Essential Requirement of BMPs-2/4 for Both Osteoblastand Osteoclast Formation in Murine Bone Marrow Cultures

from Adult Mice: Antagonism by Noggin

ETSUKO ABE,1 MATSUO YAMAMOTO,1 YASUTO TAGUCHI,1 BEATA LECKA-CZERNIK,1,2

CHARLES A. O’BRIEN,1 ARIS N. ECONOMIDES,3 NEIL STAHL,3 ROBERT L. JILKA,1,2 andSTAVROS C. MANOLAGAS1,2

ABSTRACT

Bone morphogenetic proteins (BMPs) have been heretofore implicated in the induction of osteoblast differ-entiation from uncommitted progenitors during embryonic skeletogenesis and fracture healing. We havetested the hypothesis that BMPs are also involved in the osteoblastogenesis that takes place in the bone marrowin postnatal life. To do this, we took advantage of the properties of noggin, a recently discovered protein thatbinds BMP-2 and -4 and blocks their action. Addition of human recombinant noggin to bone marrow cellcultures from normal adult mice inhibited both osteoblast and osteoclast formation; these effects werereversed by exogenous BMP-2. Consistent with these findings, BMP-2 and -4 and BMP-2/4 receptor tran-scripts and proteins were detected in these primary cultures, in a bone marrow–derived stromal/osteoblasticcell line, as well as in murine adult whole bone; noggin expression was also documented in all thesepreparations. Moreover, addition of antinoggin antibody caused an increase in osteoblast progenitor forma-tion. These findings suggest that BMP-2 and -4 are expressed in the bone marrow in postnatal life and serveto maintain the continuous supply of osteoblasts and osteoclasts; and that, in fact, BMP-2/4-induced commit-ment to the osteoblastic lineage is a prerequisite for osteoclast development. Hence, BMPs, perhaps in balancewith noggin and possibly other antagonists, may provide the tonic baseline control of the rate of boneremodeling on which other inputs (e.g., hormonal, biomechanical, etc.) operate. (J Bone Miner Res 2000;15:663–673)

Key words: bone cell progenitors, BMP receptors, colony forming unit–fibroblast (CFU-F), CFU-osteoblast(CFU-OB), bone remodeling

INTRODUCTION

BONE REMODELING, a process responsible for the renewalof the adult human skeleton approximately every 10

years, is carried out by teams of juxtaposed osteoclasts andosteoblasts, two specialized cell types that originate, respec-

tively, from hematopoietic and mesenchymal progenitors ofthe bone marrow.(1,2) A continuous and orderly supply ofthese cells is essential for skeletal homeostasis as increasedor decreased production of osteoclasts or osteoblasts and/orchanges in the rate of their apoptosis are largely responsiblefor the imbalance between bone resorption and formation

1Division of Endocrinology and Metabolism and the UAMS Center for Osteoporosis and Metabolic Bone Diseases, Little Rock,Arkansas, U.S.A.

2The Central Arkansas Veterans Health Care System, University of Arkansas for Medical Sciences, Little Rock, Arkansas, U.S.A.3Regeneron Pharmaceuticals Inc., Tarrytown, New York, U.S.A.

JOURNAL OF BONE AND MINERAL RESEARCHVolume 15, Number 4, 2000© 2000 American Society for Bone and Mineral Research

663

that underlies several systemic or localized bone diseasessuch as osteoporosis, Paget’s, metastatic, and renal bonedisease.(3–9) Even now, however, little is known about thefactors responsible for sustaining the supply of osteoblastsin postnatal life and how osteoblastogenesis and osteoclas-togenesis are coordinated to ensure a balance between for-mation and resorption during normal bone remodeling.

Bone morphogenetic proteins (BMPs), members of theTGFb (transforming growth factorb) superfamily of pro-teins, are unique among growth factors that influence os-teoblast differentiation because they can initiate this processfrom uncommitted progenitors in vitro as well as invivo.(10–12) In particular, BMP-2 and -4 are expressed dur-ing murine embryonal skeletogenesis (days 10–12) and acton cells isolated from murine limb buds to promote theirdifferentiation into osteoblasts. In addition, BMP-2 and -4are involved in fracture healing, as evidenced by theirexpression in primitive mesenchymal cells and chondro-cytes at the site of callus formation, and the ability of BMPsto accelerate the fracture healing process when suppliedexogenously.(10,11) Besides BMP-2 and -4, BMP-5, -6, and-7 may also contribute to osteoblastic cell differentiationand bone formation.(10) BMP-2/4-induced osteoblast com-mitment is mediated by the type I BMP receptor and in-volves the phosphorylation of specific transactivators (smad1, 5, and 8), which then oligomerize with smad 4, andtranslocate into the nucleus.(13) These events induce anosteoblast-specific transcription factor (CBFA-1 [core bind-ing factor], also known as Osf-2 or PEBP2aA or AML3),which in turn activates osteoblast-specific genes.(14,15)

Several proteins able to antagonize BMP action haverecently been discovered. Noggin, chordin, and cerberuswere initially found in the Spemann organizer region of theXenopusembryo and were shown to be essential for neu-ronal or head development.(16–21) Noggin and chordin in-hibit the action of BMPs by binding to them with highaffinity and preventing their interaction with their receptors.Of those BMPs tested, noggin displays specificity, in thatbinding is very tight to BMP-2 and BMP-4 (Kd 5 2 3 10211

M), weak to BMP-7, and undetectable to TGFb or IGF(insulin-like growth factor)-I.(17)

Here, we have exploited the BMP-antagonist propertiesof noggin to test the hypothesis that BMPs are involved inthe osteoblastogenesis that takes place in the bone marrowin postnatal life. Osteoclastogenesis and osteoblastogenesisproceed simultaneously in most circumstances(4,8) and theformer may not occur without the latter,(15,22) because os-teoclast development requires support from stromal/osteoblastic cells.(23) The mechanistic basis of this depen-dency has been recently explained by the discovery of amembrane-bound cytokine-like molecule, receptor activatorof NF-kB ligand (RANKL), which is present in mesenchy-mal cells and binds to a specific receptor (RANK) onhematopoietic osteoclast progenitors.(24–26)Such binding isessential and, together with M-CSF, sufficient for osteoclas-togenesis. Accordingly, we postulated that blockade ofBMP action by noggin would interfere not only with osteo-blastogenesis, but with osteoclastogenesis as well. We showthat noggin does indeed inhibit both osteoblast and oste-oclast formation in bone marrow cell cultures from adult

mice, whereas inhibition of endogenous noggin productionby a neutralizing antibody increases osteoblastogenesis.Consistent with these observations, we also show that thegenes for BMP-2 and -4, the BMP-2/4 receptor, as well asnoggin are expressed in the adult murine bone marrow, amarrow-derived stromal/osteoblastic cell line, and in mu-rine adult whole bone.

MATERIALS AND METHODS

Materials

Human recombinant BMP-2 and -6, and an anti-humanBMP-2/4 antibody, which recognizes BMP-2 and -4, wereprovided by V. Rosen (Genetic Institute, Cambridge, MA,U.S.A.). 1,25(OH)2D3 and murine soluble RANKL wereprovided by Hoffmann-LaRoche (Nutley, NJ, U.S.A.) andDr. B. Boyle (Amgen Inc., Thousand Oaks, CA, U.S.A.),respectively. Human PTH (parathyroid hormone) (1–34)and human recombinant M-CSF (macrophage-colony stim-ulating factor) were purchased from Peninsula Laborato-ries (Belmont, CA, U.S.A.) andR & D Systems (Minne-apolis, MN, U.S.A.), respectively. The doses for BMPs,1,25(OH)2D3, PTH, soluble RANKL, or M-CSF used inthese experiments were determined in preliminary studies.cDNA probes for murine BMP-4 and BMP receptor type IA(BMPR-IA) were provided by Dr. K. Miyazono (CancerInstitute, Tokyo, Japan). A murine RANKL cDNA probewas provided by Immunex Corp. (Seattle, WA, U.S.A.).

Osteoblast progenitor assays

C2C12 cells (23 104/cm2) were cultured with 100 ng/mlBMP-2, BMP-6, and/or 10–600 ng/ml noggin for 3 daysand alkaline phosphatase (AP) activity was measured bySigma kit #104 (Fig. 1).(27) For the experiment depicted(later) in Fig. 7, C2C12 cells (23 104/cm2) were precul-tured with 250 ng/ml noggin and 2 or 10mg/ml of antinog-gin antibodies (#06, #16, #18, or #21) overnight. Subse-quently, 50 ng/ml BMP-2 was added and the cultures weremaintained for 3 days. Bone marrow cells were obtainedfrom femurs of 3-month-old male Swiss Webster mice andcultured at 13 106 cells (for CFU-F determination) or 23106 cells (for CFU-OB) per 10 cm2 well in aMEM (a-minimal essential medium; Gibco-BRL, Gaithersburg, MD,U.S.A.) supplemented with 15% preselected FCS (Hyclone,Logan, UT, U.S.A.), 200mM ascorbic acid, and 10 mMb-glycerophosphate. Cultures were maintained in the ab-sence or presence of different concentrations of recombi-nant human noggin, without or with 100 ng/ml humanrecombinant BMP-2. The total number of CFU-F coloniesand the number of AP-positive CFU-F colonies were deter-mined after 10 days of culture by staining for AP, and thenumber of CFU-OB colonies was determined after 28 daysof culture by Von Kossa staining, as previously described.(8)

UAMS-33, a cell line with preosteoblastic properties, wasobtained by limiting dilution subcloning from foci of trans-formed cells that developed during long-term culture ofmurine bone marrow cells.(28) For the osteoblast differenti-ation experiments, UAMS-33 cells were plated at 23

664 ABE ET AL.

104/cm2 well and maintained for up to 8 days inaMEMcontaining 10% FCS in the presence of the indicated con-centrations of recombinant human noggin or a pegylatednoggin (PEG-noggin). PEG-noggin was prepared by cova-lent attachment of a 20-kDa polyethyleneglycol to availablelysines of human noggin to form a secondary amine link-age.(29) PEG-noggin was purified by ion-exchange chroma-tography to isolate noggin cross-linked with a single PEGmolecule (unpublished data by K. M. Bailey et al. in Re-generon Pharmaceuticals Inc.). As shown in the case ofcalcitonin, pegylation of noggin should provide a moleculethat is more stable in solution than the unpegylated nog-gin.(29) Osteoblast differentiation was assessed by measur-ing AP activity.

Osteoclast development assays

Bone marrow cells were obtained from the femurs of 3- to4-month-old mice and cultured for 8 days for the determi-nation of osteoclast formation as previously described.(3)

Cultures were maintained in the absence or the presence of1028 M 1,25(OH)2D3 or 1028 M human PTH (1–34) with-out or with 200 ng/ml PEG-noggin, 300 ng/ml BMP-2, or10 mg/ml antinoggin antibodies #06 or #21. In the experi-ments shown in Figs. 2C and 2D, osteoclast formation wasassessed in 8-day cocultures of nonadherent bone marrowcells (0.53 106 cells/cm2) and either UAMS-33 cells (13104/cm2) or murine calvaria cells (13 104/cm2). Directeffects of noggin on hematopoietic cell differentiation to-ward the osteoclastic phenotype were examined using non-adherent murine bone marrow cells (105/well). Nonadherentbone marrow cells obtained by preculturing bone marrowcells for 2 days (to remove stromal/osteoblasts) were cul-tured for 6 days with 10 ng/ml human M-CSF, 100 ng/ml

soluble RANKL, and 10 or 100 ng/ml PEG-noggin. In allthese experiments, osteoclastic cells were visualized bystaining for tartrate-resistant acid phosphatase (TRAP) us-ing Sigma kit #180.

RT-PCR analysis

RNA was prepared from freshly isolated bone marrowcells, or from 2-, 4-, 7-, 14-, and 21-day cultures of thesecells as previously described,(30) and analyzed for the ex-pression of BMP-2, BMP-4, osteocalcin, BMPR (BMPreceptor)-IA, and BMPR-IB transcripts by RT-PCR. RT-PCR was performed using primers, as detailed previ-ously for GAPDH (glyceraldehyde-3-phosphate dehydroge-nase).(30) For murine BMP-2, the primer set was: 59-CTAGTGTTGCTGCTTCCCCA (forward), 59-GAGTTC-AGGTGGTCAGCAAG (reverse); for murine BMP-4,the primer set was: 59-GCGCCGTCATTCCGGATTAC(forward), 59-CATTGTGATGGACTAGTCTG (reverse);for murine BMPR-IA, the primer set was: 59-GGCAGAA-TCTAGATAGTATGCTCC (forward), 59-GAAGTTAA-CGTGGTTTCTCCCTG (reverse); for murine BMPR-IB,the primer set was: 59-CACCAAGAAGGAGGATGGAG-AGA (forward), 59-CTACAGACAGTCACAGATAAGC(reverse); for murine osteocalcin, the primer set was: 59-TCTGACAAAGCCTTCATGTCC (forward), 59-AAA-TAGTGATACCGTAGATGCG (reverse); and for murinenoggin, the primer set was: 59-TGGACCTCATCGAA-CATCCAGAC (forward), 59-ACTTGGATGGCTTAC-ACACCATGC (reverse). Using these primers, the expectedsizes of the PCR products are exactly as depicted later inFig. 6A.

In situ RT-PCR

Bone marrow cultures were established and maintainedfor 2 weeks. After fixation with 4% paraformaldehyde,BMP-4 and osteocalcin transcripts were detected using theBMP-4 and osteocalcin primers described earlier and amethod previously detailed elsewhere.(30) As a negativecontrol, parallel cultures were processed for BMP-4 detec-tion without reverse-transcriptase.

RNase protection assay

RNA was prepared from 2-week cultures of murine bonemarrow cells and from 3-day cultures of UAMS-33 cells,and assayed by RNase protection for BMP-4 and BMPR-IAtranscripts. For the preparation of the cDNA probes, plas-mids containing the coding regions of murine BMP-4 andBMPR-IA were subcloned in Bluescript KS(1) plasmidsand were linearized. The respective riboprobes were syn-thesized in the presence of 50–100mCi of [32P]UTP (3000Ci/mmol, Amersham Corp., Arlington Heights, IL, U.S.A.),and T7 or SP6 RNA polymerase, as appropriate (Promega,Madison, WI, U.S.A.). Total RNA (30mg) was extractedfrom 3-day cultures of UAMS-33 cells and 2-week cul-tures of murine bone marrow cells. RNA and32P-labeledriboprobes in hybridization buffer (80% formaldehyde; 40mM PIPES, pH 6.4; 400 mM NaCl; 1 mM EDTA) were

FIG. 1. Effect of noggin on BMP-2- or BMP-6-inducedalkaline phosphatase in C2C12 cells. Cells (23 104/cm2)were cultured for 3 days in the presence of the indicatedconcentrations of BMP-2, -6, or noggin, and alkaline phos-phatase (AP) activity was measured. Each bar represents themeans (6 SEM) of quadruplicate determinations. Data wereanalyzed by ANOVA after establishing homogeneity ofvariances. *p , 0.05 versus cells cultured in 100 ng/mlBMP-2 alone.

665BMPs, NOGGIN, AND BONE CELL FORMATION

annealed at 45°C overnight after heating at 85°C for 5minutes. Subsequently, annealed RNAs were treated withRNase A (40mg/ml) at 30°C for 60 minutes, and theenzyme was inactivated by proteinase K (100mg) and 10%SDS. Finally, the samples were loaded onto 4.5% polyacryl-amide gels with 7 M urea after extraction by phenol andethanol.

Immunostaining

Two-week cultures of murine bone marrow cells and3-day cultures of UAMS-33 cells were immunohistostainedwith an antibody against BMP-2 and -4. Cells were fixedwith 10% formalin for 10 minutes, treated with 0.1% H2O2

for 30 minutes to remove endogenous peroxidase activity,

and blocked with 5% normal goat serum for 1 h. They werethen incubated with mouse anti-human BMP-2/4 antibodyfor 1 h and subsequently for 30 minutes with biotinylatedsecond antibody (Vector, Burlingame, CA, U.S.A.), fol-lowed by incubation with peroxidase-conjugated streptavi-din, and 3,39-diaminobenzidine tetrachloride (Santa CruzBiotechnology Inc., Santa Cruz, CA, U.S.A.). As a negativecontrol, cells were processed without primary antibody.

Northern blot analysis

Total RNA (30 mg) or polyA1 RNA (8 mg) were pre-pared from 3- or 6-day cultures of UAMS-33 cells, and wereelectrophoresed on 1% agarose gels. Northern blotting for

FIG. 2. Effect of noggin on osteoblastogenesis. Bone marrow cells (A–C) were maintained in the absence or presenceof the indicated concentrations of recombinant human noggin, without or with 100 ng/ml human recombinant BMP-2. Thetotal number of CFU-F colonies (A) or the number of AP-positive CFU-F colonies (B) were determined after 10 days ofculture by staining for AP, whereas the number of CFU-OB colonies (C) was determined after 28 days of culture by VonKossa staining. Data shown are the mean number of colonies (6 SEM) per 10 cm2 well (n 5 4 wells/group). In the presenceof 100 ng/ml noggin, AP activity (Sigma kit #104, St. Louis, MO, U.S.A.) decreased from 15.06 4.0 to 0.56 0.1 nmolp-nitrophenol hydrolyzed/minute/mg protein (p , 0.01 by Student’st-test) in CFU-F colonies formed in 10-day culturesof murine bone marrow cells. (D) UAMS-33 cell differentiation. Data shown are the means (6 SEM) of AP activity aftercorrection for cellular protein. Inset: 1 day after initiation of culture (0), cells were maintained in the absence (open circles)or presence (closed circles) of 100 ng/ml PEG-noggin for 4 days. In a subset of cultures, medium was replaced at 4 daysand cultures continued for another 4 days without or with PEG-noggin; in addition, PEG-noggin was removed frommedium of some cultures by replacing medium without PEG-noggin (closed triangle). AP activity was determined at 0, 4,and 8 days in parallel cultures. Data were analyzed by ANOVA. *p , 0.05 versus cells cultured without noggin orPEG-noggin;†p , 0.05 versus cells cultured without noggin, or cells cultured with noggin and BMP-2;‡p , 0.05 versuscells maintained for 8 days with PEG-noggin.

666 ABE ET AL.

RANKL, M-CSF, and noggin expression was performed aspreviously described.(30)

Western blot analysis

Cells were lysed in 10 mM phosphate buffer (pH 7.4),10% glycerol, 1% NP-40, 0.1% SDS, 4 mM EDTA, 0.15 MNaCl, 0.01 M NaF, 0.1% sodium orthovanadate, 1 mMPMSF, 5 mg/ml trypsin inhibitor, and 5mg/ml proteaseinhibitors, and centrifuged at 14,000g for 10 minutes. Forpreparation of bone homogenates, femurs were frozen inliquid N2 and homogenized in the above-mentioned bufferusing a Polytron homogenizer, and then centrifuged. Super-natant proteins (50–200mg) from the cell lysates or thebone homogenates were subjected to 12% SDS-PAGE(polyacrylamide gel electrophoresis) and electroblotted ontoa PVDF membrane (Millipore, Bedford, MA, U.S.A.).Membranes were incubated for 2 h atroom temperature in5% dry milk in 20 mM Tris, (pH 7.2), 0.15 M NaClcontaining 0.05% Tween 20, and subsequently with anti-BMP-2/4 antibody, or the antinoggin antibody, and thenappropriate second antibody (HRPO [horseradish per-oxidase]-goat anti-mouse IgG antibody for BMP-2/4 orHRPO-goat anti-rat IgG antibody for noggin). The boundantibody on the membrane was detected by enzyme reactionusing an enhanced chemiluminescence kit (Dupont NEN,Boston, MA, U.S.A.).(30)

RESULTS

Inhibition of osteoblastogenesis by noggin

The ability of human recombinant noggin to antagonizeBMP-mediated commitment to the osteoblast lineage, aswell as the specificity of this effect, was established usingC2C12 cells. Human recombinant noggin (100–600 ng/ml)inhibited BMP-2-induced alkaline phosphatase (AP) activ-ity, a phenotypic marker of osteoblastic cells, in a dose-dependent manner (Fig. 1). At the highest concentration ofnoggin, the effect of BMP was completely blocked. LikeBMP-2, BMP-6 stimulated AP activity in C2C12 cells;however, the effect of BMP-6 was not affected by noggin,indicating that noggin is not an antagonist of BMP-6. Asshown previously,(31) BMP-12 or -13 added alone did notinduce AP activity in C2C12 cells, nor did they interferewith the stimulatory effect of BMP-2 or -6 (data not shown).

Noggin had no effect on the number of colony formingunits–fibroblast (CFU-F) that were formed during a 10-dayprimary culture of murine bone marrow cells obtained from3-month-old mice (Fig. 2A). Noggin, however, led to adose-dependent decrease in the number of AP-positiveCFU-F colonies (Fig. 2B). Noggin also inhibited bone nod-ule formation by subsets of CFU-F colonies, designatedCFU-osteoblast (CFU-OB), in 28-day primary cultures ofbone marrow cells (Fig. 2C). Both effects were detectablewith as little as 6–10 ng/ml of noggin. Practically completesuppression of AP expression in CFU-F or bone noduleformation by noggin could be observed with 150–600 ng/ml. The inhibitory effect of noggin on AP expression inCFU-F colonies could be reversed by addition of 100 ng/ml

BMP-2, except at the highest concentration of noggin (600ng/ml), confirming competitive antagonism between nogginand BMPs (Fig. 2B). Similarly, noggin and PEG-noggindose-dependently reduced AP activity in UAMS-33 culturesmaintained under basal conditions (Fig. 2D), or in thepresence of exogenous BMP-2 (data not shown). Nogginalso prevented calcium deposition by UAMS-33 cellscultured in the presence of ascorbic acid andb-glycerophosphate for 5 days (data not shown). In linewith the expectation that pegylation would increase thestability of noggin, we found in all these experiments thatPEG-noggin was 10-fold more potent than unpegylatednoggin. Withdrawal of PEG-noggin after 4 days restored APlevels in UAMS-33 cells, suggesting that the effects of theprotein observed herein were not the result of cytotoxicactions (Fig. 2D). Consistent with this, noggin had no effecton cell viability measured by trypan blue exclusion.

Inhibition of osteoclastogenesis by noggin

Osteoclast formation induced by either 1,25(OH)2D3 orPTH was also inhibited by the addition of PEG-noggin inbone marrow cell cultures (Figs. 3A and 3B), or coculturesof nonadherent bone marrow cells, a source of osteoclastprogenitors, and either UAMS-33 cells (Fig. 3C) or neonatalmurine calvaria osteoblastic cells (Fig. 3D). UAMS-33cells can support osteoclast differentiation induced by1,25(OH)2D3, but unlike calvaria cells, they do not supportPTH-induced osteoclast development. The effect of nogginwas dose dependent and identical in experiments using cellsfrom 3- or 6-month-old mice. In contrast, exogenous humanrecombinant BMP-2 increased osteoclast formation as muchas 4- to 6-fold over baseline in PTH-stimulated primarybone marrow cell cultures (Fig. 3B), and in cocultures ofhematopoietic progenitors and calvaria cells (Fig. 3D). Anincrease, albeit smaller, was also observed in 1,25(OH)2D3-stimulated cultures (Figs. 3A and 3C). However, in differentexperiments, in which cocultures of UAMS-33 cells andnonadherent hematopoietic precursors were pretreated withBMP-2 before exposure to 1,25(OH)2D3, BMP-2 stimulatedosteoclast formation by 3- to 4-fold (data not shown).

To investigate the possibility that noggin could haveinhibited osteoclast formation in part as a result of directactions on hematopoietic cells, we employed preparationsof nonadherent murine bone marrow cells (devoid ofstromal/osteoblastic cells) and cultured them for 6 days inthe presence of human recombinant M-CSF and solublemurine RANK ligand (sRANKL)(Fig. 4). As shown previ-ously by Lacey et al.,(24) sRANKL along with M-CSF wassufficient for the induction of osteoclast formation in theabsence of stromal/osteoblastic support cells or stimuli like1,25(OH)2D3 or PTH. In this system of stromal/osteoblasticcell–independent osteoclastogenesis, noggin had no effecton osteoclast development. Consistent with the conclusionthat noggin does not exert its anti-osteoclastogenic effectsthrough direct actions on hematopoietic osteoclast precur-sors, PEG-noggin had no effect on M-CSF-stimulated pro-liferation of nonadherent hematopoietic progenitors isolatedfrom murine bone marrow (data not shown).

667BMPs, NOGGIN, AND BONE CELL FORMATION

In support of the notion that the effect of noggin onosteoclastogenesis was the result of attenuation of stromal/osteoblastic cell differentiation toward a state capable ofsupporting osteoclast development, PEG-noggin at 200ng/ml decreased the expression of the mRNA for RANKligand by 40% in 1,25(OH)2D3-treated UAMS-33 cells (Fig.5); however, at the same concentration, PEG-noggin had noeffect on M-CSF mRNA. Besides affecting the differenti-ated phenotype, the inhibitory effects of PEG-noggin onosteoclast development could be caused by a reduction inthe total number of stromal/osteoblastic cells. This possibil-ity was excluded by showing that PEG-noggin did not affectUAMS-33 or calvaria cell proliferation under the conditionsused in the cocultures of Fig. 3 (data not shown).

Expression of BMP-2/4, their receptors, and noggin inbone marrow and bone

The results presented earlier strongly suggested that thegenes encoding BMPs and their receptors are expressed inbone marrow cultures from normal adult mice and that theirproducts are required for osteoblastogenesis as well as os-

FIG. 3. Effect of PEG-noggin and BMP-2 on osteoclast formation. (A and B) Bone marrow cells from the femurs of 3-to 4-month-old mice were maintained in the absence or the presence of (A) 1028 M 1,25(OH)2D3 or (B) 1028 M humanPTH (1–34) without or with 200 ng/ml PEG-noggin or 300 ng/ml BMP-2. (C and D) Cocultures of nonadherent bonemarrow cells (0.53 106 cells/cm2) and either UAMS-33 cells (23 104/well) (C) or murine calvaria cells (13 104/cm2)(D) were stimulated with1028 M 1,25(OH)2D3 or 1028 M PTH, respectively. Osteoclastic cells were visualized by TRAPstaining (n 5 4/group). Data shown are the mean number (6 SEM) of multinucleated TRAP-positive cells/well. Data wereanalyzed by ANOVA. *p , 0.05 versus cultures stimulated with 1,25(OH)2D3 or PTH alone.

FIG. 4. Noggin does not affect hematopoietic cell differ-entiation toward the osteoclast phenotype. Nonadherent mu-rine bone marrow cells were cultured for 6 days in thepresence of the indicated reagents and osteoclast develop-ment was determined by TRAP staining. Data shown are themean number (6 SEM) of multinucleated TRAP-positivecells/well.

668 ABE ET AL.

teoclastogenesis. Therefore, we searched for BMP-2, -4,and -7 transcripts and proteins in bone marrow cells andhomogenates of bone from femurs of 3-month-old mice(Fig. 6). BMP-4 transcripts could be detected by RT-PCR inadherent bone marrow cells maintained in culture as early as4 days and BMP-2 transcripts in cells maintained for 2weeks, but not in freshly isolated bone marrow cells (Fig.6A). In contrast to BMP-2 and -4, BMP-7 transcripts werenot detected in either freshly isolated or cultured bonemarrow cells (data not shown). BMP receptor type IA wasdetected by RT-PCR in freshly isolated cells and throughoutthe culture; whereas the type IB receptor was first visualizedat 4 days. Osteocalcin mRNAs was first detected at 1 weekof culture. We previously showed that IL-6, gp130, andosteopontin transcripts are absent from freshly isolated bonemarrow cells but appear during culture.(30,32) Hence, theexpression of BMPs and BMP receptor genes on culture ofbone marrow cells is most likely the result of an expansionof the osteogenic cell population. Using in situ RT-PCRanalysis, the BMP-4 and osteocalcin transcripts were local-ized in a subset of cells present within the CFU-F colonies(Fig. 6B). Independent confirmation of the RT-PCR resultsfor BMP-4 and BMPR-IA expression in bone marrow cul-tures was obtained using RNase protection assay of RNAisolated from 2-week cultures; identical transcripts were

also observed in the UAMS-33 cells (Fig. 6C). In agreementwith the mRNA studies, the BMP-2/4 proteins were de-tected in bone marrow cells cultured for 2 weeks, as well asin UAMS-33 cells, by immunostaining with an antibodythat recognizes both BMP-2 and -4 (Fig. 6D). Moreover,BMP-4 transcripts and BMP-2/4 proteins as well as noggintranscripts and proteins could be shown by RT-PCR orNorthern blot analysis (Fig. 6E) and Western blot analysis(Fig. 6F) in bone marrow cultures, homogenates of wholefemurs from 3-month-old mice, and also in UAMS-33 cells.In UAMS-33 cells, noggin expression was highest in post-confluent cultures (6 days), at which time the activity of theosteoblast phenotypic marker alkaline phosphatase wasmaximal.

Increased osteoblastogenesis by a noggin-neutralizingantibody

Finally, based on the observations that noggin was ex-pressed in the primary murine bone marrow cell cultures,we attempted to establish the biological significance of thisphenomena using a noggin-neutralizing antibody. For thispurpose, we employed a rat-derived monoclonal antinogginantibody (#06), prepared by standard hybridoma technol-ogy, which recognizes both human and murine noggin. Theability of this antibody to neutralize the activity of nogginwas shown in the results shown in Fig. 7. Indeed, theinhibitory effect of noggin on BMP-2-induced differentia-tion of C2C12 cells toward osteoblasts could be reversedcompletely by 2–10mg/ml of the #06 antibody. In contrastto antibody #06, a different antinoggin antibody (#21) didnot exhibit neutralizing activity at the same doses as #06.Therefore, the latter was used in these studies as a negativecontrol. As shown in Fig. 8, addition of 10mg/ml of thenoggin-neutralizing antibody #06, but not antibody #21, tomurine bone marrow cell cultures caused an increase in thenumber of both AP-positive CFU-F and CFU-OB colonies(Fig. 8). However, the noggin-neutralizing antibody did notaffect osteoclast formation in bone marrow cell cultures(osteoclast number 51.56 12.1 in group with #21 vs.63.56 13.1 in group with #06).

DISCUSSION

Nature recapitulates evolutionarily successful mecha-nisms to accomplish similar tasks at different stages of life.The results of this study reinforce this truism by showingthat BMP-2 and -4, the same proteins that have been impli-cated in the induction of osteoblast formation in the embryoand during fracture repair, may be required for osteoblas-togenesis in the murine bone marrow in postnatal life. Inaddition, our data reveal that induction of mesenchymal celldifferentiation toward the osteoblast phenotype by BMP-2/4is a prerequisite for osteoclastogenesis. Thus, the require-ment of BMPs for osteoclastogenesis may offer an entirelynew perspective of how bone homeostasis is maintainedduring physiologic remodeling and may explain the in vivoobservations that osteoclastogenesis(4,8,15) and loss ofbone(22) cannot occur without osteoblastogenesis.

FIG. 5. Decrease of RANKL mRNA abundance by PEG-noggin. UAMS-33 cells were cultured for 3 days in theabsence (Cont) or presence of 1028 M 1,25(OH)2D3 or 200ng/ml PEG-noggin alone or in combination, and Northernanalysis was performed using total RNA (30mg). GAPDH,a housekeeping enzyme, was used as a control for loadingand the ratio of the mRNA of interest:GAPDH is shown ontop of the corresponding lanes.

669BMPs, NOGGIN, AND BONE CELL FORMATION

FIG. 6. Expression of BMPs, noggin, and BMP receptors in a murine preosteoblast cell line, cultured bone marrow cells,and bone homogenates. (A) RNA was prepared from freshly isolated bone marrow cells (BMC), or from 2-, 4-, 7-, 14-, and21-day cultures of these cells. (B) Two-week cultures of murine bone marrow cells were established and, after fixation,osteocalcin transcripts were detected by in situ RT-PCR. As a negative control, parallel cultures were processed for BMP-4detection without reverse-transcriptase. Original magnification5 1003. (C) RNA was prepared from 2-week cultures ofmurine bone marrow cells, and from 3-day cultures of UAMS-33 cells and assayed by RNase protection for BMP-4 andBMPR-IA transcripts. The expected bands for BMP-4, BMPR-IA, andb-actin, 290, 216, and 174 bp, respectively, areshown by arrows. Intact RNA probes used in this assay exhibit a higher molecular weight than the protected bands, andare absent after RNase digestion. (D) Two-week cultures of murine bone marrow cells, and 3-day cultures of UAMS-33cells, were immunohistostained with an antibody against BMP-2 and -4 (1Ab). As a negative control, cells were processedwithout primary antibody (2Ab). Original magnification5 1003. (E) Left and middle panel: RNA was prepared fromfemurs of neonatal, or brain and femurs of 3-month-old Swiss Webster mice and analyzed for BMP-4, noggin, and GAPDHtranscripts by RT-PCR. Right panel: Northern analysis of polyA1 RNA from 6-day cultures of UAMS-33 cells. (F) Leftpanel: Western blot analysis of noggin in lysates from UAMS-33 cells cultured for 0, 3, or 6 days, from murine bonemarrow cells cultured for 2 weeks, and from intact femurs of 3-month-old Swiss Webster mice. As a positive control, 10ng of recombinant human noggin (rNoggin) was used. Right panel: Western blot analysis of BMP-2/4 in homogenates ofadult murine femoral bone. As a positive control, 50 ng of recombinant human BMP-2 (rBMP-2) was used.

670 ABE ET AL.

In our studies, BMP-2 alone did not induce osteoclastformation and, consistent with this finding, did not inducethe expression of RANKL in stromal/osteoblastic cells.However, it did stimulate osteoclast formation (and proba-bly RANKL expression, as evidenced by the decrease ofRANKL levels by noggin) in the presence of 1,25(OH)2D3.This finding is somewhat different from the results of Ka-nataki et al.(33) and Hentunen et al.,(34) who have reportedthat BMP-2 or -7 alone has a stimulatory effect on oste-oclast formation in vitro, although in those other studies,1,25(OH)2D3 dramatically enhanced the level of osteoclas-togenesis produced by BMPs alone. This apparent discrep-ancy may result from the difference in the culture systemsused in those studies versus the present one. Even so,although both BMPs and 1,25(OH)2D3 seem to influencethe process of osteoclastogenesis through effects onstromal/osteoblastic cells, the mechanism(s) by which theyinfluence the process and the stage of cellular differentiation

at which they seem to work are quite distinct. Indeed,whereas BMP-2/4 induce uncommitted mesenchymal pro-genitors to differentiate toward the osteoblastic lineage (per-haps by inducing CBFA-1 expression), in studies not shownhere, we have found that 1,25(OH)2D3 inhibits BMP-4expression by 80% and CBFA-1 expression to undetectablelevels in bone marrow–derived UAMS-33 stromal/osteoblastic cells. Therefore, unlike BMPs, which act toinduce the differentiation of uncommitted mesenchy-mal progenitors toward the stromal/osteoblastic lineage,1,25(OH)2D3, at least in vitro, appears to act at a later stageto retain committed cells to a state of preterminal differen-tiation under which they are capable of supporting oste-oclast development. This notion is consistent with our otherstudies, indicating that the osteoclast support cells in thebone marrow are not fully differentiated osteoblasts, butrather cells expressing a preterminally differentiated pheno-type that exhibits mixed osteoblastic/adipocytic character-istics.(35)

The evidence that BMPs are required for osteoblast andosteoclast development implies that in addition to theirpreviously known roles in skeletal development and repair,these proteins may also be involved in bone remodeling. Allthese processes require osteoblasts and osteoclasts, but itappears that the production of these cells is governed by thesame proteins throughout life; consequently, biomechanicaland local signals must determine how they are deployed fordifferent purposes. The evidence that the genes for BMP-2and -4 as well as noggin are expressed in the adult murinebone marrow raises the possibility that the balance betweenBMPs and noggin provides a tonic baseline control ofosteoblastogenesis and osteoclastogenesis on which otherinputs (e.g., biomechanical, hormonal, etc.) operate, eitherby influencing the balance between BMPs and noggin orby providing independent signals that alter the pro-differentiating effects of BMPs. We are quick to point out,however, that noggin may be just one of several BMPantagonists with a role in the regulation of osteoblastogen-esis and osteoclastogenesis, because other proteins such aschordin have similar BMP antagonist properties.(16–21)Be-cause BMP-6 has been implicated in osteoblast differentia-tion,(10) in studies as shown in Fig. 1, we searched for andfound that noggin did not antagonize BMP-6 activity.Therefore, it is unlikely that the effects of noggin describedherein are the result of interference with BMP-6.

We have not definitively identified the cells expressingBMP-2 and -4 and noggin in bone marrow and intact bone.Nonetheless, the in situ RT-PCR analysis of the bone mar-row cultures suggests that they are of the mesenchymal/osteoblastic lineage. Consistent with this, we detectedBMP-4, BMPR-IA, as well as noggin transcripts and pro-teins in UAMS-33 cells, a bone marrow–derived preosteo-blastic line.(28) The level of expression of these genes inUAMS-33 cultures, a homogeneous preparation, was con-siderably higher when compared with the heterogeneousprimary bone marrow cell cultures. This observation addssupport to the contention that most of the cells expressingthese genes in the bone marrow are of the mesenchymal/osteoblastic lineage. The observation that expression ofnoggin increased in the UAMS-33 cells during culture sug-

FIG. 7. Effect of antinoggin antibody on AP activity inC2C12 cells. C2C12 (23 104/cm2) were cultured with 250ng/ml noggin and 2 or 10mg/ml of antinoggin antibodies#06, #16, #18, and #21 overnight. A 50 ng/ml aliquot ofBMP-2 was added to the indicated cultures. Three days afterBMP-2 addition, AP activity was measured and expressedas the means (6 SEM). *p , 0.05.

FIG. 8. Effect of antinoggin antibody on CFU-F andCFU-OB colony number in murine bone marrow cell cul-tures. Bone marrow cells were cultured in the absence(Cont) or presence of 10mg/ml of antinoggin antibodies #21or #06. The number of AP-positive CFU-F and CFU-OBcolonies were determined at 10 and 28 days of culture,respectively. Each bar represents the means (6 SEM) of thenumber of colonies present in four 10-cm2 well plates. *p ,0.05.

671BMPs, NOGGIN, AND BONE CELL FORMATION

gests that noggin expression by osteoblastic cells may in-crease as they advance to more differentiated stages (Fig.6F). More important, the evidence that neutralization ofendogenous noggin increased the number of CFU-F andCFU-OB colonies in our murine bone marrow cell cultures(Fig. 8) strongly suggests that the biological role of BMPsis counterbalanced by noggin. In fact, if noggin is producedby terminally differentiated osteoblasts, lining cells, or os-teocytes, it could serve to restrict inappropriate osteoblas-togenesis, and thereby target remodeling to appropriatesites. In support of the contention that the balance betweenBMPs and noggin has important biologic implications, itwas recently shown that absence of regulated BMP activityin mice lacking noggin leads to failure of joint development;(36) and that BMPs as well as TGFb induce the expressionof noggin in osteoblastic cells.(37) Moreover, in line with theview that noggin may serve to counteract the effects ofBMPs in vivo, PEG-noggin administration to mice inhibitsbone formation in subcutaneous implants of BMP-2-impregnated matrigel, a solubilized basement-membranepreparation.(38) Definitive evidence, however, for or againstthe importance of these proteins in adult bone remodelingmust, of course, await the results of future studies thatexamine the effects of administration of BMP-2 or -4 ornoggin to adult animals.

In agreement with the evidence that osteoblastogenesis isa prerequisite for osteoclastogenesis—highlighted herein bythe evidence that the effects of noggin on osteoclastogenesisare the result of antagonism of BMP actions on mesenchy-mal cells, as opposed to direct actions on hematopoieticprogenitors—we have found and reported elsewhere that thepromoters of both the murine and human RANK ligandgene contain two functional CBFA-1 binding sites; and thatmutation of these sites abrogates the transcriptional activityof the RANKL promoter.(39) We, therefore, propose that aBMP3 CBFA-13 RANK ligand gene expression cascadein cells of the stromal/osteoblastic lineage constitutes themolecular basis of the linkage between osteoblastogenesisand osteoclastogenesis, with the last component probablyrequiring additional stimuli such as 1,25(OH)2D3, PTH, orgp130-activating cytokines.

In conclusion, the evidence presented in this study sug-gests that the balance between BMPs and their antagonistsis a critical determinant of osteoblastogenesis as well asosteoclastogenesis, and thereby the rate of bone remodeling.It is therefore possible that novel therapies for bone diseasesmay be developed, based on the manipulation of the balancebetween these proteins.

ACKNOWLEDGMENTS

The authors thank Drs. A. Michael Parfitt, George Yan-copoulos, and David Roodman for critical review of themanuscript before its submission; Catherine Smith for tech-nical assistance; and Tonya Smith for help with the prepa-ration of the manuscript. This work was supported by theNational Institutes of Health (grants P01 AG/AR13918 andR01 AR43003) and the Department of Veterans Affairs.

REFERENCES

1. Manolagas SC, Jilka RL 1995 Bone marrow, cytokines, andbone remodeling: Emerging insights into the pathophysiologyof osteoporosis. N Eng J Med332:305–311.

2. Parfitt AM 1994 Osteonal and hemi-osteonal remodeling: Thespatial and temporal framework for signal traffic in adulthuman bone. J Cell Biochem55:273–286.

3. Jilka RL, Hangoc G, Girasole G, Passere G, Williams DC,Abrams JS, Boyce B, Broxmeyer H, Manolagas, SC 1992Increased osteoclast development after estrogen loss: Media-tion by interleukin-6. Science257:88–91.

4. Jilka RL, Weinstein RS, Takahashi K, Parfitt AM, ManolagasSC 1996 Linkage of decreased bone mass with impairedosteoblastogenesis in a murine model of accelerated senes-cence. J Clin Invest97:1732–1740.

5. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC 1998Inhibition of osteoblastogenesis and promotion of apoptosis ofosteoblasts and osteocytes by glucocorticoids: Potential mech-anisms of their deleterious effects on bone. J Clin Invest102:274–282.

6. Parfitt AM 1990 Bone forming cells in clinical conditions. In:Hall BK (ed.) Bone, The Osteoblast and Osteocyte, vol. 1.Telford and CRC Press, Boca Raton, FL, U.S.A., pp. 351–429.

7. Mundy GR 1996 Role of cytokines, parathyroid hormone andgrowth factors in malignancy. In: Bilezekian JP, Raisz LG,Rodan GA (eds.) Principles of Bone Biology. Academic Press,San Diego, CA, U.S.A., pp. 827–836.

8. Jilka RL, Takahashi K, Munshi M, Williams DC, RobersonPK, Manolagas SC 1998 Loss of estrogen upregulates osteo-blastogenesis in the murine bone marrow: Evidence for auton-omy from factors released during bone resorption. J ClinInvest101:1942–1959.

9. Hughes DE, Dai A, Tiffee JC, Li HH, Mundy GR, Boyce BF1996 Estrogen promotes apoptosis of murine osteoclasts me-diated by TGF-beta. Nat Med2:1132–1136.

10. Rosen V, Cox K, Hattersley G 1996 Bone morphogeneticproteins. In: Bilezikian JP, Raisz LG, Rodan GA (eds.) Prin-ciples of Bone Biology. Academic Press, San Diego, CA,U.S.A., pp. 661–671.

11. Kirker-Head CA, Gerhart TN, Schelling SH, Henng GE, WangE, Holtrop ME 1995 Long-term healing of bone using recom-binant human bone morphogenetic protein 2. Clin Orthop318:222–230.

12. Nifuji A, Kellermann O, Kuboki Y, Wozney JM, Noda M1997 Perturbation of BMP signaling in somitogenesis resultedin vertebral and rib malformations in the axial skeletal forma-tion. J Bone Miner Res12:332–334.

13. Nishimura R, Kato Y, Chen D, Harris SE, Mundy GR, YonedaT 1998 Smad5 and DPC4 are key molecules in mediatingBMP-2-induced osteoblastic differentiation of the pluripotentmesenchymal precursor cell line C2C12. J Biol Chem273:1872–1879.

14. Ducy P, Zhang R, Geoffriy Y, Ridall AL, Karsenty G 1997Osf2/cbfa1: A transcriptional activator of osteoblast differen-tiation. Cell89:747–754.

15. Komori TH, Yagi H, Nomura S, Yamaguchi A, Sasaki K,Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, SatoM, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T 1997Targeted disruption of cbfa1 results in a complete lack of boneformation owing to maturational arrest of osteoblasts. Cell89:755–764.

16. Piccolo S, Sasai Y, Lu B, De Robertis EM 1996 Dorsoventralpatterning inXenopus:Inhibition of ventral signals by directbinding of chordin to BMP-4. Cell86:589–598.

17. Zimmerman LB, DeJesus-Escobar JM, Harland RM 1996 TheSpemann organizer signal, noggin binds and inactivates bonemorphogenetic protein 4. Cell86:599–606.

672 ABE ET AL.

18. Holley SA, Neul JL, Attisano L, Wrana JL, Sasai Y, O’ConnorMB, De Robertis EM, Ferguson EL 1996 TheXenopusdor-salizing factor noggin ventralizesDrosophilaembryos by pre-venting DPP from activating its receptor. Cell86:607–617.

19. Re’em-Kalma Y, Lamb T, Frank D 1996 Competition betweennoggin and bone morphogenetic protein 4 activities may reg-ulate dorsalization duringXenopusdevelopment. Proc NatlAcad Sci USA92:12141–12145.

20. Valenzuela DM, Economides AN, Rojas E, Lamb TM, NumezL, Jones P, Ip NY, Espinosa III R, Brannan CI, Gilbert DJ,Copeland NG, Jenkins NA, Le Beau MM, Harland RM, Yan-copoulos GD 1995 Identification of mammalian noggin and itsexpression in the adult nervous system. J Neurosci15:6077–6084.

21. Bouwmeester T, Kim S, Sasai Y, Lu B, De Robertis EM 1996Cerberus is a head-inducing secreted factor expressed in theanterior endoderm of Spemann’s organizer. Nature382:595–601.

22. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC 1997 Theeffect of androgen deficiency on murine bone remodeling andbone mineral density are mediated via cells of the osteoblasticlineage. Endocrinology138:4013–4021.

23. Suda T, Takahashi N, Martin TJ 1995 Modulation of oste-oclast differentiation: Update 1995. Endocr Rev4:266–270.

24. Lacey DL, Timms E, Tan H-L, Kelley MJ, Dunstan CR,Burgess T, Elliot R, Colombero A, Elliot G, Scully S, Hsu H,Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, QianY-X, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J,Delaney J, Boyle WJ 1998 Osteoprotegerin ligand is a cyto-kine that regulates osteoclast differentiation and activation.Cell 93:165–176.

25. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M,Mochizuki S-I, Tomoyasu A, Yano K, Goto M, Murakami A,Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N,Suda T 1998 Osteoclast differentiation factor is a ligand forosteoprotegerin/osteoclastogenesis-inhibitory factor and isidentical to TRANCE/RANKL. Proc Natl Acad Sci USA95:3597–3602.

26. Nagasaki N, Kinoshita M, Yamaguchi K, Shima N, Yasuda H,Yano K, Morinaga T, Higashi K 1998 RANK is the essentialsignaling receptor for osteoclast differentiation factor in oste-oclastogenesis. Biochem Biophys Res Commun252:395–400.

27. Katagiri T, Yamaguchi A, Komaki M, Abe E, Takahashi N,Ikeda T, Rosen V, Wozney JM, Fujisawa-Sehara A, Suda T1994 Bone morphogenetic protein-2 converts the differentia-tion pathway of C2C12 myoblasts into the osteoblast lineage.J Cell Biol 127:1755–1766.

28. Lecka-Czernik B, Gubrij I, Moerman EA, Kajkenova O, Lip-schitz DA, Manolagas SC, Jilka RL 1999 Inhibition of Osf2/Cbfa 1 expression and terminal osteoblast differentiation byPPARg2. J Cell Biochem74:357–371.

29. Lee KC, Tak KK, Park MO, Lee JT, Woo BH, Yoo SD, LeeHS, Deluca PP 1999 Preparation and characterization ofpolyethylene-glycol-modified salmon calcitonins. Pharm DevTech4:269–275.

30. Lin S-C, Yamate T, Taguchi Y, Borba V, Girasole G, O’BrienCA, Bellido T, Abe E, Manolagas SC 1997 Regulation of the

gp130 subunits of the IL-6 receptor by sex steroids in themurine bone marrow. J Clin Invest100:1980–1990.

31. Inada M, Katagiri T, Akiyama S, Namiki M, Komaki M,Yamaguchi A, Komori K, Rosen V, Suda T 1996 Bone mor-phogenetic protein-12 and -13 inhibit terminal differentiationof myoblasts, but not induce their differentiation into osteo-blasts. Biochem Biophys Res Commun222:317–322.

32. Yamate T, Mocharla H, Taguchi Y, Igietseme JU, ManolagasSC, Abe E 1997 Osteopontin expression by osteoclast andosteoblast progenitors in the murine bone marrow: Demonstra-tion of its requirement for osteoclastogenesis and its increasefollowing ovariectomy. Endocrinology138:3047–3055.

33. Kanatani M, Sugimoto T, Kobayashi T, Nishiyama K, FukaseM, Kumegawa M, Chihara K 1995 Stimulatory effect of bonemorphogenetic protein-2 on osteoclast-like cell formation andbone-resorbing activity. J Bone Miner Res10:1681–1190.

34. Hentunen TA, Lakkakorpi PT, Tuukkanen J, Lehenkari PP,Sampath TK, Vaananen HK 1995 Effects of recombinanthuman osteogenic protein-1 on the differentiation ofosteoclast-like cells and bone resorption. Biochem BiophysRes Commun209:433–443.

35. Gbriji I, Lecka-Czrnik B, Kajkenova O, O’Brien CA, Lips-chitz D, Manolagas SC, Jilka RL 1997 Isolation of clonal celllines with distinct differentiation potential of the osteoclasts,adipocytes, or both: Association of the osteoclast supportfunction with bipotentiality [abstract]. J Bone Miner Res12:S183.

36. Brunet LJ, McMahon JA, McMahon AP, Harland RM 1998Noggin, cartilage morphogenesis, and joint formation in themammalian skeleton. Science280:1455–1457.

37. Gazzerro E, Gangji V, Canalis E 1998 Bone morphogeneticproteins induce the expression of noggin, which limits theiractivity in cultured rat osteoblasts. J Clin Invest102:2106–2114.

38. Kimble RB, Liu X, Gannon FA, Caplan FS, Shore EM, Econo-mides AN, Stahl N 1998 Noggin inhibits bone formation in amodel of heterotopic ossification [abstract]. J Bone Miner Res12:S244.

39. O’Brien CA, Farrar NC, Manolagas SC 1998 Identification ofan OSF-2 binding site in the murine RANKL/OPGL genepromoter: A potential link between osteoblastogenesis andosteoclastogenesis [abstract]. J Bone Miner Res12:S149.

Address reprint requests to:Etsuko Abe, Ph.D.

Division of Endocrinology and Metabolism, Slot 587University of Arkansas for Medical Sciences

4301 West Markham StreetLittle Rock, AR 72205-7199, U.S.A.

Received in original form June 23, 1999; in revised form Novem-ber 12, 1999; accepted December 10, 1999.

673BMPs, NOGGIN, AND BONE CELL FORMATION