Fibronectin regulates calvarial osteoblast differentiation

1369Journal of Cell Science 109, 1369-1380 (1996)Printed in Great Britain © The Company of Biologists Limited 1996JCS4126

Fibronectin regulates calvarial osteoblast differentiation

Amr M. Moursi1, Caroline H. Damsky1,2,*, Jonathan Lull3,4, Deborah Zimmerman1, Stephen B. Doty5,Shin-ichi Aota6 and Ruth K. Globus3,4

1Department of Stomatology, University of California, San Francisco, CA 94143-0512, USA2Department of Anatomy, University of California, San Francisco, CA 94143, USA3Department of Medicine, University of California, San Francisco, CA 94143, USA4NASA-Ames Research Center, Moffett Field, CA, USA5Hospital for Special Surgery, New York, NY, USA6National Institute of Dental Research, Bethesda, MD, USA

*Author for correspondence at address 1

The secretion of fibronectin by differentiating osteoblastsand its accumulation at sites of osteogenesis suggest thatfibronectin participates in bone formation. To test thisdirectly, we determined whether fibronectin-cell interac-tions regulate progressive differentiation of cultured fetalrat calvarial osteoblasts. Spatial distributions of α5integrin subunit, fibronectin, osteopontin (bone sialopro-tein I) and osteocalcin (bone Gla-protein) were similar infetal rat calvaria and mineralized, bone-like nodulesformed by cultured osteoblasts. Addition of anti-fibronectin antibodies to cultures at confluence reducedsubsequent formation of nodules to less than 10% ofcontrol values, showing that fibronectin is required fornormal nodule morphogenesis. Anti-fibronectin antibodiesselectively inhibited steady-state expression of mRNA forgenes associated with osteoblast differentiation; mRNAlevels for alkaline phosphatase and osteocalcin were sup-pressed, whereas fibronectin, type I collagen and osteo-pontin were unaffected. To identify functionally relevantdomains of fibronectin, we treated cells with solublefibronectin fragments and peptides. Cell-bindingfibronectin fragments (type III repeats 6-10) containing the

Arg-Gly-Asp (RGD) sequence blocked both noduleinitiation and maturation, whether or not they contained afunctional synergy site. In contrast, addition of the RGD-containing peptide GRGDSPK alone did not inhibit noduleinitiation, although it did block nodule maturation. Thus,in addition to the RGD sequence, other features of the largecell-binding fragments contribute to the full osteogeniceffects of fibronectin. Nodule formation and osteoblastdifferentiation resumed after anti-fibronectin antibodies orGRGDSPK peptides were omitted from the media, showingthat the inhibition was reversible and the treatments werenot cytotoxic. Outside the central cell-binding domain,peptides from the IIICS region and antibodies to the Nterminus did not inhibit nodule formation. We concludethat osteoblasts interact with the central cell-bindingdomain of endogenously produced fibronectin during earlystages of differentiation, and that these interactionsregulate both normal morphogenesis and gene expression.

Key words: Fibronectin, Osteoblast differentiation, RGD sequence,Fetal rat calvarial osteoblast

SUMMARY

INTRODUCTION

Osteogenesis involves the recruitment of mesenchymal cells tothe osteoblast lineage and their progressive differentiation toproduce a mineralized extracellular matrix (ECM). StructuralECM proteins, cytokines sequestered in the ECM, and ECM-degrading proteases and their inhibitors are all potential regu-lators of osteoblast differentiation and function. The observa-tion that the production and composition of ECM are carefullymodulated during osteoblastic differentiation (Stein et al.,1990) suggests an important regulatory role for osteoblast-ECM interactions. The functional significance of these inter-actions, however, is poorly understood.

One ECM protein that may provide information toosteoblasts during their differentiation is fibronectin (FN). FN

is a heterodimeric ECM glycoprotein that has several cell- andmatrix-binding domains (Fig. 1) (Hynes, 1990). Sequences inat least three different regions of FN bind to the cell surfacevia two classes of transmembrane receptors: integrins and cell-surface proteoglycans (Damsky and Werb, 1992; Hynes,1992). The arginine-glycine-aspartic acid (RGD) sequence,present in FN type III repeat 10 (FNIII10) of the central cell-binding region, recognizes several members of the integrinfamily. These include the specific integrin FN receptor α5β1as well as several αv-containing integrins, which also interactwith other RGD-containing ECM components besides FN. Asecond sequence in FNIII9, referred to as the synergy sequence,has been found to selectively enhance binding of the FN RGDsequence to α5β1 (Bowditch et al., 1994; Aota et al., 1994).The C-terminal heparin-binding region of FN (FNIII11-14)

1370 A. M. Moursi and others

contains several sequences that bind to transmembrane cell-surface proteoglycans, including CD44 and members of thesyndecan family (Iida et al., 1992; Jalkanen and Jalkanen,1992). This region also has low-affinity binding activity for theα4β1 integrin (Mould and Humphries, 1991). Finally, thevariably spliced IIICS connecting sequence contains the CS-1high-affinity binding sequence for α4β1 (Komoriya et al.,1991). Distinct functions for several of these cell-binding FNdomains have been demonstrated. For example, the centralcell-binding and heparin-binding regions of FN cooperate topromote the formation of focal contact sites in cultured fibro-blasts (Woods et al., 1993). The central cell-binding region andthe CS-1 sequence cooperate to establish a low basal level ofcollagenase expression in synovial fibroblasts (Huhtala et al.,1995). These and other studies in many systems have shownthat cell interactions with intact FN, or specific FN fragments,can initiate signals that affect cytoskeletal organization, cellmotility, tissue-specific gene expression, and matrix remodel-ing (Adams and Watt, 1993; Damsky and Werb, 1992;Homandberg et al., 1993; Huhtala et al., 1995; Juliano andHaskill, 1993; Werb et al., 1989).

The distribution of FN in areas of skeletogenesis suggeststhat it may be involved in early stages of bone formation. FNis detected in the periosteum of cultured rat calvaria(Gronowicz et al., 1991), where it could participate in regula-tion of osteoblast recruitment and commitment to terminaldifferentiation. FN is also localized to the non-mineralized,dense ECM of osteoid surrounding implants (Weiss and Reddi,1980), where it could help to regulate matrix assembly. A rolein matrix organization is also suggested by ultrastructuralimmunolocalization studies which demonstrate an associationof FN with individual type I collagen fibrils in the matrix ofmature bone tissue during endochondral ossification in youngrats (Nordahl et al., 1995). In culture, FN is synthesized bychicken osteoblasts and accumulates in the ECM during cellproliferation and early differentiation; synthesis is sharplyreduced as cells mature (Winnard et al., 1995). In addition, theexpression of FN, as well as of type I collagen, increases duringthe early stages of differentiation in rat calvarial osteoblasts(Stein et al., 1990). Furthermore, factors known to regulateosteoblast differentiation, such as parathyroid hormone,estrogen, glucocorticoids, 1, 25-dihydroxy vitamin D3, andtransforming growth factor beta (TGF-β), affect both FNexpression and osteoblast attachment to FN (Breen et al., 1994;Eielson et al., 1994; Franceschi et al., 1987; Gronowicz et al.,1991; Gronowicz and McCarthy, 1995; Shalhoub et al., 1992).

Integrins are likely to transduce the signals generated by FN.Although there are conflicting data, osteoblasts in normalhuman bone have been reported to express integrin receptorsfor FN; α3β1 (Clover et al., 1992), α4β1 (Grzesik and Robey,1994), α5β1 (Hughes et al., 1993; Grzesik and Robey, 1994)and αvβ3 (Grzesik and Robey, 1994). Similarly, normal humanosteoblasts in culture express α3β1 (Clover et al., 1992; Saitoet al., 1994), α4β1 (Grzesik and Robey, 1994), α5β1 (Grzesikand Robey, 1994; Saito et al., 1994) and αvβ3 (Grzesik andRobey, 1994). In addition, αvβ5 has been identified in theosteoblast cultures (Saito et al., 1994) but was not localized toosteoblasts in human tissue (Grzesik and Robey, 1994). In rat,the FN receptors αvβ3 and αvβ5 were localized to normal bonetissue (Hultenby et al., 1993) and α5β1 was identified in normalosteoblast cultures (Brighton and Albelda, 1992).

These correlative data suggest the hypothesis that interac-tion of osteoblasts with FN generates signals that are requiredfor normal morphogenesis and gene regulation during osteo-genesis and bone remodeling. To test this hypothesis directlyand to study the role of FN in osteoblast differentiation, weused a well-characterized rat calvarial osteoblast model(Bellows et al., 1986; Stein et al., 1990) in which mineralized,bone-like nodules form. We disrupted FN-osteoblast interac-tions in these cultured osteoblasts by using anti-FN antibodiesand fragments of FN. Here we report that perturbing interac-tions between osteoblasts and the central cell-binding regionof endogenously produced FN inhibited mineralized nodulemorphogenesis and gene expression in vitro. These findingsprovide novel evidence that FN plays a role in regulatingosteoblast differentiation.

MATERIALS AND METHODS

CellsOsteoblasts were isolated from 21-day-old fetal rat calvariae asdescribed by Bellows et al. (1986), with several modifications.Following an initial treatment of calvariae for 10 minutes at 37°C with570 U/ml collagenase (Worthington Biochemical Corp., Freehold,NJ), the cells released from calvariae by two 10 minute, and two 20minute sequential collagenase digestions were pooled, and thenfiltered through both 100 µm and 37 µm Nitex filters. Cells weregrown overnight at 25,000 cells/cm2 on plastic dishes (Corning,Corning, NY). The cells were then detached using 0.05% trypsin and0.53 mM EDTA in Hanks’ buffered salt solution (Gibco, GrandIsland, NY). For the differentiation experiments, cells were plated ata density of 36,000 cells/cm2 in 35 mm dishes or 24-well plates coatedwith 0.2% gelatin (bovine skin, type B, Sigma Chemical Co., St Louis,MO). For some experiments, cells were plated on 8-well chamberslides, 0.81 cm2 per well, (Permanox, Nunc Inc., Naperville, IL) thatwere coated with 0.2% gelatin cross-linked with cyanamide (Sigma)(Macklis et al., 1985) as previously described (Globus et al., 1989).The cells were grown in alpha-Minimum Essential Medium supple-mented with 10% heat-inactivated fetal calf serum (Gibco). After con-fluence (3 days) the medium was further supplemented with freshlyprepared ascorbic acid (50 µg/ml) and β-glycerophosphate (3 mM)(Sigma) to trigger differentiation. Addition of this level of β-glyc-erophosphate, which is lower than that used in most previous studies,promotes both high cell viability over the 2-4-week culture period andan orderly deposition of mineral extracellularly, in association withfibrillar collagen (see Fig. 2E). Unless otherwise noted, antibodies,fragments, or small peptides were added with every medium change(every 2-3 days) until termination of the experiment. Multiple con-centrations of each reagent were tested and the minimum doseproviding maximal inhibition was used throughout.

ECM ligands and antibodiesGRGDSPK and GRADSP peptides, the 120 kDa central cell-bindingfragment of FN (120 FN) consisting of approximately FN type IIIrepeats 3-11, and rabbit anti-rat FN antiserum were purchased fromGibco/BRL (Gaithersburg, MD). An additional rabbit anti-rat FNantiserum was purchased from Chemicon International Inc.(Temecula, CA). The IgG fractions of the anti-FN antisera werepurified by binding to Protein A-Sepharose (Pharmacia Biotech Inc.,Piscataway, NJ) and eluting at pH 3.0. Shorter central cell-binding FNfragments containing FN type III repeats 6-10 (FNIII6-10) with, andwithout, a 16-amino-acid substitution at the synergy site wereprepared as described previously (Aota et al., 1994). Rabbit anti-ratosteocalcin antiserum was kindly provided by Dr K. Nishimoto (Uni-versity of Tennessee, Memphis, TN). Monoclonal mouse anti-rat

1371Fibronectin regulates calvarial osteoblast differentiation

osteopontin antibody (MP111B10) was obtained from the Develop-mental Studies Hybridoma Bank (University of Iowa, Iowa City, IA).Monoclonal mouse anti-rat FN antibody (BR5.4) was kindly providedby Dr Paul Johnson (University of California, San Francisco). Rabbitanti-mouse α5 antibody was obtained from Chemicon InternationalInc. The CS-1 peptide (DELPQLVTLPHPNLHGPEILDVPST) andtwo control peptides, CS-1C (DELPQLVTLPHPNLHGPDSTEVPIL)and CS-1S (LTEHTHQLPLEPVDGVLSDDPPPLI), were syn-thesized at the Howard Hughes Medical Institute (University of Cal-ifornia, San Francisco). They were conjugated to ovalbumin asdescribed previously (Huhtala et al., 1995). Rabbit anti-humanantibody to the 70 kDa amino-terminal region of FN was provided byDr Deane Mosher (University of Wisconsin, Madison, WI).

HistologyCalvaria from 21-day-old rat fetuses were imbedded in Tissue TekOCT (Miles, Inc., Elkhart, IN) and cryosectioned at 8 µm. Tissuesections were then fixed in 4% paraformaldehyde in PBS for 30minutes. Cells prepared for immunocytochemistry were grown onplastic chamber slides, fixed with 4% formaldehyde, permeabilizedwith cold methanol, and then rinsed three times with PBS. Sampleswere incubated with 5% sucrose in PBS and then carefully removedas an intact sheet, embedded in Tissue Tek OCT, frozen in isopen-tane in a liquid nitrogen bath, and cryosectioned at 8 µm. Cellsprepared for toluidine blue staining and electron microscopy werefixed with a solution of 2% paraformaldehyde, 0.5% glutaraldehydein 0.05 M cacodylate buffer, pH 7.4, for 12-18 hours at 4°C. The cellsand their associated matrix were removed from the dishes/slides andplaced into 2% aqueous osmium tetroxide for 1 hour, dehydrated ina graded series of alcohols, and embedded in Spurr’s or Araldite resin.Sections 1 µm thick were obtained from each block and stained with1% toluidine blue containing sodium borate at 55°C. For electronmicroscopy, sections approximately 70-80 nm thick were collected onwater containing bromothymol blue with pH adjusted to 8.0 or above.This prevented loss of mineral from the thin section into the collect-ing water. These sections were counter-stained with lead citrate anduranyl acetate and examined in a Philips CM-12 electron microscope.

ImmunocytochemistryCryostat sections of fixed calvarial tissue and osteoblast cultures werewashed with PBS, incubated with 0.1 M glycine (Sigma) in PBS, andthen incubated for 1 hour in a blocking solution containing 0.5%casein (Sigma), 5.0% BSA (Sigma), 0.1% Tween-20 (Fisher Scien-tific, Fair Lawn, NJ) and 2.0% donkey serum (Jackson ImmunoRe-search, West Grove, PA). Slides were incubated in primary antibodydiluted in the blocking solution (anti-α5 anti-serum, 1:100; mouseanti-rat FN ascites, 1:200; anti-osteopontin hybridoma supernatant1:1; anti-osteocalcin anti-serum 1:100) overnight at 4°C. Next, slideswere washed in PBS for 30 minutes, incubated in blocking solutionfor 1 hour, and then incubated for 30 minutes at room temperaturewith rhodamine-conjugated goat anti-mouse or donkey anti-rabbitsecondary antibodies (Jackson ImmunoResearch) diluted 1:200 in theblocking solution. Finally, slides were washed with PBS for 15minutes and coverslips were placed with Aqua Polymount (Poly-sciences Inc., Warrington, PA).

Quantification of nodule formationTo calculate the surface area of the osteoblast cultures occupied bynodules, a photographic image of the central 80% of the entire wellwas captured with a ScanMaker 1850S scanner (Microtek, Taiwan,Rep. of China) linked to a Macintosh IIci computer (Apple ComputerInc., Cupertino, CA) and then analyzed with NIH Image version 1.57(National Institutes of Health, public domain). Since the nodules weremore phase dense than the surrounding inter-nodular areas we wereable to determine nodule surface area by measuring regions in theculture that displayed a density greater than a set threshold. Thisthreshold value was determined and verified by comparing threshold-

derived surface areas to manually-circumscribed surface areas. Thistechnique was reproducible between investigators and duplicate wells.Surface areas were calculated from the average of three to four inde-pendent experiments containing duplicate dishes or wells ± s.e.m. Pvalues were determined by Student’s t-test.

Northern analysis Total RNA (10 µg/lane) was fractionated on 1% agarose/formalde-hyde gels, blotted to a nylon membrane (Hybond N, Amersham,Arlington Heights, IL), and then incubated at 68°C with an [α-32P]dCTP-labeled cDNA probe (see below) in hybridization solution(QuikHyb, Stratagene, La Jolla, CA) for 1 hour. The final wash of thefilters was at high stringency in 0.3 M sodium chloride, 30 mMsodium citrate, 0.1% sodium dodecyl sulfate, pH 7.0 at 60°C for 30minutes. The filters were exposed to film (X-OMAT AR, EastmanKodak Co., Rochester, NY) for 1-9 days with enhancer screens. Uni-formity of sample loading was confirmed by densitometric scanningof the 28 S ribosomal RNA bands stained on the gel with acridineorange. Densitometric analysis of the steady-state expression ofmRNA was performed by capturing the video image of the autoradi-ograph and then analyzing the bands using NIH Image version 1.57on a Macintosh IIci computer. The data shown are the means of twoto three experiments ± s.e.m.

The following cDNA fragments were used for probing the northernblots. A 363-bp cDNA encoding a fragment of rat osteocalcin (boneγ-carboxyglutamic acid protein) (Pan and Price, 1985) was a gift fromP. Price (University of California, San Diego, CA). A 520 bp cDNA(pRAP-1) for rat alkaline phosphatase (Noda et al., 1987) was a giftfrom M. Noda (Tokyo Medical and Dental University, Tokyo) and G.Rodan (Merck Research Labs, West Point, PA). A 1,600 bp cDNA(pα1R1) encoding a fragment of the α1(I) chain of rat procollagen(Genovese et al., 1984) was provided by B. Kream (University ofConnecticut, CT). A 350 bp cDNA for rat osteopontin cDNA (Ridallet al., 1995) was a gift from A. Ridall (University of Texas HealthScience Center, Houston, TX). A 100 bp cDNA for rat FN (Peters etal., 1995) was provided by P. Johnson (University of California, SanFrancisco).

Cell numberCell number was determined on day 10. Cells were washed with Ca2+-and Mg2+- free PBS and treated with 1 mg/ml collagenase for 1 hourat 37°C. An equal volume of 0.05% trypsin was then added to the col-lagenase, and the cells were incubated in this mixture for an additional30 minutes. Detached cells were then agitated to obtain a cell sus-pension. Exclusion of 0.1% trypan blue was used to determine cellviability during counting. Cells from duplicate or triplicate wells ofthree independent experiments were counted using a hemocytometer.Data are reported as mean ± s.e.m.

RESULTS

A fetal rat calvarial osteoblast culture model ofosteogenesis preserves normal spatial regulation ofECM components, including FNWhen osteoblasts isolated from fetal rat calvariae are plated atnear confluent density, they proliferate and differentiate inculture over a 14-16 day period to form multilayered sheets ofcells that include prominent mineralized nodules (Bellows etal., 1986; Stein et al., 1990). Cells isolated as described inMaterials and Methods formed confluent monolayers within 3days (Fig. 2A), at which time ascorbic acid and β-gly-cerophosphate were added. By days 6-8, discrete clusters ofcells first appeared that possessed a markedly rounder shapethan surrounding cells (nodule initiation) (Fig. 2B). These


COOH

COOH

NH 2

NH 2

SS

CollagenGelatin

RGDMatrixassembly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

α5β1

Type I repeat Type II repeat Type III repeat

CS-1

HBD IIICS

HBD

CBD

Sy

Fig. 1. Structure of fibronectin showingcell- and matrix-binding domains. CBD,cell-binding domain; Sy, ‘synergy’sequence; RGD, arginine-glycine-asparticacid; HBD, C-terminal heparin-bindingdomain; CS-1, connecting segmentsequence 1; IIICS, type III connectingsegment.

clusters of round cells subsequently became increasingly phasedense over the next week as observed by phase microscopy(nodule maturation), and mineral deposition was evident byday 14 (Fig. 2C), as shown by von Kossa staining (Fig. 2D).Fully mature nodules (assessed at days 20-29) shared importantfeatures of normal woven bone, as judged by electronmicroscopy. The osteoblast surface membrane was surroundedby an ECM rich in banded type I collagen. Areas of mineral-ization were associated with collagen fibrils (Fig. 2E, arrows),as is found in normal bone tissue.

To determine whether the spatially regulated distribution ofECM components characteristic of bone in vivo was preservedin this in vitro model, cross-sections of 21 day-old fetal ratcalvaria and osteoblast cultures were stained with antibodiesagainst the α5 integrin subunit of the FN receptor α5β1 andECM components characteristic of early and late stages ofosteogenesis. In the calvaria the strongest staining for α5 wasobserved in the periosteal surface and osteoblast layer adjacentto mineralized matrix. Less intense staining was detectedamong cells within the mineralized matrix (Fig. 3A). Similarly,in day 8 and day 14 osteoblast cultures (similar to culturesdepicted en face in Fig. 2B,C) α5 staining was most intense inthe internodular areas and the periphery of nodules with someintra-nodular cells also staining (Fig. 3C,E). In the case of FN,the predominant area of tissue staining was the periostealsurface adjacent to bone. Little staining was evident in the min-eralized matrix (Fig. 3B). This pattern was similar to thatdemonstrated by osteoblast cultures where FN was localizedto internodular areas and at the edges of nodules, but not withinthe nodule core (Fig. 3D,F), where mature osteocyte-like cellssurrounded by mineralized matrix were present. In contrast,staining for both osteopontin (Fig. 3G) and osteocalcin (Fig.3H), which are characteristic of mature, mineralized bonematrix, was restricted largely to the nodular region. Thisspatially regulated distribution of ECM components servesboth to validate the culture system as a suitable in vitro modelin which to study the role of FN-osteoblast interactions, and toreinforce the hypothesis that FN is likely to play a role in rel-atively early stages of osteogenesis.

Osteoblast interactions with FN are required fornodule formation in vitroTo determine whether there is a functional role for FN-osteoblast interactions in the progressive differentiation ofosteoblasts, we added a polyclonal anti-FN antibody (100µg/ml) or control rabbit IgG (100 µg/ml) to cultures ofconfluent day 3 osteoblasts. Ascorbate and β-glycerophosphate

were also added to the medium at this time to induce differ-entiation and mineralization. Antibody was replenished everyother day at each medium change. Samples were photographedevery other day and harvested at several time points to evaluateproliferation, extent of nodule formation, and the expression ofgenes associated with osteoblast differentiation.

In control, IgG-treated cultures, nodule initiation wasevident by phase or Hoffman optics microscopy by day 8 (Fig.4A). These nodules appeared increasingly phase-dense byphase contrast microscopy and mineralization could beobserved in the maturing nodules by day 14 (Fig. 4C) withoutstaining. In cultures treated with anti-FN antibody, noduleformation was inhibited to less than 10% of control values atboth early (day 10) and late (day 20) stages (Fig. 5). Anti-ratFN antibodies from two different commercial sourcesproduced similar results. To determine whether soluble FNpresent in fetal bovine serum affected the function of the anti-rat FN antibodies, we repeated the experiments using serumdepleted FN on a gelatin-Sepharose column. The results weresimilar (data not shown), suggesting that the antibody is inter-acting primarily with the FN produced by the rat osteoblasts,as opposed to the soluble FN in the fetal bovine serum addedto the medium.

To determine whether the inhibition of nodule formation byantibody treatment was reversible, cell cultures treated withanti-FN antibody from day 3 to day 14 were allowed to recoverfor an additional 6-9 days in medium that did not includeantibody. Nodule formation recovered significantly by day 20(Fig. 4H), when compared both with anti-FN antibody-treatedday 14 cultures (Fig. 4D) and with parallel cultures exposed toanti-FN antibody continuously through day 20 (Fig. 4F).

The effect of FN on nodule formation is a property ofthe central cell-binding domainThe data above documented an important role for osteoblast-FN interactions in regulating nodule morphogenesis. Todetermine which domain(s) in FN are required for normalnodule formation, we first added large fragments derived fromthe RGD-containing central cell-binding region of FN to themedium of confluent osteoblast cultures (day 3). Fragment wasreplenished every other day at each medium change. Additionof a 120 kDa fragment (120 FN), encompassing approximatelyFNIII3-11 (see Figs 1 and 9), at 100 µg/ml (0.83 µM) reducednodule formation to 10% of control values as assessed on day10 and day 20 (Fig. 5). Nodule formation resumed after thecultures were provided with medium without the fragment,although the extent of recovery was less consistent than that


Fig. 2. Fetal rat calvarial osteoblasts are a model for osteogenesis in culture. Isolated rat calvarial osteoblasts were plated on gelatin andreached confluence at day 3. Ascorbic acid and β-glycerophosphate were added at day 3 and cultures were photographed at the time pointsindicated. (A) Day 3, confluent monolayer of osteoblasts. (B) Day 8, nodule formation (arrow) was first detected after 6-8 days. (C) Day 14,nodules became increasingly phase dense, and staining with the von Kossa method demonstrated mineral deposition in the nodules (D).(E) Thin-section electron micrograph showing part of a mature osteoblast and its associated mineralized ECM after 29 days in culture. Notethe abundant banded type I collagen fibrils (arrows) and mineral deposition (m) associated with the collagenous ECM. Bars: (A-D) 100 µm;(E) 0.6 µm.

observed following treatment with anti-FN antibody (Fig. 5).The RGD sequence present in FNIII10 is the major cell-recog-nition motif in the cell-binding domain, although additionalsequences present in FNIII9, such as the synergy site, have

been shown to influence integrin receptor specificity andbinding affinity (Bowditch et al., 1994; Aota et al., 1994).Addition of the smaller cell-binding fragments FNIII6-10 at 200µg/ml (4.0 µM) inhibited nodule formation to 15% of control


(not shown), whether or not they contained a functionalsynergy site. These soluble fragments are likely to function byinterfering with interactions between osteoblasts and the cell-binding domain of intact FN present in their pericellular ECM.Therefore, these data suggest an important role for the centralcell-binding domain of FN in regulating osteoblast differen-tiation.

To determine effects on osteoblast differentiation of theRGD sequence alone, we began addition of the peptideGRGDSPK (100 µg/ml) to osteoblast cultures on either day 3or day 8. Peptide was replenished every other day at eachmedium change. GRGDSPK perturbed nodule formation inthis assay, when compared with a control peptide, GRADSP(100 µg/ml). However, the morphological effects were distinctfrom those produced by treatment with polyclonal anti-FNantibody (Fig. 6). When added at day 3, GRGDSPK did not

Fig. 3. Extracellular matrix components havedistinct distribution patterns in fetal ratcalvaria and differentiated osteoblastcultures. Cryostat cross-sections of fetal 21day-old calvaria were incubated withantibodies against the α5 integrin subunit (A)and FN (B). α5 staining was strongest in theperiosteal surface adjacent to bone (P). FNstaining was strongest in the periostealsurface adjacent to bone, with littlelocalization in the mineralized tissue (MN). 8day osteoblast cultures, corresponding tonodule initiation, were incubated withantibodies against the α5 integrin subunit (C)and FN (D). 14-day osteoblast cultures wereincubated with antibodies against α5 (E), FN(F), osteopontin (G) and osteocalcin (H). Incultured osteoblasts α5 staining was mostintense in the internodular (IN) regions andthe periphery of the nodules (N). Some cellswithin the nodule (N) also stained for α5. FNwas detected in the internodular region (IN)and around the periphery of the nodule (N),but not within the nodule itself. Staining forosteopontin and osteocalcin was confinedlargely to the core of the nodule (N). Allsamples were incubated with the appropriatesecondary antibodies conjugated torhodamine. MS, marrow space. Bars, 40 µm.

block nodule initiation as detected by the formation of clustersof rounded cells (Fig. 6B); however, the nodules that formedby day 8 were no longer evident by day 10. This inhibition ofnodule maturation was also observed when GRGDSPK wasadded at day 8, after nodule initiation had already occurred (notshown). In addition, the number of nodules detected byHoffman optics and phase microscopy (without staining) wasreduced to less than 5% of control values (Figs 5 and 6C,D).The peptide was omitted from the media of selected culturesat day 14 and the cultures were allowed to continue for an addi-tional 6-9 days. Nodule formation recovered significantly byday 20, when compared both with day 10 cultures and withparallel cultures exposed to RGD continuously through day 20(Fig. 6E-H). These data suggest that GRGDSPK acts after theinitiation of nodule formation, at a critical step in whichintegrin-mediated interactions are required for stabilization and


Fig. 4. Anti-FN antibody inhibits nodule formation. Cultures were incubated with control rabbit IgG (A,C,E,G) or rabbit anti-rat FN antibody(B,D,F,H) starting on day 3. Cultures were photographed on days 8 (A,B) using Hoffman optics, 10×. Day 14 (C,D) 4×; and day 20 (E-H) 1.5×cultures were photographed using phase optics, without staining. In control cultures, nodules were evident by day 8 and became increasinglyphase dense, whereas in experimental cultures anti-FN antibody suppressed nodule formation. (G,H) D20-R, cultures were treated with controlor anti-FN antibody from day 3 to day 14, then allowed to recover by omitting the antibody from the medium during days 14-20. Densenodules were able to form following removal of anti-FN antibody. Bars, 400 µm.

continued maturation of nodules. By contrast, the polyclonalanti-FN antibody blocks both initiation and maturation ofnodules.

To evaluate at higher resolution the effects of the RGDpeptide treatment on osteoblast differentiation and ECMorganization, we examined treated and control cultures histo-logically in toluidine blue-stained sections. Both control andGRGDSPK-treated cultures formed multilayered sheets ofcells (Fig. 7A,B). GRGDSPK-treated cultures had very fewnodules, and those that formed were small and poorlydeveloped (Fig. 7B). In contrast, only the control culturesformed numerous nodules containing large depositions offibrillar collagen (Fig. 7A, arrow), which on ultrastructuralobservation were shown to have a typical banded pattern (Fig.2E, arrows).

To determine whether cell number was affected by the anti-FN antibody or GRGDSPK peptide, we treated cultures on day3 then harvested them on day 10, when nodules were well-formed in control cultures and nodule inhibition was evident

in treated cultures. The numbers of cells in treated and controlcultures were not significantly different, indicating that anti-FN antibody and GRGDSPK were not toxic and that theirinhibitory effects on nodule formation did not reflect largechanges in cell proliferation or cell detachment.

To evaluate potential roles in osteoblast differentiation forFN domains outside the cell-binding region, we first used anantibody against the N-terminal 70 kDa region of FN (200µg/ml). This region includes the N-terminal heparin/fibrin-binding and matrix assembly region as well as the adjacentcollagen-binding domain of FN. The antibody against thisregion blocks FN matrix assembly in fibroblasts (Mosher et al.,1991). However, we found that this antibody had no effect onthe formation of nodules by osteoblasts (not shown), suggest-ing that the N-terminal 70 kDa region of FN is not required forthis process. Similarly, addition of the CS-1 peptide (200µg/ml) from the variably spliced IIICS domain that interactswith α4β1 integrin also did not affect nodule formation. Thispeptide is active, however, in regulating the expression of col-


0

2

4

6

8

% S

UR

FA

CE

AR

EA

D10 D20 D20 Rec.

Control

Anti-FNAntibody

RGDPeptide

120FN

*

*

**

*

*

*

*

# *#

Fig. 5. Anti-FN antibody, 120 FN fragment, and GRGDSPK peptidereduce surface area of mineralized nodules. Live cultures werephotographed and the percentage surface area of the well (0.81 cm2)containing mineralized nodules was calculated at the time pointsindicated. Control, combined data from IgG-, BSA- and GRADSP-treated cultures, all of which displayed similar values. D20 Rec., day20 cultures were allowed to recover for 6 days with antibody orpeptide omitted from the medium. Asterisk denotes statisticallysignificant difference (P<0.05) between treated and control cultures;# denotes statistically significant difference (P<0.05) between D20and D20 Rec. Results are the average of three to four independentexperiments ± s.e.m.

lagenase in response to FN in synovial fibroblasts (Huhtala etal., 1995). Taken together, these results indicate that the centralcell-binding domain has a preeminent role in regulating nodulemorphogenesis.

The disruption of osteoblast-FN interactionsreduces expression of genes characteristic ofdifferentiating osteoblasts Our results thus far indicated that perturbing osteoblast-FNinteractions interferes with the morphogenesis of mineralizednodules. To determine whether interfering with FN- and RGD-dependent interactions also alters the pattern of geneexpression characteristic of osteoblasts, we used northernblotting to assess the steady-state expression of mRNA forgenes characteristic of early (alkaline phosphatase, FN andtype I collagen) or late (osteopontin, osteocalcin) stages ofdifferentiation. Data are presented as the values for treatedcultures as a percentage of control at the indicated time points(Fig. 8A,C). Alkaline phosphatase (AP) levels in controlcultures were elevated by day 3 (not shown) and remained highat day 10. When assessed on day 10, AP levels in culturestreated with anti-FN antibody (100 µg/ml) were 25% of controllevels (Fig. 8A), indicating that anti-FN activity suppressed APexpression. Consistent with the results of others (Stein et al.,1990), osteocalcin (OC) expression was low in control culturesthrough day 10 and then rose at later time points. Whenassessed on days 10 and 14, treatment with anti-FN antibodysuppressed the rise in expression of osteocalcin mRNA, to 25%and 20%, of IgG-treated controls, respectively (Fig. 8A,B).However, anti-FN treatment had no significant effect on theexpression of mRNA for type I collagen, osteopontin or FNitself (not shown). When anti-FN antibody was removed at day

14 and the cultures were allowed to recover, the mRNA levelsfor AP and OC increased to control levels by day 20, showingthat the inhibitory effects of the antibody were reversible (Fig.8A,B). Collectively, these results indicate that interfering withosteoblast-FN interactions disrupts morphogenetic processesinvolved in early nodule formation and selectively suppressesthe expression of specific genes associated with the osteoblastphenotype, while leaving others unaffected.

In contrast to the inhibitory effects of anti-FN antibody,GRGDSPK (100 µg/ml) treatment had only a modestinhibitory effect on the expression of AP and OC after 20 daysin culture (Fig. 8C). As with anti-FN antibody treatment,GRGDSPK treatment did not markedly inhibit expression ofFN, type I collagen or osteopontin. These differences betweenthe effects of anti-FN antibody and GRGDSPK on geneexpression during the early stages of nodule formation (day 10)are consistent with the finding that they exert distinct effectson nodule morphogenesis.

DISCUSSION

The data presented in this study strongly support our hypoth-esis that FN plays a crucial role in the progressive differen-tiation of osteoblasts. In order to inhibit interactions betweenFN and osteoblasts, we first used a polyclonal anti-FN antibodythat is expected to bind to multiple sites on FN. This antibodyinterfered with both the formation of mineralized nodules(nodule morphogenesis) and the expression of genes charac-teristic of the osteoblast phenotype. Since the cultures resumednodule morphogenesis once the antibody was removed, theinhibition was reversible and not cytotoxic. Furthermore,addition of a function-perturbing antibody (Young et al., 1994)to tenascin-c, an ECM component localized to developing boneand produced by osteoblasts (Mackie et al., 1992), did notinhibit nodule formation, though after day 14 some wellsshowed slight detachment of the entire multi-layered cell sheetfrom the surface. Nodules in these detached cultures appearednormal (data not shown). Thus, inhibition of nodule formationby anti-FN antibody was not simply a generic response to per-turbation of cell-ECM interactions; instead, we conclude thatblocking cell-FN interactions in osteoblast cultures inhibits oneor more steps required for differentiation to progress.

As an alternative approach to perturbing osteoblast interac-tions with FN, and to identify functionally relevant domains,we then used fragments of FN from the central cell-bindingregion (120FN). This region contains the RGD sequence inFNIII10, as well as sequences in FNIII9 reported to beimportant for optimal interaction of the central cell-bindingregion with specific integrin receptors (Bowditch et al., 1994),including α5β1 (Aota et al., 1994). Addition of either 120FN(corresponding approximately to FNIII3-11), or a smallerfragment containing FNIII6-10, to osteoblast cultures blockednodule formation (summarized in Fig. 9), indicating that thecentral cell-binding domain of FN plays an important role inregulating normal nodule morphogenesis. In contrast, additionof either an antibody against the N-terminal 70 kDa matrixassembly/collagen-binding region or the CS-1 peptide, whichinteracts with the α4β1 integrin, did not affect noduleformation (Fig. 9), suggesting that these domains of FN are notrequired for nodule morphogenesis. Not all FN domains were


Fig. 6. GRGDSPK peptide blocks nodule maturation, but not nodule initiation. Day 3 cultures were treated with control peptide GRADSP(A,C,E,G) or GRGDSPK (B,D,F,H). (A,B) Day 8, nodule initiation (arrows) was observed in both cultures. (C,D) Day 10, nodules continued toform and mature in the GRADSP-treated cultures (C), but nodules were no longer detected in the GRGDSPK-treated cultures (D). (E,F) Day20, nodules appeared increasingly phase dense in the GRADSP-treated cultures (E) but formation of dense nodules was not evident inGRGDSPK-treated cultures (F). (G,H) Day 20 Recovered. Peptides were omitted from the medium starting on day 14, and cultures werephotographed at day 20. Cultures originally treated with GRADSP and exposed to peptide-free medium for 6 days resembled those treatedcontinuously with this peptide (G). Cultures originally treated with GRGDSPK and allowed to recover for 6 days (H) were able to formnodules. Day 8 and 10 images are 10× using Hoffman optics and Day 20 and 20R images are 1.5× using phase microscopy, all without staining.Bars, 400 µm.

tested in this assay, so we cannot rule out important contribu-tions to nodule morphogenesis from regions outside the cell-binding domain, such as the C-terminal heparin-binding

Fig. 7. GRGDSPK peptide reduces nodulenumber and size. Cross sections ofglutaraldehyde-fixed, plastic embeddedday 19 osteoblast cultures treated with (A)control peptide GRADSP or (B)GRGDSPK starting at day 3. A prominentdensely stained nodule (A, arrow) isshown in the GRADSP-treated culturealong with adjacent multilayeredinternodular areas. Nodules were rarelydetected in the GRGDSPK-treated cultures(B), although multilayering of the cells isevident (insert). Bars, 100 µm.

domain. For example, it has been reported that humanosteoblasts attach to the heparin-binding domain as well as tothe cell-binding domain of FN (Dalton et al., 1995). However,


D8-10 D14-16 D20-23 D20-23 Rec.

0

10

20

30

40

50

60

70

80

90

100

AP APAPAPOC OC OC OC

110

A

% o

f Con

trol

Anti-FN/IgG

D8-10 D14-16 D20-23 D20-23 Rec.

0

20

40

60

80

100

120

140

AP APAPAPOC OC OC OC

C

% o

f Con

trol

RGD/RAD

Fig. 8. Expression of mRNA for osteoblast-associated genes in anti-FN antibody-treated (A) and GRGDSPK-treated cultures (C). TotalRNA was analyzed by northern blotting followed by densitometry.Results are depicted as a percentage of the control values (GRADSP-and IgG-antibody-treated) obtained from cultures at the time pointsindicated. Results are the average of at least two independentexperiments ± s.e.m. AP, alkaline phosphatase; OC, osteocalcin.(B) Northern analysis of IgG- and anti-FN antibody-treated culturesdemonstrating osteocalcin expression. 28 S ribosomal mRNA wasused to correct for discrepancies in loading. Day 23 Rec., day 23cultures were allowed to recover for 6 days with antibody or peptideomitted from the medium.

the finding that no effect was observed with addition of eitherthe anti-70 kDa antibody or the CS-1 peptide confirms that theinhibitory effects of the cell-binding fragments are specific andpoint to a preeminent role for this region in regulatingosteoblastic differentiation.

The RGD-containing peptide, GRGDSPK, also perturbedmorphogenesis, reinforcing our conclusion that the central cell-binding domain of FN plays a critical role in regulatingosteoblast differentiation in these cells. However, the effectsof GRGDSPK on nodule formation were distinct from those

of the polyclonal anti-FN antibody and the large cell-bindingfragments, suggesting that other sequences in the cell-bindingdomain may also contribute to the overall process of noduleformation. When this small peptide was added to osteoblastcultures on day 3, nodules started to form by day 6-8, asassessed by the appearance of organized clusters of round cells.With further continuous treatment with GRGDSPK, however,the clusters failed to mature into mineralized nodules and wereno longer visually evident by day 10 when viewed by Hoffmanoptics or phase microscopy. The ability of an RGD peptide toalter osteoblast morphology and disrupt nodule formation incultured osteoblasts is consistent with the results of Gronowiczand DeRome (1994), who showed that treatment of fetal ratcalvaria with GRGDSPK in tissue culture caused changes incell shape, disrupted the organization of the surface layer ofosteoblasts, and reduced the rate of bone formation.

Although GRGDSPK treatment did not block noduleinitiation, nodule maturation was inhibited whetherGRGDSPK or the 120 kDa fragment was added before (day 3)or after (day 6-8) the clusters of round cells first appeared (datanot shown). Thus, GRGDSPK appears to exert inhibitoryeffects at a later stage of cell differentiation than does the 120kDa fragment, although both ultimately suppress the formationof mature nodules. Taken together, these results suggest thatnodule morphogenesis is a multi-step process. The initiationstep, which entails changes in cell shape and organization,appears to be a reversible process and must be reinforced bysubsequent events that stabilize the nodule and promote furtherdifferentiation. A possible explanation for the discrepancybetween the onset of the inhibitory effects of GRGDSPK andthe larger cell-binding domain fragments may be that theaffinity of the larger fragments for their receptors is greaterthan that of the small RGD peptides. This is supported by datashowing that FN fragments containing a normal synergysequence for α5β1, present in FNIII9, in addition to the RGDsequence in FNIII10, bind with a higher affinity to α5β1 thanfragments in which the synergy sequence has been mutated(Aota et al., 1994). Our results, however, indicate that thecentral cell-binding fragment, even without a functionalsynergy site, is sufficient to inhibit nodule formation. It ispossible, though, that the larger cell-binding fragments are ableto assume a more native conformation and thereby interactmore strongly with their cell-surface receptors. In addition tothis difference in integrin- binding affinity, there may also exista difference in integrin-binding specificity. For example, thelarge cell-binding fragments may bind to a different repertoireof integrin FN receptors than the small RGD-containingpeptide (Pytela et al., 1985). Further studies will be requiredto sort out the relative importance of binding affinity andreceptor specificity in determining the distinct effects of smallRGD peptides and large RGD-containing FN fragments onnodule formation in osteoblast cultures.

Besides blocking nodule initiation and morphogenesis,treatment with anti-FN antibody suppressed steady-statemRNA levels of alkaline phosphatase and osteocalcin, twogenes characteristic of the osteoblast phenotype. Alkalinephosphatase expression was actively suppressed, since itsmRNA level was lower at day 10 in cultures treated with anti-FN antibody than it was at day 3 in control cultures. Anti-FNtreatment delayed the rise in expression of osteocalcin, a lateosteoblast marker. In contrast, anti-FN antibody had no sig-


+-

+

+

-

-

+-

+

+

+

-

Anti-70 kD

FN III

RGD

6-10

S y

Anti-FN

and

CS-1

S y RGD

Fragments

Peptides

Inhibits Nodule

Initiation

InhibitsNodule

Maturation

FN 120

Antibodies

GRGDSPK

CS-1

S y RGD

RGD S y CS-1Heparin/CollagenBinding

NH2 COOHFN Molecule

Fig. 9. Summary of the effects of anti-FN antibodies andFN fragments and peptides on nodule formation. Solid barrepresents FN molecule. Striped bars represent regions ofFN recognized by the antibodies listed. Open barsrepresent FN fragments and peptides. Anti-FN Ab, anti-fibronectin antibody; anti-70 kDa Ab, antibody to the N-terminal 70 kDa region of fibronectin; 120 FN, 120 kDafragment of fibronectin; FN III6-10, type III repeats 6-10 ofFN; Sy, ‘synergy’ sequence; RGD, arginine-glycine-aspartic acid; CS-1, connecting segment 1. See text formore information.

nificant effect on the expression of mRNA for FN, type Icollagen or osteopontin, important ECM components that arecharacteristic of the osteoblastic phenotype. Thus, the effectsof anti-FN antibody on osteoblast gene expression areselective. Taken together, these data suggest that addition ofanti-FN antibody initiates a shift in the pattern of geneexpression toward a less mature osteoblast.

Surprisingly, continuous treatment with GRGDSPK had noeffect on expression of mRNAs for any of the genes testedprior to day 20, despite the fact that nodules were not detectedbeyond day 10. These data suggest that, in contrast to nodulemorphogenesis, the program of osteoblast gene expression isrelatively stable in the continued presence of GRGDSPK. Fur-thermore, these results indicate that morphogenesis and bio-chemical differentiation of osteoblasts can be uncoupled byfactors that perturb those integrin-ECM interactions that areRGD-dependent.

Our results suggest a functional role for integrins as trans-ducers of the signals generated by FN in this culture model,since the central cell-binding domain of FN was critical forinhibition of osteoblast differentiation and since an RGDpeptide also suppressed nodule morphogenesis. We found thatthe distribution of the specific FN receptor α5β1 was similarto that reported by others in both human (Hughes et al., 1993;Grzesik and Robey, 1994; Saito et al., 1994) and rat (Brightonand Albelda, 1992) osteoblasts. Another FN receptor identifiedin human bone and cultured osteoblasts, α4β1(Grzesik andRobey, 1994), was not detected in either rat calvarial tissue orosteoblasts in culture. Future studies will be aimed at usingfunction-perturbing antibodies that recognize β1 and αv-asso-ciated integrin subunits in this normal rat osteoblast system.Using this approach in MG-63 human osteosarcoma cells, theα5β1 FN receptor was reported to promote osteoblastic differ-entiation (Dedhar, 1989; Dedhar et al., 1987, 1989). In thosestudies, amplification of α5β1 expression was associated withincreased mineralization in MG-63 cultures. In addition, anti-α5 and anti-β1 antibodies blocked induction of alkaline phos-phatase in response to IL-1 in MG-63 cells. Thus, the α5β1integrin is a likely candidate for signaling at least some of theresponses of rat calvarial osteoblasts to FN.

Evidence from these studies indicates that FN acts as anautocrine factor that provides either permissive or instructivecues required for osteoblast differentiation, although ourresults do not preclude a regulatory role for other bone matrixcomponents. Since FN expression appears to be highlylocalized to the bone surfaces in vivo and at the periphery ofnodules in vitro, it is possible that FN promotes a step in therecruitment or migration of osteoblast precursors. FN may alsopromote the synthesis and/or organization of an ECM by

osteoblasts that is permissive for signaling by growth factors,such as members of the insulin growth factor, TGF-β or fibro-blast growth factor families, which are known to reside in theECM of bone (Erlebacher et al., 1995). In addition, it ispossible that contact of osteoblasts with FN via integrinsincreases expression of TGF-β/bone morphogenetic proteinfamily members which in turn enhance osteoblast differen-tiation. Such cooperative signaling by ECM and growth/differ-entiation factors has been demonstrated elegantly in the caseof the mammary gland (Streuli et al., 1991). In that system, theexpression of milk proteins in pregnant mammary epitheliumrequires both basement membrane and lactogenic hormones.

In summary, this study demonstrates that osteoblast-FNinteractions are required for both morphogenesis and geneexpression in this model of osteoblast differentiation. Thereappear to be critical periods in the differentiation program ofosteoblasts during which specific adhesion receptor-FN inter-actions provide important regulatory signals required fornormal morphogenesis and gene expression. Further investi-gation of the role of osteoblast-FN interactions and how theymay collaborate with growth/differentiation factors shouldprovide important new insights into the regulation of tissue-specific differentiation in osteoblasts.

This work was supported by the National Institute for DentalResearch (P50-DE10306) and the National Aeronautic and SpaceAdministration (NAGW-4625; NCC 2-655). A. Moursi is supportedby a National Institute for Dental Research Training Grant (T32DE07204). D. Zimmerman is supported by a fellowship from theArthritis Foundation. We thank Drs E. Morey-Holton and A.Malouvier for their comments and support, Ms Evangeline Leash forediting the manuscript, and Mr Chanh Dinh for excellent assistancein preparing the figures. The authors appreciate gifts of antibodies,fragments and cDNA probes from colleagues as indicated in theMaterials and Methods.

REFERENCES

Adams, J. C. and Watt, F. M. (1993). Regulation of development anddifferentiation by the extracellular matrix. Development 117, 1183-1198.

Aota, S., Nomizu, M. and Yamada, K. M. (1994). The short amino acidsequence Pro-His-Ser-Arg-Asn in human fibronectin enhances cell-adhesivefunction. J. Biol. Chem. 269, 24756-24761.

Bellows, C. G., Aubin, J. E., Heersche, J. N. and Antosz, M. E. (1986).Mineralized bone nodules formed in vitro from enzymatically released ratcalvaria cell populations. Calcif. Tissue Int. 38, 143-154.

Bowditch, R. D., Hariharan, M., Tominna, E. F., Smith, J. W., Yamada, K.M., Getzoff, E. D. and Ginsberg, M. H. (1994). Identification of a novelintegrin binding site in fibronectin. Differential utilization by beta 3integrins. J. Biol. Chem. 269, 10856-10863.

Breen, E. C., Ignotz, R. A., McCabe, L., Stein, J. L., Stein, G. S. and Lian, J.B. (1994). TGF beta alters growth and differentiation related gene expression


in proliferating osteoblasts in vitro, preventing development of the maturebone phenotype. J. Cell Physiol. 160, 323-335.

Brighton, C. T. and Albelda, S. M. (1992). Identification of integrin cell-substratum adhesion receptors on cultured rat bone cells. J. Orthop. Res. 10,766-773.

Clover, J., Dodds, R. A. and Gowen, M. (1992). Integrin subunit expressionby human osteoblasts and osteoclasts in situ and in culture. J. Cell Sci. 103,267-271.

Dalton, B. A., McFarland, C. D., Underwood, P. A. and Steele, J. G. (1995).Role of the heparin-binding domain of fibronectin in attachment andspreading of human bone-derived cells. J. Cell Sci. 108, 2083-2092.

Damsky, C. H. and Werb, Z. (1992). Signal transduction by integrin receptorsfor extracellular matrix: cooperative processing of extracellular information.Curr. Opin. Cell Biol. 4, 772-781.

Dedhar, S., Argraves, W. S., Suzuki, S., Ruoslahti, E. and Pierschbacher,M. D. (1987). Human osteosarcoma cells resistant to detachment by an Arg-Gly-Asp-containing peptide overproduce the fibronectin receptor. J. CellBiol. 105, 1175-1182.

Dedhar, S. (1989). Signal transduction via the beta 1 integrins is a requiredintermediate in interleukin-1 beta induction of alkaline phosphatase activityin human osteosarcoma cells. Exp. Cell Res. 183, 207-214.

Dedhar, S., Mitchell, M. D. and Pierschbacher, M. D. (1989). The osteoblast-like differentiated phenotype of a variant of MG-63 osteosarcoma cell linecorrelated with altered adhesive properties. Connect. Tissue Res. 20, 49-61.

Eielson, C., Kaplan, D., Mitnick, M. A., Paliwal, I. and Insogna, K. (1994).Estrogen modulates parathyroid hormone-induced fibronectin production inhuman and rat osteoblast-like cells. Endocrinology 135, 1639-1644.

Erlebacher, A., Filvaroff, E. H., Gitleman, S. E. and Derynck, R. (1995).Toward a molecular understanding of skeletal development. Cell 80, 371-378.

Franceschi, R. T., Linson, C. J., Peter, T. C. and Romano, P. R. (1987).Regulation of cellular adhesion and fibronectin synthesis by 1 alpha,25-dihydroxyvitamin D3. J. Biol. Chem. 262, 4165-4171.

Genovese, C., Rowe, D. and Kream, B. (1984). Construction of DNAsequences complementary to rat alpha 1 and alpha 2 collagen mRNA andtheir use in studying the regulation of type I collagen synthesis by 1,25-dihydroxyvitamin D. Biochemistry 23, 6210-6216.

Globus, R. K., Plouet, J. and Gospodarowicz, D. (1989). Cultured bovinebone cells synthesize basic fibroblast growth factor and store it in theirextracellular matrix. Endocrinology 124, 1539-1547.

Gronowicz, G. A., DeRome, M. E. and McCarthy, M. B. (1991).Glucocorticoids inhibit fibronectin synthesis and messenger ribonucleic acidlevels in cultured fetal rat parietal bones. Endocrinology 128, 1107-1114.

Gronowicz, G. A. and DeRome, M. E. (1994). Synthetic peptide containingArg-Gly-Asp inhibits bone formation and resorption in a mineralizing organculture system of fetal rat parietal bones. J. Bone Miner. Res. 9, 193-201.

Gronowicz, G. A. and McCarthy, M. B. (1995). Glucocorticoids inhibit theattachment of osteoblasts to bone extracellular matrix proteins and decreasebeta 1-integrin levels. Endocrinology 136, 598-608.

Grzesik, W. J. and Robey, P. G. (1994). Bone matrix RGD glycoproteins:immunolocalization and interaction with human primary osteoblastic bonecells in vitro. J. Bone Miner. Res. 9, 487-496.

Hultenby, K., Reinholt, F. P. and Heinegard, D. (1993). Distribution ofintegrin subunits on rat metaphyseal osteoclasts and osteoblasts. Eur. J. Cell.Biol. 62, 86-93.

Homandberg, G. A., Meyers, R. and Williams, J. M. (1993). Intraarticularinjection of fibronectin fragments causes severe depletion of cartilageproteoglycans in vivo. J. Rheumatol. 20, 1378-1382.

Hughes, D. E., Salter, D. M., Dedhar, S. and Simpson, R. (1993). Integrinexpression in human bone. J. Bone Miner. Res. 8, 527-533.

Huhtala, P., Humphries, M. J., McCarthy, J. B., Tremble, P. M., Werb, Z.and Damsky, C. H. (1995). Cooperative signaling by alpha 5 beta 1 andalpha 4 beta 1 integrins regulates metalloproteinase gene expression infibroblasts adhering to fibronectin. J. Cell Biol. 129, 867-879.

Hynes, R. O. (1990). Fibronectins. Springer-Verlag, New York. 546 pp.Hynes, R. O. (1992). Integrins: versatility, modulation, and signaling in cell

adhesion. Cell 69, 11-25.Iida, J., Skubitz, A. P., Furcht, L. T., Wayner, E. A. and McCarthy, J. B.

(1992). Coordinate role for cell surface chondroitin sulfate proteoglycan andalpha 4 beta 1 integrin in mediating melanoma cell adhesion to fibronectin. J.Cell Biol. 118, 431-444.

Jalkanen, S. and Jalkanen, M. (1992). Lymphocyte CD44 binds the COOH-terminal heparin-binding domain of fibronectin. J. Cell Biol. 116, 817-825.

Juliano, R. L. and Haskill, S. (1993). Signal transduction from theextracellular matrix. J. Cell Biol. 120, 577-585.

Komoriya, A., Green, L. J., Mervic, M., Yamada, S. S., Yamada, K. M. andHumphries, M. J. (1991). The minimal essential sequence for a major celltype-specific adhesion site (CS1) within the alternatively spliced type IIIconnecting segment domain of fibronectin is leucine-aspartic acid-valine. J.Biol Chem. 266, 15075-15079.

Mackie, E. J. and Tucker, R. P. (1992). Tenascin in bone morphogenesis:expression by osteoblasts and cell type-specific expression of splice variants.J. Cell Sci. 103, 765-771.

Macklis, J. D., Sidman, R. L. and Shine, H. D. (1985). Cross-linked collagensurface for cell culture that is stable, uniform, and optically superior toconventional surfaces. In Vitro Cell. Dev. Biol. 21, 189-194.

Mosher, D. F., Fogerty, F. J., Chernousov, M. A. and Barry, E. L. (1991).Assembly of fibronectin into extracellular matrix. Ann. NY Acad. Sci. 614,167-180.

Mould, A. P. and Humphries, M. J. (1991). Identification of a novelrecognition sequence for the integrin alpha 4 beta 1 in the COOH-terminalheparin-binding domain of fibronectin. EMBO J. 10, 4089-4095.

Noda, M., Yoon, K., Thiede, M., Buenaga, R., Weiss, M., Henthorn, P.,Harris, H. and Rodan, G. A. (1987). cDNA cloning of alkaline phosphatasefrom rat osteosarcoma (ROS 17/2.8) cells. J. Bone Miner. Res. 2, 161-164.

Nordahl, J., Mengarelli-Widholm, S., Hultenby, K. and Reinholt, F. P.(1995). Ultrastructural immunolocalization of fibronectin in epiphyseal andmetaphyseal bone of young rats. Calcif. Tissue. Int. 57, 442-449.

Pan, L. C. and Price, P. A. (1985). The propeptide of rat bone gamma-carboxyglutamic acid protein shares homology with other vitamin K-dependent protein precursors. Proc. Nat. Acad. Sci. USA 82, 6109-6113.

Peters, J. H., Trevithick, J. E., Johnson, P. and Hynes, R. O. (1995).Expression of the alternatively spliced EIIIB segment of fibronectin. CellAdhes. Commun. 3, 67-89.

Pytela, R., Pierschbacher, M. D. and Ruoslahti, E. (1985). A 125/115-kDacell surface receptor specific for vitronectin interacts with the arginine-glycine-aspartic acid adhesion sequence derived from fibronectin. Proc. Nat.Acad. Sci. USA 82, 5766-5770.

Ridall, A. L., Daane, E. L., Dickinson, D. P. and Butler, W. T. (1995).Characterization of the rat osteopontin gene. Evidence for two vitamin Dresponse elements. Ann. NY Acad. Sci. 760, 59-66.

Saito, T., Albelda, S. M. and Brighton, C. T. (1994). Identification of integrinreceptors on cultured human bone cells. J. Orthop. Res. 12, 384-394.

Shalhoub, V., Conlon, D., Tassinari, M., Quinn, C., Partridge, N., Stein, G.S. and Lian, J. B. (1992). Glucocorticoids promote development of theosteoblast phenotype by selectively modulating expression of cell growthand differentiation associated genes. J. Cell. Biochem. 50, 425-440.

Stein, G. S., Lian, J. B. and Owen, T. A. (1990). Relationship of cell growth tothe regulation of tissue-specific gene expression during osteoblastdifferentiation. FASEB J. 4, 3111-3123.

Streuli, C. H., Bailey, N. and Bissell, M. J. (1991). Control of mammaryepithelial differentiation: basement membrane induces tissue-specific geneexpression in the absence of cell-cell interaction and morphological polarity.J. Cell Biol. 115, 1383-1395.

Weiss, R. E. and Reddi, A. H. (1980). Synthesis and localization of fibronectinduring collagenous matrix-mesenchymal cell interaction and differentiationof cartilage and bone in vivo. Proc. Nat. Acad. Sci. USA 77, 2074-2078.

Werb, Z., Tremble, P. M., Behrendtsen, O., Crowley, E. and Damsky, C. H.(1989). Signal transduction through the fibronectin receptor inducescollagenase and stromelysin gene expression. J. Cell Biol. 109, 877-889.

Winnard, R. G., Gerstenfeld, L. C., Toma, C. D. and Franceschi R. T. (1995).Fibronectin gene expression, synthesis and accumulation during in vitrodifferentiation of chicken osteoblasts. J. Bone Miner. Res. 10, 1969-1977.

Woods, A., McCarthy, J. B., Furcht, L. T. and Couchman, J. R. (1993). Asynthetic peptide from the COOH-terminal heparin-binding domain offibronectin promotes focal adhesion formation. Mol. Biol. Cell. 4, 605-613.

Young, S. L., Chang, L. Y. and Erickson, H. P. (1994). Tenascin-C in ratlung: distribution, ontogeny and role in branching morphogenesis. Dev. Biol.161, 615-625.

(Received 24 October 1995 - Accepted 27 February 1996)

Documents

Fibronectin regulates calvarial osteoblast differentiation