage of using ELVAX is that its release kinetics can be manipulated by altering certain variables during its preparation (i.e. increasing water loading). When undiluted enoxacin was incorporated into ELVAX, approximately 10% of the total molecule was released by day 30 ( Figure 5 1 A ). In contras t, when diluted enoxacin was used for the preparation of ELVAX enoxacin almost 100% of the molecule was released by day 25 ( Figure 5 1 B). Our data show that enoxacin can be incorporated into the polymer and perhaps more
96 importantly that ELVAX can be mani pulated for the localized sustained release of enoxacin. By surgically in corporating a strip of ELVAX enoxacin into the dental tissues it is possible to mai ntain a sustainedrelease of enoxacin over a two week span or more into neighboring tissues. It will be important to determine the effect of the localized sustained released enoxacin (or BENX) in bone resorption. This sustaineddelivery system would serve as a novel platform for the use of bioactive molecules in the dental clinic. Potential applications include: retention after orthodontic treatment, prevention of alveolar bone loss from periodontal infections, or inhibition of internal/external root resorption. Although there is much work to be done in order to fully elucidate the mechanism by which enoxacin blocks osteoclasts formation and function, I have identified enoxacin a s a novel type of selective inhibitor of osteoclasts. Enoxacin, or other molecules with similar activity, may prove useful for the treatment of osteoporosis and other bone pathol ogies. Because enoxacin is a potent inhibitor of the growth and metastasis of cancers (159) and selectively inhibits osteoclasts (117) it is a particularly attractive candidate to test as a therapeutic agent for the treatment of bone cancer
97 Figure 5 1. ENX can be incorporated into a biologically inert polymer The ELVAX ENX polymer was prepared according to the protocol by Langer and Folkman (204) A 0.07g slice of polymer was incubated in 300 L 1X PBS at 37oC for 24 hrs. After each incubation period, the amount of ENX released from the polymer was measured by spectrophotometry (290nm) and the polymer transferred into fresh PBS. Ten milligrams of enoxacin were incorporated into the polymer. A) Decreased water loading during polymer preparation resulted in only 10% of the total enoxacinrelease after a 3 0 day period whereas; B) increased water loading resulted in 93% of enoxacin released by day 25 from the polymer.
98 APPENDIX ENOXACIN ALT ERS THE SURFACE EXPR ESSION O F DCSTAMP A PROTEIN LINKED T O CELL FUSION The seven transmembrane region receptor DC STAMP is preferentially expressed in osteoclasts after stimulation by RANKL. Osteoclasts from DCSTAMPdeficient mice display lack of fusion, but comparable TRAP activity, formation of ruffled plasma membranes and actin rings indicating that DC STAMP is not involved in osteoclast activation, except for cellular fusion (160) S urface protein expression of DC STAMP was analyzed by western blot and FACS analysis on RANKLstimulated osteoclasts after 5 days of cult ure treated with 50 M enoxacin or vehicle. The immunoblots showed that enoxacin did not affect the overall expression of DC STAMP in the RANKL stimulated cells ( Figure 2 3 ). T hr ee distinct populations of RANK expressing cells were evident on the FACS analysis and categorized into low, medium, and high DC STAM P expressing populations ( Figure S 1 ). Overall, high RANK expressing cells showed a significant increase in DC STAMP expressed on their cell surface. Interestingly, when cells from the highRANKL expr essing population where treated with enoxacin, a significantly greater expression of surface DC STAMP was observed ( Figure S 2 ). Materials and Methods Real Time PCR Total RNA from RAW 264.7 cells was isolated with an RNeasy Mini Kit (Qiagen, Valencia, CA ) and real time quantitative PCR (qPCR) was performed using the TaqMan Gene Expression Master Mix Reagents, and detection System (Applied Biosystems, Foster City, CA) (31). Probes and primers for the mouse genes that encode TRAP5b
99 (Mm00475698_m1), a3subu nit (Mm00469406_m1), Lplastin (Mm01310735_m1), ADAM8 (Mm00545762_m1), B2subunit ( Mm00431987_m1), WASH (Mm 01239734_m1) and hypoxanthine phosphoribosyltransferase (HPRT) (Mm00446968_m1) were obtained from Applied Biosciences. RAW 264.7 cells were treated as indicated in the figure legends, RNA was isolated and assayed on day 5 and the level of each of the genes relative to HPRT was determined. The experiment was analyzed stat istically using t tests of the Ct values. P values < 0.05 were consider ed significant. Flow Cytometry Cells were allowed to dissociate from the bottom of UpCell coated plates (Thermo Scientific, Rochester, NY) at room temperature for 3045 minutes. Suspended cells were washed with FACS Buffer [1x PBS + 5% FBS + 0.372g EDTA] and allowed to incubate with 1:200 dilution of antibodies for 1hr at 4oC in the dark, biotinconjugated rat anti mouse RANK (eBioscience cat #136612 81) with Alexafluor 647conjugated streptavidin (Invitrogen, S21374 ), and rabbit anti mouse DC STAMP (sc 25579 ) with Alexafluor 488conjugated goat anti rabbit IgG (Invitrogen, A11008). Cells were acquired using a FACSCalibur flow cytometer (BD Biosciences) and analyzed using FCS Express (De Novo Software).
100 Figure A 1 Enoxacin has only minor effec ts on mRNA levels of many osteoclast genes Raw 264.7 cells were treated with RANKL and either vehicle o enoxacin. Messenger RNA was prepared and V ATPase and osteoclast selective genes were analyzed by qPCR. The level of each gene relative to HPRT1 was determined. The asterisks represent the relative expression of each mRNA in the enoxacintreated cultures relative to control cultures. Statistical analysis was done using t tests of the Ct values. Copied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D, Bradaschia Correa V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permission.
101 Figure A 2 Enoxacin alters the distribution of DC S T A MP. Bone marrow derived osteoclasts seeded on UpCell plates were differentiated for 6 days with M CSF and RANK L. Osteoclasts were then treated with 50uM enoxacin or vehicle alone for 72 hours. Cells were probed with anti RANK and anti DCSTAMP antibodies a nd acquired using a FACSCalibur flow cytometer. Data was analyzed using FCS Express software. A. Forward versus side scatter of osteoclasts showing gating on live cells (red) with or without enoxacin treatment. B. RANK versus DC STAMP mean fluorescence intensity (MFI) gated on live cells with or without enoxacin treatment. Blue square indicates RANK+ cells used for histograms. C. Histograms showing MFI or each mouse with (red lines) or without (black lines) enoxacin treatment. Gated on RANK+ live cells. D. Data shown as an average of MFI in each gated category: DC STAMP low, DC STAMP, medium, DC STAMP high expression. *p value = 0.0089. Unpaired t test with Welchs correction. n=4 mice per group. Copied from ( Toro EJ, Zuo J, Ostrov DA, Catalfamo D BradaschiaCorrea V, AranaChavez V, Caridad AR, Neubert JK, Wronski TJ, Wallet SM, Holliday LS. Enoxacin directly inhibits osteoclastogenesis without inducing apoptosis. J Biol Chem 2012 May 18;287(21):17894904. Epub 2012 Apr 2) with permission.
102 LIST OF REFERENCES Reference List 1. Beresford, J. N. (1989) Clin. Orthop. Relat Res. 270 280 2. Roodman, G. D. (2006) Ann. N. Y. Acad. Sci. 1068, 100109 3. Villa, A., Gue rrini, M. M., Cassani, B., Pangrazio, A., and Sobacchi, C. (2009) Calcif. Tissue Int. 84, 1 12 4. Mundy, G. R. (2000) Adv. Drug Deliv. Rev. 42, 165173 5. Sato, K. and Takayanagi, H. (2006) Curr. Opin. Rheumatol. 18, 419 426 6. Schett, G. (2007) Arthritis Res. Ther. 9, 203 7. Bartold, P. M., Cantley, M. D., and Haynes, D. R. (2010) Periodontol. 2000. 53, 55 69 8. Teitelbaum, S. L. (2007) Am. J. Pathol. 170, 427 435 9. Kong, Y. Y., Yoshida, H., Sarosi, I., Tan, H. L., Timms, E., Capparelli, C., Morony, S., Oliveira dos Santos, A. J., Van, G., Itie, A., Khoo, W., Wakeham, A., Dunstan, C. R., Lacey, D. L., Mak, T. W., Boyle, W. J., and Penninger, J. M. (1999) Nature 397, 315 323 10. Hsu, H., Lacey, D. L., Dunstan, C. R., Solovyev, I., Colombero, A ., Timms, E., Tan, H. L., Elliott, G., Kelley, M. J., Sarosi, I., Wang, L., Xia, X. Z., Elliott, R., Chiu, L., Black, T., Scully, S., Capparelli, C., Morony, S., Shimamoto, G., Bass, M. B., and Boyle, W. J. (1999) Proc. Natl. Acad. Sci. U. S. A 96, 354035 45 11. Bucay, N., Sarosi, I., Dunstan, C. R., Morony, S., Tarpley, J., Capparelli, C., Scully, S., Tan, H. L., Xu, W., Lacey, D. L., Boyle, W. J., and Simonet, W. S. (1998) Genes Dev. 12, 12601268 12. Li, J., Sarosi, I., Yan, X. Q., Morony, S., Capparel li, C., Tan, H. L., McCabe, S., Elliott, R., Scully, S., Van, G., Kaufman, S., Juan, S. C., Sun, Y., Tarpley, J., Martin, L., Christensen, K., McCabe, J., Kostenuik, P., Hsu, H., Fletcher, F., Dunstan, C. R., Lacey, D. L., and Boyle, W. J. (2000) Proc. Nat l. Acad. Sci. U. S. A 97, 15661571 13. Lorenzo, J. A., Sousa, S. L., Fonseca, J. M., Hock, J. M., and Medlock, E. S. (1987) J. Clin. Invest 80, 160164
103 14. Matsuo, K. and Irie, N. (2008) Arch. Biochem. Biophys. 473, 201209 15. Simonet, W. S., Lacey, D. L., Dunstan, C. R., Kelley, M., Chang, M. S., Luthy, R., Nguyen, H. Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H. L., Trail, G., Sullivan, J., Davy, E., Bucay, N., Renshaw Gegg, L., Hughes, T. M., Hill, D., Pattison, W., Campbell, P., Sander, S., Van, G., Tarpley, J., Derby, P., Lee, R., and Boyle, W. J. (1997) Cell 89, 309319 16. Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y. X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W. J. (1998) Cell 93, 165176 17. Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Mochizuki, S., Tomoyasu, A., Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998) Proc. Natl. Acad. Sci. U. S. A 95, 3 597 3602 18. Naito, M., Hayashi, S., Yoshida, H., Nishikawa, S., Shultz, L. D., and Takahashi, K. (1991) Am. J. Pathol. 139, 657667 19. Tondravi, M. M., McKercher, S. R., Anderson, K., Erdmann, J. M., Quiroz, M., Maki, R., and Teitelbaum, S. L. (1997) N ature 386, 8184 20. Zhang, D. E., Hetherington, C. J., Chen, H. M., and Tenen, D. G. (1994) Mol. Cell Biol. 14, 373 381 21. Watanabe, H., Tatsumi, K., Yokoi, H., Higuchi, T., Iwai, M., Fukuoka, M., Fujiwara, H., Fujita, K., Nakayama, H., Mori, T., and F ujita, J. (1997) Biol. Reprod. 57, 13941400 22. Okada, S., Wang, Z. Q., Grigoriadis, A. E., Wagner, E. F., and von, R. T. (1994) Mol. Cell Biol. 14, 382 390 23. David, J. P., Sabapathy, K., Hoffmann, O., Idarraga, M. H., and Wagner, E. F. (2002) J. Cell Sci. 115, 43174325 24. Lamothe, B., Besse, A., Campos, A. D., Webster, W. K., Wu, H., and Darnay, B. G. (2007) J. Biol. Chem. 282, 41024112 25. Takayanagi, H., Kim, S., Koga, T., Nishina, H., Isshiki, M., Yoshida, H., Saiura, A., Isobe, M., Yokochi, T ., Inoue, J., Wagner, E. F., Mak, T. W., Kodama, T., and Taniguchi, T. (2002) Dev. Cell 3, 889 901 26. Takayanagi, H. (2007) Ann. N. Y. Acad. Sci. 1116, 227237
104 27. Kim, Y., Sato, K., Asagiri, M., Morita, I., Soma, K., and Takayanagi, H. (2005) J. Biol. Chem. 280, 32905 32913 28. Matsumoto, M., Kogawa, M., Wada, S., Takayanagi, H., Tsujimoto, M., Katayama, S., Hisatake, K., and Nogi, Y. (2004) J. Biol. Chem. 279, 4596945979 29. Crotti, T. N., Sharma, S. M., Fleming, J. D., Flannery, M. R., Ostrowski, M. C., Goldring, S. R., and McHugh, K. P. (2008) J. Cell Physiol 215, 636644 30. McHugh, K. P., Hodivala Dilke, K., Zheng, M. H., Namba, N., Lam, J., Novack, D., Feng, X., Ross, F. P., Hynes, R. O., and Teitelbaum, S. L. (2000) J. Clin. Invest 105, 433 440 31. Horton, M. A. (1997) Int. J. Biochem. Cell Biol. 29, 721 725 32. Horton, M. A. (2001) Proc. Nutr. Soc. 60, 275281 33. Davies, J., Warwick, J., Totty, N., Philp, R., Helfri ch, M., and Horton, M. (1989) J. Cell Biol. 109, 18171826 34. Clover, J., Dodds, R. A., and Gowen, M. (1992) J. Cell Sci. 103 ( Pt 1), 267 271 35. Hynes, R. O. (1987) Cell 48, 549 554 36. Sato, M., Sardana, M. K., Grasser, W. A., Garsky, V. M., Murray, J. M., and Gould, R. J. (1990) J. Cell Biol. 111, 1713 1723 37. Horton, M. A., Taylor, M. L., Arnett, T. R., and Helfrich, M. H. (1991) Exp. Cell Res. 195, 368 375 38. Scott, B. L. (1967) J. Ultrastruct. Res. 19, 417431 39. GONZALES, F. and KARNOVSKY, M. J. (1961) J. Biophys. Biochem. Cytol. 9, 299316 40. HANCOX, N. M. (1946) J. Physiol 105, 6671 41. Dudley, H. R. and Spiro, D. (1961) J. Biophys. Biochem. Cytol. 11, 627649 42. King, G. J. and Holtrop, M. E. (1975) J. Cell Biol. 66, 445 451 43. Saltel, F., Destaing, O., Bard, F., Eichert, D., and Jurdic, P. (2004) Mol. Biol. Cell 15, 52315241 44. Jurdic, P., Saltel, F., Chabadel, A., and Destaing, O. (2006) Eur. J. Cell Biol. 85, 195202 45. Spinardi, L. and Marchisio, P. C. (2006) Eur. J. Cel l Biol. 85, 191 194
105 46. Hurst, I. R., Zuo, J., Jiang, J., and Holliday, L. S. (2004) J. Bone Miner. Res. 19, 499506 47. Bromme, D., Okamoto, K., Wang, B. B., and Biroc, S. (1996) J. Biol. Chem. 271, 21262132 48. Drake, F. H., Dodds, R. A., James, I. E ., Connor, J. R., Debouck, C., Richardson, S., Lee Rykaczewski, E., Coleman, L., Rieman, D., Barthlow, R., Hastings, G., and Gowen, M. (1996) J. Biol. Chem. 271, 1251112516 49. Holliday, L. S., Welgus, H. G., Hanna, J., Lee, B. S., Lu, M., Jeffrey, J. J., and Gluck, S. L. (2003) Calcif. Tissue Int. 72, 206 214 50. Holtrop, M. E. and Raisz, L. G. (1979) Calcif. Tissue Int. 29, 201 205 51. Holtrop, M. E., Cox, K. A., Eilon, G., Simmons, H. A., and Raisz, L. G. (1981) Metab Bone Dis. Relat Res. 3, 123 129 52. Wagner, C. A., Finberg, K. E., Breton, S., Marshansky, V., Brown, D., and Geibel, J. P. (2004) Physiol Rev. 84, 12631314 53. Gluck, S. L., Lee, B. S., Wang, S. P., Underhill, D., Nemoto, J., and Holliday, L. S. (1998) Acta Physiol Scand. Suppl 643, 203 212 54. Pietrement, C., Sun Wada, G. H., Silva, N. D., McKee, M., Marshansky, V., Brown, D., Futai, M., and Breton, S. (2006) Biol. Reprod. 74, 185 194 55. Hinton, A., Sennoune, S. R., Bond, S., Fang, M., Reuveni, M., Sahagian, G. G., Jay, D., Martin ez Zaguilan, R., and Forgac, M. (2009) J. Biol. Chem. 284, 1640016408 56. Hinton, A., Bond, S., and Forgac, M. (2009) Pflugers Arch. 457, 589598 57. Nishi, T. and Forgac, M. (2002) Nat. Rev. Mol. Cell Biol. 3, 94103 58. Saroussi, S. and Nelson, N. (2009) Pflugers Arch. 457, 581587 59. Perzov, N., Padler Karavani, V., Nelson, H., and Nelson, N. (2001) FEBS Lett. 504, 223228 60. Wilkens, S. (2005) Adv. Protein Chem. 71, 345 382 61. Abrahams, J. P., Leslie, A. G., Lutter, R., and Walker, J. E. (1994 ) Nature 370, 621 628 62. Abrahams, J. P., Lutter, R., Todd, R. J., van Raaij, M. J., Leslie, A. G., and Walker, J. E. (1993) EMBO J. 12, 17751780
106 63. Braig, K., Menz, R. I., Montgomery, M. G., Leslie, A. G., and Walker, J. E. (2000) Structure. 8, 567 573 64. Cabezon, E., Montgomery, M. G., Leslie, A. G., and Walker, J. E. (2003) Nat. Struct. Biol. 10, 744 750 65. Ichikawa, N., Chisuwa, N., Tanase, M., and Nakamura, M. (2005) J. Biochem. 138, 201207 66. Yagi, H., Kajiwara, N., Tanaka, H., Tsuki hara, T., Kato Yamada, Y., Yoshida, M., and Akutsu, H. (2007) Proc. Natl. Acad. Sci. U. S. A 104, 11233 11238 67. Gruber, G., Wieczorek, H., Harvey, W. R., and Muller, V. (2001) J. Exp. Biol. 204, 25972605 68. Cipriano, D. J., Wang, Y., Bond, S., Hinton A., Jefferies, K. C., Qi, J., and Forgac, M. (2008) Biochim. Biophys. Acta 1777, 599604 69. Nakamura, I., Takahashi, N., Udagawa, N., Moriyama, Y., Kurokawa, T., Jimi, E., Sasaki, T., and Suda, T. (1997) FEBS Lett. 401, 207212 70. Zuo, J., Jiang, J., Chen, S. H., Vergara, S., Gong, Y., Xue, J., Huang, H., Kaku, M., and Holliday, L. S. (2006) J. Bone Miner. Res. 21, 714 721 71. Manolson, M. F., Wu, B., Proteau, D., Taillon, B. E., Roberts, B. T., Hoyt, M. A., and Jones, E. W. (1994) J. Biol. Chem. 269, 14064 14074 72. Manolson, M. F., Yu, H., Chen, W., Yao, Y., Li, K., Lees, R. L., and Heersche, J. N. (2003) J. Biol. Chem. 278, 4927149278 7 3. Manolson, M. F., Proteau, D., Preston, R. A., Stenbit, A., Roberts, B. T., Hoyt, M. A., Preuss, D., Mulholland, J., Botstein, D., and Jones, E. W. (1992) J. Biol. Chem. 267, 1429414303 74. Toyomura, T., Murata, Y., Yamamoto, A., Oka, T., Sun Wada, G. H., Wada, Y., and Futai, M. (2003) J. Biol. Chem. 278, 22023 22030 75. Li, Y. P., Chen, W., Liang, Y., Li, E., and Stashenko, P. (1999) Nat. Genet. 23, 447451 76. Smith, A. N., Skaug, J., Choate, K. A., Nayir, A., Bakkaloglu, A., Ozen, S., Hulton, S. A., Sanjad, S. A., Al Sabban, E. A., Lifton, R. P., Scherer, S. W., and Karet, F. E. (2000) Nat. Genet. 26, 7175 77. Oka, T., Murata, Y., Namba, M., Yoshimizu, T., Toyomura, T., Yamamoto, A., SunWada, G. H., Hamasaki, N., Wada, Y., and Futai M. (2001) J. Biol. Chem. 276, 4005040054
107 78. Lee, B. S., Holliday, L. S., Ojikutu, B., Krits, I., and Gluck, S. L. (1996) Am. J. Physiol 270, C382C388 79. Serrano, E. M., Ricofort, R. D., Zuo, J., Ochotny, N., Manolson, M. F., and Holliday, L. S. (2009) Biochem. Biophys. Res. Commun. 389, 193 197 80. Sun Wada, G. H., Toyomura, T., Murata, Y., Yamamoto, A., Futai, M., and Wada, Y. (2006) J. Cell Sci. 119, 4531 4540 81. Kornak, U., Schulz, A., Friedrich, W., Uhlhaas, S., Kremens, B., Voit, T., Hasan, C., Bode, U., Jentsch, T. J., and Kubisch, C. (2000) Hum. Mol. Genet. 9, 20592063 82. Scimeca, J. C., Quincey, D., Parrinello, H., Romatet, D., Grosgeorge, J., Gaudray, P., Philip, N., Fischer, A., and Carle, G. F. (2003) Hum. Mutat. 21, 151 157 83. Blair, H. C., Teitelbaum, S. L., Ghiselli, R., and Gluck, S. (1989) Science 245, 855 857 84. Baron, R., Neff, L., Louvard, D., and Courtoy, P. J. (1985) J. Cell Biol. 101, 22102222 85. Lee, B. S., Holliday, L. S., Krits, I., and Gluck, S. L. (1999) J. Bone Miner. Res. 14, 21272136 86. Lee, B. S., Gluck, S. L., and Holliday, L. S. (1999) J. Biol. Chem. 274, 29164 29171 87. Lakkakorpi, P. T. and Vaananen, H. K. ( 1991) J. Bone Miner. Res. 6, 817 826 88. Chen, S. H., Bubb, M. R., Yarmola, E. G., Zuo, J., Jiang, J., Lee, B. S., Lu, M., Gluck, S. L., Hurst, I. R., and Holliday, L. S. (2004) J. Biol. Chem. 279, 79887998 89. Chen, D., Zhao, M., and Mundy, G. R. (2004) Growth Factors 22, 233241 90. Holliday, L. S., Lu, M., Lee, B. S., Nelson, R. D., Solivan, S., Zhang, L., and Gluck, S. L. (2000) J. Biol. Chem. 275, 32331 32337 91. Broehan, G., Zimoch, L., Wessels, A., Ertas, B., and Merzendorfer, H. (2007) J. Exp. Biol. 210, 36363643 92. Vitavska, O., Merzendorfer, H., and Wieczorek, H. (2005) J. Biol. Chem. 280, 10701076 93. Vitavska, O., Wieczorek, H., and Merzendorfer, H. (2003) J. Biol. Chem. 278, 1849918505 94. Zuo, J., Vergara, S., Kohno, S., and Holliday, L. S. (2008) J. Exp. Biol. 211, 11021108
108 95. Ma, B., Qian, D., Nan, Q., Tan, C., An, L., and Xiang, Y. (2012) J. Biol. Chem. 96. HurtadoLorenzo, A., Skinner, M., El, A. J., Futai, M., Sun Wada, G. H., Bourgoin, S., Casanova, J., Wildeman, A., Bechoua, S., Ausiello, D. A., Brown, D., and Marshansky, V. (2006) Nat. Cell Biol. 8, 124 136 97. Merkulova, M., HurtadoLorenzo, A., Hosokawa, H., Zhuang, Z., Brown, D., Ausiello, D. A., and Marshansky, V. (2011) Am. J. Physiol Cell Physiol 300, C1442C1455 98. Donaldson, J. G. (2003) J. Biol. Chem. 278, 4157341576 99. Sabe, H. (2003) J. Biochem. 134, 485 489 100. Schweitzer, J. K., Sedgw ick, A. E., and D'Souza Schorey, C. (2011) Semin. Cell Dev. Biol. 22, 3947 101. Lu, M., Holliday, L. S., Zhang, L., Dunn, W. A., Jr., and Gluck, S. L. (2001) J. Biol. Chem. 276, 3040730413 102. Lu, M., Sautin, Y. Y., Holliday, L. S., and Gluck, S. L. (2004) J. Biol. Chem. 279, 87328739 103. Lu, M., Ammar, D., Ives, H., Albrecht, F., and Gluck, S. L. (2007) J. Biol. Chem. 282, 2449524503 104. Su, Y., Zhou, A., Al Lamki, R. S., and Karet, F. E. (2003) J. Biol. Chem. 278, 2001320018 105. Su, Y., Blake Palmer, K. G., Sorrell, S., Javid, B., Bowers, K., Zhou, A., Chang, S. H., Qamar, S., and Karet, F. E. (2008) Am. J. Physiol Renal Physiol 295, F950 F958 106. Fulgenzi, G., Graciotti, L., Corsi, A., and Granata, A. L. (2001) J. Muscle Res. Cell Motil. 22, 391 397 107. Bronstein, W. W. and Knull, H. R. (1981) Can. J. Biochem. 59, 494499 108. Waingeh, V. F., Gustafson, C. D., Kozliak, E. I., Lowe, S. L., Knull, H. R., and Thomasson, K. A. (2006) Biophys. J. 90, 1371 1384 109. Carnell, M., Zech, T., Calaminus, S. D., Ura, S., Hagedorn, M., Johnston, S. A., May, R. C., Soldati, T., Machesky, L. M., and Insall, R. H. (2011) J. Cell Biol. 193, 831839 110. Brown, D., Breton, S., Ausiello, D. A., and Marshansky, V. (2009) Traffic. 10, 275 284 111. Schluter, K., Schleicher, M., and Jockusch, B. M. (1998) J. Cell Sci. 111 ( Pt 22), 32613273
109 112. Kane, P. M., Yamashiro, C. T., and Stevens, T. H. (1989) J. Biol. Chem. 264, 1923619244 113. Liu, J. and Kane, P. M. (1996) Biochemistry 35, 1093810948 114. Liu, M., Tarsio, M., Charsky, C. M., and Kane, P. M. (2005) J. Biol. Chem. 280, 3697836985 115. Liu, Q., Kane, P. M., Newman, P. R., and Forgac, M. (1996) J. Biol. Chem. 271, 20182022 116. Liu, Q., Leng, X. H., Newman, P. R., Vasilyeva, E., Kane, P. M., and Forgac, M. (1997) J. Biol. Chem. 272, 1175011756 117. Ostrov, D. A., Magis, A. T., Wronski, T. J., Chan, E. K., Toro, E. J., Donatelli, R. E., Sajek, K., Haroun, I. N., Nagib, M. I., Piedrahita, A., Harris, A., and Holliday, L. S. (2009) J. Med. Chem. 52, 51445151 118. Schutt, C. E., Myslik, J. C., Rozycki, M. D., Goonesekere, N. C., and Lindberg, U. (1993) Nature 365, 810 816 119. Dias, R. and de Azevedo, W. F. J. (2008) Curr. Dru g Targets. 9, 10401047 120. Ewing, T. J., Makino, S., Skillman, A. G., and Kuntz, I. D. (2001) J. Comput. Aided Mol. Des 15, 411 428 121. Rockey, W. M. and Elcock, A. H. (2006) Curr. Protein Pept. Sci. 7, 437 457 122. Henwood, J. M. and Monk, J. P. (1988) Drugs 36, 3266 123. Rafalsky, V., Andreeva, I., and Rjabkova, E. (2006) Cochrane. Database. Syst. Rev. CD003597 124. Carbon, C. (2001) Chemotherapy 47 Suppl 3, 9 14 125. Wiebe, S. H., Hafezi, M., Sandhu, H. S., Sims, S. M., and Dixon, S. J. (1996) Oral Dis. 2, 167 180 126. Andreasen, J. O. and Andreasen, F. M. (1992) Proc. Finn. Dent. Soc. 88 Suppl 1, 95114 127. Karras, J. C., Miller, J. R., Hodges, J. S., Beyer, J. P., and Larson, B. E. (2009) Am. J. Orthod. Dentofacial Orthop. 136, 843 847 128. Dunn, M. D., Park, C. H., Kostenuik, P. J., Kapila, S., and Giannobile, W. V. (2007) Bone 41, 446 455 129. Dolce, C., Vakani, A., Archer, L., Morris Wiman, J. A., and Holliday L. S. (2003) J. Dent. Res. 82, 682 686
110 130. Holliday, L. S., Vakani, A., Archer, L., and Dolce, C. (2003) J. Dent. Res. 82, 687691 131. Russell, R. G., Watts, N. B., Ebetino, F. H., and Rogers, M. J. (2008) Osteoporos. Int. 19, 733 759 132. Ghoneima, A. A., Allam, E. S., Zunt, S. L., and Windsor, L. J. (2010) Orthod. Craniofac. Res. 13, 1 10 133. Bowman, E. J. and Bowman, B. J. (2005) J. Bioenerg. Biomembr. 37, 431 435 134. Huss, M. and Wieczorek, H. (2009) J. Exp. Biol. 212, 341 346 135. Grzybowski, B. A., Ishchenko, A. V., Shimada, J., and Shakhnovich, E. I. (2002) Acc. Chem. Res. 35, 261 269 136. Kirchmair, J., Distinto, S., Schuster, D., Spitzer, G., Langer, T., and Wolber, G. (2008) Curr. Med. Chem. 15, 20402053 137. Foloppe, N. and Hubbard, R. (2006) Curr. Med. Chem. 13, 35833608 138. Forster, M. and Mulloy, B. (2006) Biochem. Soc. Trans. 34, 431 434 139. Fukunishi, Y. (2010) Expert. Opin. Drug Metab Toxicol. 6, 835 849 140. Kroemer, R. T. (2007) Curr. Protein Pept. Sci. 8, 312 328 141. Huentelman, M. J., Zubcevic, J., Hernandez Prada, J. A., Xiao, X., Dimitrov, D. S., Raizada, M. K., and Ostrov, D. A. (2004) Hypertension 44, 903 906 142. Knox, A. J., Meegan, M. J., and Lloyd, D. G. (2006) Curr. Top. Med. Chem. 6, 217243 143. Ostrov, D. A., Hernandez Prada, J. A., Corsino, P. E., Finton, K. A., Le, N., and Rowe, T. C. (2007) Antimicrob. Agents Chemother. 51, 36883698 144. Tomlinson, S. M., Malmstrom, R. D., Russo, A., Mueller, N., Pang, Y. P., and Watowich, S. J. (2009) Anti viral Res. 82, 110 114 145. Toro, E. J., Ostrov, D. A., Wronski, T. J., and Holliday, L. S. (2012) Curr. Protein Pept. Sci. 13, 180 191 146. Vaananen, H. K., Karhukorpi, E. K., Sundquist, K., Wallmark, B., Roininen, I., Hentunen, T., Tuukkanen, J., and Lakkakorpi, P. (1990) J. Cell Biol. 111, 13051311 147. Ma, B., Xiang, Y., and An, L. (2011) Cell Signal. 23, 12441256
111 148. Ochotny, N., Van, V. A., Chan, N., Yao, Y., Morel, M., Kartner, N., von Schroeder, H. P., Heersche, J. N., and Manolson, M. F. (2006) J. Biol. Chem. 281, 2610226111 149. Chin, N. X. and Neu, H. C. (1983) Antimicrob. Agents Chemother. 24, 754 763 150. Holliday, L. S., Dean, A. D., Greenwald, J. E., and Glucks, S. L. (1995) J. Biol. Chem. 270, 1898318989 151. Li, J. and Yuan, J. (2008) Oncogene 27, 61946206 152. Lee, S. H., Rho, J., Jeong, D., Sul, J. Y., Kim, T., Kim, N., Kang, J. S., Miyamoto, T., Suda, T., Lee, S. K., Pignolo, R. J., Koczon Jaremko, B., Lorenzo, J., and Choi, Y. (2006) Nat. Med. 12, 14031409 153. Nakamura, I., Sasaki, T., Tanaka, S., Takahashi, N., Jimi, E., Kurokawa, T., Kita, Y., Ihara, S., Suda, T., and Fukui, Y. (1997) J. Cel l Physiol 172, 230 239 154. Toro, E. J., Zuo, J., Ostrov, D. A., Catalfamo, D., Bradaschia Correa, V., AranaChavez, V., Caridad, A. R., Neubert, J. K., Wronski, T. J., Wallet, S. M., and Holliday, L. S. (2012) J. Biol. Chem. 287, 1789417904 155. Oddie, G. W., Schenk, G., Angel, N. Z., Walsh, N., Guddat, L. W., de, J. J., Cassady, A. I., Hamilton, S. E., and Hume, D. A. (2000) Bone 27, 575 584 156. Tehrani, S., Faccio, R., Chandrasekar, I., Ross, F. P., and Cooper, J. A. (2006) Mol. Biol. Cell 17, 2882 2895 157. Matsubara, T., Myoui, A., Ikeda, F., Hata, K., Yoshikawa, H., Nishimura, R., and Yoneda, T. (2006) J. Bone Miner. Metab 24, 368372 158. Ma, T., Sadashivaiah, K., Madayiputhiya, N., and Chellaiah, M. A. (2010) J. Biol. Chem. 285, 2991129924 159. Melo, S., Villanueva, A., Moutinho, C., Davalos, V., Spizzo, R., Ivan, C., Rossi, S., Setien, F., Casanovas, O., Simo Riudalbas, L., Carmona, J., Carrere, J., Vidal, A., Aytes, A., Puertas, S., Ropero, S., Kalluri, R., Croce, C. M., Calin, G. A., and Esteller, M. (2011) Proc. Natl. Acad. Sci. U. S. A 108, 43944399 160. Yagi, M., Miyamoto, T., Toyama, Y., and Suda, T. (2006) J. Bone Miner. Metab 24, 355358 161. Kukita, T., Wada, N., Kukita, A., Kakimoto, T., Sandra, F., Toh, K., Nagata, K., Iijima, T., Horiuchi, M., Matsusaki, H., Hieshima, K., Yoshie, O., and Nomiyama, H. (2004) J. Exp. Med. 200, 941 946 162. Mensah, K. A., Ritchlin, C. T., and Schwarz, E. M. (2010) J. Cell Physiol 223, 7683
112 163. Miyamoto, T. (2006) Mod. Rheumatol. 16, 341 342 1 64. Ljusberg, J., Ek Rylander, B., and Andersson, G. (1999) Biochem. J. 343 Pt 1, 63 69 165. Babb, S. G., Matsudaira, P., Sato, M., Correia, I., and Lim, S. S. (1997) Cell Motil. Cytoskeleton 37, 308325 166. Hanein, D., Volkmann, N., Goldsmith, S., Michon, A. M., Lehman, W., Craig, R., DeRosier, D., Almo, S., and Matsudaira, P. (1998) Nat. Struct. Biol. 5, 787 792 167. Pacaud, M. and Derancourt, J. (1993) Biochemistry 32, 34483455 168. Shinomiya, H., Hirata H., Saito, S., Yagisawa, H., and Nakano, M. (1994) Biochem. Biophys. Res. Commun. 202, 1631 1638 169. Kanzaki, H., Chiba, M., Arai, K., Takahashi, I., Haruyama, N., Nishimura, M., and Mitani, H. (2006) Gene Ther. 13, 678685 170. Kanzaki, H., Chiba, M. Takahashi, I., Haruyama, N., Nishimura, M., and Mitani, H. (2004) J. Dent. Res. 83, 920 925 171. Madan, M. S., Liu, Z. J., Gu, G. M., and King, G. J. (2007) Am. J. Orthod. Dentofacial Orthop. 131, 8 10 172. Hirabayashi, H., Takahashi, T., Fujisaki, J., Masunaga, T., Sato, S., Hiroi, J., Tokunaga, Y., Kimura, S., and Hata, T. (2001) J. Control Release 70, 183191 173. Hirabayashi, H., Sawamoto, T., Fujisaki, J., Tokunaga, Y., Kimura, S., and Hata, T. (2002) Biopharm. Drug Dispos. 23, 307315 174. Gil, L., Han, Y., Opas, E. E., Rodan, G. A., Ruel, R., Seedor, J. G., Tyler, P. C., and Young, R. N. (1999) Bioorg. Med. Chem. 7, 901 919 175. Page, P. C., McKenzie, M. J., and Gallagher, J. A. (2001) J. Org. Chem. 66, 37043708 176. Herczegh, P., Buxton, T. B., McPherson, J. C., III, Kovacs Kulyassa, A., Brewer, P. D., Sztaricskai, F., Stroebel, G. G., Plowman, K. M., Farcasiu, D., and Hartmann, J. F. (2002) J. Med. Chem. 45, 23382341 177. Houghton, T. J., Tanaka, K. S., Kang, T., Dietrich, E., Lafontaine, Y., Delorme, D., Ferreira, S. S., Viens, F., Arhin, F. F., Sarmiento, I., Lehoux, D., Fadhil, I., Laquerre, K., Liu, J., Ostiguy, V., Poirier, H., Moeck, G., Parr, T. R., Jr., and Far, A. R. (2008) J. Med. Chem. 51, 6955 6969
113 178. Ostrov, D. A., Magis, A. T., Wronski, T. J., Chan, E. K., Toro, E. J., Donatelli, R. E., Sajek, K., Haroun, I. N., Nagib, M. I., Piedrahita, A., Harris, A., and Holliday, L. S. (2009) J. Med. Chem. 52, 51445151 179. Holliday, L. S., Dean, A. D., Greenwald, J. E., and Glucks, S. L. (1995) J. Biol. Chem. 270, 1898318989 180. Reszka, A. A. and Rodan, G. A. (2004) Mini. Rev. Med. Chem. 4, 711 719 181. Baron, R., Ferrari, S., and Russell, R. G. (2011) Bone 48, 677 692 182. Azuma, Y., Sato, H., Oue, Y., Okabe, K., Ohta, T., Tsuchimoto, M., and Kiyoki, M. (1995) Bone 16, 235 245 183. Frith, J. C., Monkkonen, J., Auriola, S., Monkkonen, H., and Rogers, M. J. (2001) Arthritis Rheum. 44, 22012210 184. Rogers, M. J., Frith, J. C., Luckman, S. P., Coxon, F. P., Benford, H. L., Monkkonen, J., Auriola, S., Chilton, K. M., and Russell, R. G. (1999) Bone 24, 73S 79S 185. Wise, G. E. and King, G. J. (2008) J. Dent. Res. 87, 414 434 186. Reitan, K. (1967) Am. J. Orthod. 53 721745 187. Sirisoontorn, I., Hotokezaka, H., Hashimoto, M., Gonzales, C., Luppanapornlarp, S., Darendeliler, M. A., and Yoshida, N. (2012) Am. J. Orthod. Dentofacial Orthop. 141, 563573 188. Migliorati, C. A. (2003) J. Clin. Oncol. 21, 42534254 189. Migliorati, C. A., Schubert, M. M., Peterson, D. E., and Seneda, L. M. (2005) Cancer 104, 8393 190. Hoff, A. O., Toth, B. B., Altundag, K., Johnson, M. M., Warneke, C. L., Hu, M., Nooka, A., Sayegh, G., Guarneri, V., Desrouleaux, K., Cui, J., Adamus A., Gagel, R. F., and Hortobagyi, G. N. (2008) J. Bone Miner. Res. 23, 826836 191. King, G. J., Keeling, S. D., McCoy, E. A., and Ward, T. H. (1991) Am. J. Orthod. Dentofacial Orthop. 99, 456 465 192. Martino, S., Cauley, J. A., BarrettConnor, E., Po wles, T. J., Mershon, J., Disch, D., Secrest, R. J., and Cummings, S. R. (2004) J. Natl. Cancer Inst. 96, 17511761 193. Migliorati, C. A., Epstein, J. B., Abt, E., and Berenson, J. R. (2011) Nat. Rev. Endocrinol. 7, 3442 194. Chu, S., Ishikawa, H., Kim T., and Yoshida, S. (2000) Cleft Palate Craniofac. J. 37, 488496
1 14 195. Sims, N. A. and Gooi, J. H. (2008) Semin. Cell Dev. Biol. 19, 444 451 196. Lakkakorpi, P., Tuukkanen, J., Hentunen, T., Jarvelin, K., and Vaananen, K. (1989) J. Bone Miner. Res. 4, 817825 197. Vaananen, H. K. and Horton, M. (1995) J. Cell Sci. 108 ( Pt 8), 27292732 198. Hayman, A. R. (2008) Autoimmunity 41, 218223 199. Brown, D., Paunescu, T. G., Breton, S., and Marshansky, V. (2009) J. Exp. Biol. 212, 17621772 200. Shan, G., Li, Y., Zhang, J., Li, W., Szulwach, K. E., Duan, R., Faghihi, M. A., Khalil, A. M., Lu, L., Paroo, Z., Chan, A. W., Shi, Z., Liu, Q., Wahlestedt, C., He, C., and Jin, P. (2008) Nat. Biotechnol. 26, 933 940 201. Sennoune, S. R., Luo, D., and Mar tinez Zaguilan, R. (2004) Cell Biochem. Biophys. 40, 185 206 202. Linder, S. and Aepfelbacher, M. (2003) Trends Cell Biol. 13, 376 385 203. Aguirre, J. I., Akhter, M. P., Kimmel, D. B., Pingel, J. E., Williams, A., Jorgensen, M., Kesavalu, L., and Wronsk i, T. J. (2012) J. Bone Miner. Res. 204. Langer, R. and Folkman, J. (1976) Nature 263, 797 800
115 BIOGRAPHICAL SKETCH Edgardo J Toro Qui ones was born in 1973 in Bayam n Puerto Rico, oldest son of journalist and photographer Jos Toro Romanacce and adm inistrative assistant Melba Qui ones Maldonado. Edgardo attended the University of Missouri Kansas City and graduated with a BA in biology in 1998. Immediately after completing his undergraduate education, he worked at the National Institutes of Healt h (NIDCR, Bethesda, MD) under supervision of Dr. Abner Louis Notkins in diabetes research. He then graduated with honors from the University of Puerto Rico School of Dental Medicine, San Juan, Puerto Rico, in 2004 with a DMD degree. After graduating from dental school, Edgardo went on to complete a 2year residency in Pediatric Dentistry and subsequently a post doctoral program leading to an M.S. in clinical research. He studied the relationship between the enzyme urease produced by oral bacteria and the development of dental caries in children. Edgardo joined the University of Florida Inter Disciplinary Program in biomedical sciences in 2008 and pursued his Ph.D work in the laboratory of Dr. L. Shannon Holliday studying the effect of the small molecule e noxacin in osteoclasts. Since joining the Holliday Lab he has been involved in training new lab members and residents. He has presented his work at several national and international conferences and has published his research.