Abstract

Understanding the factors that control osteoblastic behavior is centrally important in establishment of successful osseointegration. Pharmacogenetic control of the osteoblast to increase the mineral content around dental implants may offer a unique advantage to clinicians in improving osseointegration success and decreasing time before mechanical loading. This in vitro pilot study has screened for bioactive peptides derived from bone morphogenetic protein 7 (BMP-7) (also called osteogenic protein 1 [OP-1]). Thirteen overlapping peptides of BMP-7 were synthesized and covalently coupled to glass coverslip substrates using silane chemistry. The rate and relative amount of mineralization were compared by von Kossa analysis using primary rat calvarial osteoblastic cell populations during a 7- to 21-day period. In addition, bone sialoprotein (BSP) and osteocalcin (OC) gene expression was measured from osteoblastic cells grown on peptide-immobilized glass coverslips by reverse transcriptase–polymerase chain reaction. Initial results from mineralization studies suggested the BMP-7–derived peptides were able to support mineralization to varying degrees with enhanced peptide-induced mineralization from the C- and N-termini of the BMP-7 molecule. Analysis of these peptide regions indicated that these peptides comprised the finger 1 and 2 domains of OP-1, which contribute toward ligand-receptor interaction. Further analysis of gene expression from select peptide-immobilized substrates indicated that peptides from the C-terminus of BMP-7 were capable of supporting BSP and OC messenger RNA expression. These studies indicate that BMP-7 peptides covalently bound to solid substrates may provide the biological basis to immobilize peptides to titanium implants to induce osteoblastic differentiation and mineralization in a more predictable fashion.

ORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemIN VITRO MINERALIZATION STUDIES WITHSUBSTRATE-IMMOBILIZED BONEMORPHOGENETIC PROTEIN PEPTIDESRESEARCHKeith Kirkwood, DDS, PhDBrian Rheude, DDSYoung Joon Kim, DDS, PhDKyle White, MEngKay C. Dee, PhDKEY WORDSOsteoblastBMP-7 peptidesKeith Kirkwood, DDS, PhD, Brian Rheude,DDS, and Young Joon Kim, DDS, PhD, areat the Department of Periodontics andEndodontics, State University of New York atBuffalo. Address correspondence to DrKirkwood at Departments of Periodontics andEndodontics and Oral Biology and Departmentof Pharmacology and Toxicology, StateUniversity of New York at Buffalo, 250 SquireHall, Buffalo, NY 14214-3008.Young Joon Kim, DDS, PhD, is at theDepartment of Periodontics, Chonnam NationalUniversity, Kwang-Ju, Korea.Kyle White, MEng, and Kay C. Dee, PhD,are at the Department of BiomedicalEngineering, Tulane University, New Orleans,LA.these substrates with variousmolecules that attract,orim 29_205 Mp_57File # 05emUnderstanding the factors that control osteoblastic behavior is centrally important inestablishment of successful osseointegration. Pharmacogenetic control of theosteoblast to increase the mineral content around dental implants may offer a uniqueadvantage to clinicians in improving osseointegration success and decreasing timebefore mechanical loading. This in vitro pilot study has screened for bioactivepeptides derived from bone morphogenetic protein 7 (BMP-7) (also called osteogenicprotein 1 [OP-1]). Thirteen overlapping peptides of BMP-7 were synthesized andcovalently coupled to glass coverslip substrates using silane chemistry. The rate andrelative amount of mineralization were compared by von Kossa analysis usingprimary rat calvarial osteoblastic cell populations during a 7- to 21-day period. Inaddition, bone sialoprotein (BSP) and osteocalcin (OC) gene expression wasmeasured from osteoblastic cells grown on peptide-immobilized glass coverslips byreverse transcriptase-polymerase chain reaction. Initial results from mineralizationstudies suggested the BMP-7-derived peptides were able to support mineralizationto varying degrees with enhanced peptide-induced mineralization from the C- andN-termini of the BMP-7 molecule. Analysis of these peptide regions indicated thatthese peptides comprised the finger 1 and 2 domains of OP-1, which contributetoward ligand-receptor interaction. Further analysis of gene expression from selectpeptide-immobilized substrates indicated that peptides from the C-terminus ofBMP-7 were capable of supporting BSP and OC messenger RNA expression. Thesestudies indicate that BMP-7 peptides covalently bound to solid substrates mayprovide the biological basis to immobilize peptides to titanium implants to induceosteoblastic differentiation and mineralization in a more predictable fashion.modifications have been developed toencourage direct bone apposition to titaniumsurfaces. Osteoblast differenti- T o b d betitanium,strengtheneand tissuy cesatingcan Throughout the years, many surfh fas able ct redi p a in e v beha ding cl in rials, te ma lant pIhNTRODUCTIONe boundary betweeu nim- bind, and activate specific cell typesion.actoe1Journal of Oral Implantology 57TABLE 1ORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemBMP-7 PEPTIDES SUPPORT OSTEOBLAST DIFFERENTIATIONSequences of BMP-7-derived peptides*SequencePeptideNo.123456STGSKQRSQN RSKTPKNQEARSKTPKNQEA LRMANVAENSLRMANVAENS SSDQRQACKKSSDQRQACKK HELYVSFRDLHELYVSFRDL GWQDWIIAPEGWQDWIIAPE GYAAYYCEGEGYAAYYCEGE CAFPLNSYMN 7*The mature form of the human bone morphagenic protein 7 (BMP-7) osteogenic protein-1 (OP-1) sequence (NCBI accession No. P18075) was divided into 13 different peptides thatwere 15 amino acids in length and that overlapped another peptide by 10 sequences. Theresulting peptides (designated 1 through 13) were used in the present study. The BMP-7protein sequence is as follows:1 STGSKQRSQN RSKTPKNQEA LRMANVAENS SSDQRQACKK HELYVSFRDL GWQDWIIAPE61 GYAAYYCEGE CAFPLNSYMN ATNHAIVQTL VHFINPETVP KPCCAPTQLN AISVLYFDDSTABLE 2Amplification primer sets used in polymerase chain reaction*Sequence (59-39) PrimerGAPDHsense (1)GAPDHantisense (2)OCsenseOCantisenseBSPsenseBSPantisenseCACCATGGAGAAGGCCGGGGGACGGACACATTGGGGGTAGTCTGACAAACCTTCATGTCCAAATAGTGATACCGTAGATGCGAACAATCCGTGCCACTCAGGAGGGGGCTTCACTGAT*GAPDH indicates glyceraldehyde-3-phosphate dehydrogenase; OC, osteocalcin; and BSP,bone sialoprotein. GenBank accession Nos. are provided for DNA sequences used in reversetranscriptase-polymerase chain reation.ation and mineralization have beendemonstrated on titanium, glass, andhydroxyapatite-coated implant surfaces.2,3 Using various methods, includingimmunocytochemical analysis andelectron microscopy, no significant differenceshave been observed betweenimplant surface coatings and mineralizationin tissue culture.4 Thus, morerecent experimental strategies have focusedon recombinant protein use toenhance bone formation around titaniumimplants.Several growth factors have beenused to induce bone formation aroundtitanium implants. These include combinationsof insulinlike growth factorand platelet-derived growth factor in abeagle dog model,5 transforminggrowth factor b1 (TGF-b1) when appliedto titanium surfaces,6 and basicfibroblast growth factor to increase angiogenesisin the area adjacent to the58 Vol. XXIX/No. Two/2003Expectedbase pairs Accesion No.NM023964 418NM013414 198NM012881 1068implant to increase subsequent boneformation.7 However, the most potentinducers of ectopic bone formationaround titanium implants described todate are members of the bone morphogeneticprotein (BMP) family of proteingrowth factors.8,9 The osteogenic capacityof many BMPs has been proven byimplanting individual recombinantBMPs in a rat ectopic assay and usingstringent criteria. BMP 2, 4 through 7,and 9 have been shown to be bone inductivemolecules.10 Moreover, both recombinantosteogenic protein 1 (OP-1)and partially purified osteogenic proteinpreparations have been shown toenhance osseointegration around dentalimplants.11,12Several studies during the past fewyears have focused on the role of BMPsduring osteoblast differentiation. Osteoblastsexpress various phenotypicmarkers, such as alkaline phosphataseSequencePeptideNo.891011CAFPLNSYMN ATNHAIVQTLATNHAIVQTL VHFINPETVPVHFINPETVP KPCCAPTQLNKPCCAPTQLN AISVLYFDDSAISVLYFDDS SNVILKKYRNSNVILKKYRN MVVRACGCH1213orim 29_205 Mp_58File # 05emand synthesize collagenous and noncollagenousbone matrix proteins, includingosteocalcin (OC).13 Osteoblastdifferentiation is regulated by many localfactors in a paracrine and/or autocrinefashion.14 BMP-7 (or OP-1) hasbeen shown to increase alkaline phosphataseactivity, parathyroid hormoneresponsiveness, and OC production inosteoblastic model systems,2,13-15 suggestingthat BMP-7 stimulated osteoblastdifferentiation from precursorcells.Covalent chemical strategies are oftenused to modify biomaterial surfaceswith bioactive compounds. The rationaleof these strategies has beensupported by previous studies thatshowed that molecules covalently immobilizedon surfaces are retainedmore dependably than molecules thatare simply adsorbed.1,16,17 The use ofbiomimetic engineered surfaces havemainly used adhesive peptides, includingargine-glycine-aspartate or RGDpeptide sequences, which mimic integrinreceptors.18-21 These peptides havebeen used either adsorbed or covalentlyattached on different substrates, includingtitanium. Relatively fewerstudies have used BMP-derived peptidesto determine if these peptides areosteogenic. A recent study has used anexpressed BMP-2 peptide containing102 amino acids of the C-terminal endof the molecule to induce bone formation,22 and others have studied the osteogenicgrowth peptide, which is apositively charged 14-amino acidgrowth polypeptide generated as aposttranslational cleavage product usingan alternative translational initiationcodon of the histone H4 gene.23-25Therefore, the present study sought toidentify the minimal bioactive sequencesfrom BMP-7 for potential immobilizationon biomaterials and use in controlof osteoblast functions.METHODS AND MATERIALSPeptide design and synthesisThirteen overlapping peptides of 20amino acids each (Table 1) derived121 SNVILKKYRN MVVRACGCHORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro Systemfrom the mature form of humanBMP-7 molecule (amino acid residues293-431, National Center for BiotechnologyInformation accession No.P18075) were custom synthesized byand purchased from American PeptideCompany Inc (Sunnyvale, Calif).Substrate preparationPeptides were covalently immobilizedon borosilicate glass coverslips (FisherScientific, Pittsburgh, Pa) using establishedtechniques.19,20 Briefly, substrateswere treated with 2% 3-aminopropyltriethoxysilaneand 2% triethylaminein acetone (chemicals from SigmaChemical Company, St Louis, Mo) underan argon environment to yield''aminated'' substrates or substratescovered with a polymerized silane layerthat possessed functional aminegroups. Peptides were then covalentlybound to the aminated substrates duringincubation with a 25:25:1 (vol/vol/vol) solution of peptide (0.1 mM in dryN,N-dimethylformamide), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide(2.5 mg/mL21 in dry N,N-dimethylformamide),and N-ethyl morpholine (allchemicals from Sigma). Adsorbed peptideswere removed using 4 M ureaand 1 M sodium chloride. Control substratesused in the present study includedplain and aminated glass coverslips.All substrates were stored underargon or nitrogen until sterilization(via immersion in 70% ethanol) anduse in cell culture experiments.Cell culture of primary ratcalvarial cellsOsteoblast-enriched cell preparationswere obtained from Sprague-Dawley21- or 28-day fetal rat calvaria by sequentialcollagenase digestion (Type II,Invitrogen Life Technologies, Carlsbad,Calif) in bone cell buffer, pH 7.4, as describedpreviously.26 The resultant cellsfrom the third and fourth 15-minutecollagenase digestions have osteoblasticcharacter, including high alkalinephosphatase activity and the ability toform collagen and bone in vitro.26These cells were pooled and culturedin BGJb media (Life Technologies) supplementedwith 10% heat-inactivatedfetal bovine serum (Sigma), 100 U/mLof penicillin, and 100 mg/mL of streptomycinat 378C in a humidified atmosphereof 5% carbon dioxide and95% air.Mineralization assayPrimary rat calvarial cells were seededin 6-well dishes containing peptide-immobilizedor control coverslips at aninitial cell density of 2500 cells perwell. On confluency, the media waschanged to include supplements of bglycerophosphate(10 mM, Sigma) andL-ascorbate (50 mg/mL, Sigma). Exogenouslyadded human recombinantBMP-7 (50 mg/mL), received as a giftfrom D. Rueger, Stryker Biotech Corporation(Hopkinton, Mass), was usedas a positive control in these studies.Media was changed every 2 to 3 days.After 14 and 21 days of culture, thecells were fixed in 4% formalin in sodiumphosphate buffer and stained forcalcium salts using the von Kossamethod. Digital images of the fixedand stained cultures were acquired usinga Sony CCD-3 camera and a NikonSMZ-U stereomicroscope at 35 magnification.Optimas image-processingsoftware (Media Cybernetics, Carlsbad,Calif) was used to detect andquantify darkly stained (mineralized)regions of the culture surface areas.One-way analysis of variance was usedto determine whether the culture conditions(ie, type of substrate) affectedthe degree of mineralization.Total RNA isolation of primary ratcalvarial cellsPrimary rat calvarial cells were platedinto 6-well dishes containing coverslipswith immobilized peptides orcontrol coverslips. Total RNA frompeptide-immobilized and control osteoblastcell populations was isolated usingTrizol reagent (Life Technologies)14 days after confluency in mineralizingmedia. Total RNA was visualizedfor intactness by ethidium bromidestaining following gel electrophoresis.orim 29_205 Mp_59File # 05emKeith Kirkwood et alRNA was quantitated by spectroscopy(SmartSpec 3000, BioRad, Hercules,Calif).Reverse transcription-polymerasechain reaction analysisA total of 5 mg of total RNA was usedfor complementary DNA synthesis witholigo (dT)12-18 primer and SuperscriptII (RnaseH-) (Life Technologies) in reversetranscription (RT) reactions. A totalof 2 mL of the RT product was usedas a template for polymerase chain reaction(PCR) amplification of BSP, OC,and GAPDH gene products (Table 1).Standard PCR conditions were used.Semiquantitative comparison withGAPDH, an unregulated housekeepinggene, was made to assess changes ingene expression as a function of peptidetreatment on agarose gels. ThePCR results were quantitated usingBioRad's PhosphorImager system andtheir Molecular Analyst software version1.5 to assess relative differences.The RT-PCR products were assessed bysubcloning RT-PCR products intopGEM-T (Promega) and DNA sequencingto determine authenticity.RESULTSRepresentative samples of mineralizationare shown in Figure 1. Calciumdeposition was confirmed by EDXanalysis (data not shown). In Figures 2and 3, we show mineralization datafrom 14- and 21-day periods normalizedto tissue culture plastic (TCP).Variable induction on mineralizationoccurred on BMP-7-immobilized substratesat both 14-and 21-day periods.One-way analysis of variance con-firmed that after 14 days, the type ofcell culture substrate affected (P 5.056, which does not meet the commonlyused significance criteria of P ,.05 but is still appropriate to report) thedegree of mineralization exhibited byosteoblastic cells. After 21 days, this effectwas no longer evident (P . .10).The analysis of variance is used to determinewhether a general experimentalfactor (in this case, the varied surfacesof the coverslips) affected a mea-Journal of Oral Implantology 59Using RT-PCR analysis, peptides 1, 6,11, 12, and 13 were used to evaluatethe potential of these peptides to induceor support messenger RNA expressionof osteoblastic-specific genesassociated with noncollagenous bonematrix formation, namely bone sialoprotein(BSP) and OC. In Figure 4,BMP-7, used as a positive control, wasable to induce OC expression comparedwith controls of plain glass. Inaddition, all immobilized BMP-7 peptidesand controls were able to supportBSP expression, but only peptides 6and 11 were able to support OC expressioncompared with controls after14 days.DISCUSSIONORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemBMP-7 PEPTIDES SUPPORT OSTEOBLAST DIFFERENTIATIONvon Kossa staining. Photographs were taken at 53 magnification.sured result not to compare the effectsof individual treatments (ie, to comparethe effect of one peptide to another).Statistical comparison of individualtreatments can be accomplished by anumber of post hoc techniques, for example,a t test. Although the variancewithin observed responses was toogreat for t tests to confirm significantdifferences between the amount ofmineralization on substrates modifiedwith any one peptide vs another, atrend toward enhanced mineralizationwas observed on peptides for theN-terminus (1) and toward the C-terminus(11, 12, and 13). These experimentswere performed in duplicate totriplicate with 3 separate experiments.From this initial screening, 5 candidateBMP-7 peptides were used insubsequent studies of gene expression.60 Vol. XXIX/No. Two/2003FIGURE 1. Representative in vitro mineralization data obtained from 21-day cultures of primary rat calvarial cells grown on osteogenicprotein 1-immobilized peptides. Indicated peptide numbers correspond to peptides listed in Table 1. Mineralization was determined byorim 29_205 Mp_60File # 05emseveral investigators in the field of tissueengineering. The common themein engineering cell and tissue behaviorat biomedical device surfaces has beento alter the surface of the material toselectively interact with specific celltypes through molecular recognitionsignals. Thus, one of the initial steps indesigning bioengineered surfaces is themolecular component of the material.In the present study, our central objectivewas to determine if humanBMP-7-derived peptides had the abilityto support or induce biomineralizationthrough in vitro studies. BMP-7is a well-known member of the TGF-bsuperfamily capable of inducing boneformation and osteoblastic cellular differentiationand secretion of bone matrixproteins.The mature chain of human BMP-7,Design strategies for creating biomimeticmaterials have been the focus ofORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemFIGURES 2 AND 3. FIGURE 2. The 14-day mineralization experiments on osteogenic protein1 (OP-1)-immobilized peptides. Primary rat calvarial cells were grown on immobilizedpeptides for 14 days after confluency. Plain glass (PG), tissue culture plastic (TCP), aminophased glass (APG), peptides 1 through 13, and OP-1 (50 mg/mL) control were evaluated.FIGURE 3. The 21-day mineralization experiments on OP-1-immobilized peptides. Primaryrat calvarial cells were grown on immobilized peptides for 14 days after confluency. ThePG, TCP, APG, peptides 1 through 13, and OP-1 (50 mg/mL) control were evaluated.lacking the signal and propeptide sequences,was used in this study. Usingthe linear sequence, as shown in Table1, BMP-7 was divided into 13 peptides,each of 20 amino acids in length andoverlapping by 10 amino acids. Otherstudies had shown that peptides fromBMP-2 containing 102 amino acids werecapable of inducing in vivo bone formation.22 If a minimal linear sequencewas sufficient for BMP-7-induced osteogenesisactivity, then tissue engineeringstrategies involving surface modifi-cation would be readily applicable forbiomedical devices.Covalent strategies in tissue engineeringhave been used to immobilizebiomolecules on various surfaces tocontrol cellular behavior. Adhesivepeptides containing the Arg-Gly-Asp(RGD) sequence have been the mostwidely studied to date.19,28,29 The RGDsequence mediates attachment of manytypes of cells, including osteoblasts,and is found in many cell membraneand extracellular matrix proteins, includingfibronectin, type I collagen, osteopontin,and bone sialoprotein.30Other non-RGD peptides have alsobeen examined alone or in combinationorim 29_205 Mp_61File # 05emKeith Kirkwood et alwith RGD peptides.20,31-35 More recently,immobilized proteins, such asBMP-4, have been successfully immobilizedonto solid substrates to modifycellular functions, including proliferationand differentiation.36 The presentstudy has used BMP-7 (OP-1)-derivedpeptides with rat calvarial osteoblasticcells to evaluate in vitro mineralization.The primary rat calvarial cell modelis a well-established in vitro modelof osteoblast differentiation and mineralization.Osteoblasts isolated fromthe calvaria of 21-day fetal rats differentiatein tissue culture by the productionand deposition of collagen-basedmatrix and the appearance of nodulesconsisting of multiple layers of cellswithin a mineralized extracellular matrix.Using this model system, we evaluatedBMP-7 peptides once covalentlyattached to glass substrates for the abilityto induce or support osteoblasticdifferentiation as measured by noduleformation and noncollagenous bonematrix gene expression. As shown inFigures 2 and 3, at both 14- and 21-dayperiods, peptides 6, 11, 12, and 13 wereconsistently able to support mineralizationcompared with controls. In addition,OP-1 was able to induce mineralizationconsistently greater thanany peptides tested. Moreover, our resultswith mineralized nodule formationwere supported by studies thataddressed gene expression. As shownin Figure 4, following 14 days in tissueculture, osteoblasts grown on OP-1-derivedpeptides 6, 11, 12, and to a lesserextent 1 and 13 were able to supportBSP gene expression. In addition, peptides6 and 11 were able to support OCgene expression. These results are consistentwith the data obtained frommineralization studies. Thus, these resultssuggest a role for shorter BMP-7peptides to support mineralization.The present study has screenedBMP-7 (OP-1) for bioactivity relative toin vitro mineralization induction. Numerousstudies have shown OP-1 to inducebone formation. Some studieshave addressed the efficacy relative toOP-1 to induce bone formation adja-Journal of Oral Implantology 61ORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemBMP-7 PEPTIDES SUPPORT OSTEOBLAST DIFFERENTIATIONcent to dental implants. These studiesused native or recombinant OP-1.Morerecent studies have shown the in vitrosynthesized BMP-2 containing the 102amino acids of the C-terminal portionof the BMP-2 molecule was capable ofinduction of ectopic bone formation invivo.22 The present study representsthe first attempt to synthesize peptidesand explore the possibility of inductionof bone formation. As shown in Figures2 and 3, OP-1-derived peptides 1,6, 11, 12, and 13 were able to induce orsupport rat calvarial osteoblastic cellmineralization and expression of noncollagenousgenes consistent with theability to synthesize bone matrix after14 and 21 days in culture.As a member of the TGF-b superfamilyof genes, OP-1 contains 7 conservedcysteines (Cys) in the C-terminaldomain that are capable of exertingits biological effect through the interactionand association of specific type62 Vol. XXIX/No. Two/2003FIGURE 4. Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of bone sialoprotein (BSP), osteocalcin (OC), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) messenger RNA expression in primary rat calvarial cells. Cells were cultured on immobilizedOP-1-derived peptides for 14 days after confluency in mineralizing conditions. The RT-PCR-specific products for BSP, OC, andGAPDH were analyzed by agarose gel electrophoresis and digitized using Molecular Analyst software. Representative data from 2independent experiments are presented.orim 29_205 Mp_62File # 05emC-terminal end of the mature OP-1,similar to findings with the largerBMP-2 peptide.22 Moreover, both C-terminalregions of BMP-2 and BMP-7show a high degree of structural homologyin this region.38 We speculatethat combinations of OP-1 peptidesfrom the C-terminal may offer additionalbone-inducing activity. Ongoingstudies will address if these peptidescan initiate BMP-mediated signalingprocesses known to be involved inBMP-7 signaling events, such as phosphorylationof SMAD-1 and SMAD-5or MAP kinases.The preliminary evidence fromthis study may provide a novel set ofOP-1-derived peptides that are capableof inducing bone formation. However,additional studies are needed tosubstantiate these findings and toidentify the molecular mechanisms bywhich these peptides may mediatetheir events.I and II serine-threonine kinase receptors.Various members of the TGF-bfamily, including OP-1, have beenshown to bind to specific type II receptors,which then complex with type Ireceptors to initiate signal transduction.The 3-dimensional structure ofthe OP-1 was determined throughX-ray crystallography at 2.8-A resolution.37 These authors proposed that theregion of this molecule that contributesthe ligand binding domain consists ofpart of finger 1 and finger 2 (Figure 5).The most intriguing finding of this articleis that peptides 6, 11, 12, and 13map the finger 1 (peptide 6) and finger2 (peptides 11, 12, and 13) regions. Theresults of this study are consistent withthe ability of these peptides to participatein ligand-receptor interactionsand subsequent signal transductionevents. In addition, peptides from thisstudy that show promise in bone-inducingactivity are mostly from theORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemACKNOWLEDGMENTSThis study was supported by grantsfrom the American Academy of ImplantDentistry Foundation and theAcademy of Osseointegration Foundation.We acknowledge Mr ThomasMartin for his assistance with figureassembly.REFERENCES1. Puleo DA, Nanci A. Understandingand controlling the bone-implantFIGURE 5. Bioactive osteogenic protein 1 (bone morphogenetic protein 7 [BMP-7]) peptides map to regions of the BMP-7 molecule involvedin ligand-receptor interaction regions. Peptide 6 is located within the finger 1 domain and peptides 11, 12, and 13 comprise the finger 2domain. Both finger 1 and finger 2 are believed to participate in ligand-receptor interaction and signal transduction (see ''Discussion''section for details). Image was captured from the Molecular Modelling Database using the Cn3D 4.0 application program.orim 29_205 Mp_63File # 05emKeith Kirkwood et alson SW, et al. Correlative microscopicinvestigation of the interface betweentitanium alloy and the osteoblast-osteoblastmatrix using mineralizing culturesof primary fetal bovine mandibularosteoblasts. Int J Oral MaxillofacImplants. 1995;10:655-665.5. Giannobile WV. Periodontal tissueengineering by growth factors.Bone. 1996;19(1 suppl):23S-37S.6. Lind M, Deleuran B, Thestrup-Pedersen K, et al. Chemotaxis of humanosteoblasts: effects of osteotropicJournal of Oral Implantology 63interface. Biomaterials. 1999;20:2311-2321.2. Davies JE, Lowenberg B, ShigaA. The bone-titanium interface in vitro.J Biomed Mater Res. 1990;24:1289-1306.3. Cooper LF, Masuda T, WhitsonSW, Yliheikkila P, Felton DA. Formationof mineralizing osteoblast cultures onmachined, titanium oxide grit-blasted,and plasma-sprayed titanium surfaces.Int J Oral Maxillofac Implants. 1999;14:37-47.4. Yliheikkila PK, Felton DA, Whit-immobilized on silanized Co-Cr-Mo. JBiomed Mater Res. 1995;29:951-957.17. Puleo DA. Biochemical surfacemodification of Co-Cr-Mo. Biomaterials.1996;17:217-222.18. Puleo DA, Bizios R. RGDS tetrapeptidebinds to osteoblasts and inhibitsfibronectin-mediated adhesion.Bone. 1991;12:271-276.19. Dee KC, Rueger DC, AndersenTT, Bizios R. Conditions which promotemineralization at the bone-implantinterface: a model in vitro study.Biomaterials. 1996;17:209-215.20. Dee KC, Andersen TT, Bizios R.Design and function of novel osteoblast-adhesive peptides for chemicalmodification of biomaterials. J BiomedMater Res. 1998;40:371-377.21. Dee KC, Anderson TT, Bizios R.Osteoblast population migration characteristicson substrates modified withimmobilized adhesive peptides. Biomaterials.1999;20:221-227.22. Zhang B, Pu Q, Zhu BF, ChenNC, Chen SM. Bone-inducing activityof human bone morphogenetic protein-2 102 peptide. Sheng Wu Gong ChengXue Bao. 2001;17:678-682.23. Greenberg Z, Gavish H, MuhlradA, et al. Isolation of osteogenicgrowth peptide from osteoblasticMC3T3 E1 cell cultures and demonstrationof osteogenic growth peptidebinding proteins. J Cell Biochem. 1997;65:359-367.24. Bab I, Gavish H, Namdar-AttarM, et al. Isolation of mitogenically activeC-terminal truncated pentapeptideof osteogenic growth peptide from humanplasma and culture medium ofmurine osteoblastic cells. J Pept Res.1999;54:408-414.25. Fei QM, Cui DF, Chen TY, ChenZW. Synthetic osteogenic growth peptidestimulates osteoblast osteogenicactivity and enhances fracture healingin rabbits. Sheng Wu Hua Xue Yu ShengWu Wu Li Xue Bao (Shanghai). 2001;33:415-420.26. Asahina I, Sampath TK, NishimuraI, Hauschka PV. Human osteogenicprotein-1 induces both chondroblasticand osteoblastic differentiationORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro SystemBMP-7 PEPTIDES SUPPORT OSTEOBLAST DIFFERENTIATIONgrowth factors. APMIS. 1995;103:140-146.7. Wang JS. Basic fibroblast growthfactor for stimulation of bone formationin osteoinductive or conductive implants.Acta Orthop Scand Suppl. 1996;269:1-33.8. Rutherford RB, Ryan ME, KennedyJE, Tucker MM, Charette MF.Platelet-derived growth factor anddexamethasone combined with a collagenmatrix induce regeneration of theperiodontium in monkeys. J Clin Periodontol.1993;20:537-544.9. Sailer HF, Kolb E. Application ofpurified bone morphogenetic protein(BMP) in cranio-maxillo-facial surgery:BMP in compromised surgical reconstructionsusing titanium implants. JCraniomaxillofac Surg. 1994;22:2-11.10. Rosen V, Thies RS, Lyons K.Signaling pathways in skeletal formation:a role for BMP receptors. Ann NY Acad Sci. 1996;785:59-69.11. Rutherford RB, Sampath TK,Rueger DC, Taylor TD. Use of bovineosteogenic protein to promote rapidosseointegration of endosseous dentalimplants. Int J Oral Maxillofac Implants.1992;7:297-301.12. Xiang W, Baolin L, Yan J, YangX. The effect of bone morphogeneticprotein on osseointegration of titaniumimplants. J Oral Maxillofac Surg. 1993;51:647-651.13. Aubin JE, Liu F, Malaval L,Gupta AK. Osteoblast and chondroblastdifferentiation. Bone. 1995;17(2suppl):77S-83S.14. Takuwa Y, Ohse C, Wang EA,Wozney JM, Yamashita K. Bone morphogeneticprotein-2 stimulates alkalinephosphatase activity and collagensynthesis in cultured osteoblastic cells,MC3T3-E1. Biochem Biophys Res Commun.1991;174:96-101.15. Nakase T, Takaoka K, MasuharaK, et al. Interleukin-1 beta enhancesand tumor necrosis factor-alphainhibits bone morphogenetic protein-2-induced alkaline phosphatase activityin MC3T3-E1 osteoblastic cells. Bone.1997;21:17-21.16. Puleo DA. Activity of enzyme64 Vol. XXIX/No. Two/2003orim 29_205 Mp_64File # 05emof osteoprogenitor cells derived fromnewborn rat calvaria. J Cell Biol. 1993;123:921-933.27. Luben RA,Wong GL, Cohn DV.Biochemical characterization with parathormoneand calcitonin of isolatedbone cells: provisional identification ofosteoclasts and osteoblasts. Endocrinology.1976;99:526-534.28. Maheshwari G, Brown G, LauffenburgerDA, Wells A, Griffith LG.Cell adhesion and motility depend onnanoscale RGD clustering. J Cell Sci.2000;113(pt 10):1677-1686.29. Massia SP, Stark J. ImmobilizedRGD peptides on surface-grafted dextranpromote biospecific cell attachment.J Biomed Mater Res. 2001;56:390-399.30. Craig WS, Cheng S, MullenDG, Blevitt J, Pierschbacher MD. Conceptand progress in the developmentof RGD-containing peptide pharmaceuticals.Biopolymers. 1995;37:157-175.31. Massia SP, Hubbell JA. Covalentsurface immobilization of Arg-Gly-Asp- and Tyr-Ile-Gly-Ser-Arg-containingpeptides to obtain well-definedcell-adhesive substrates. Anal Biochem.1990;187:292-301.32. Massia SP, Rao SS, Hubbell JA.Covalently immobilized laminin peptideTyr-Ile-Gly-Ser-Arg (YIGSR) supportscell spreading and co-localizationof the 67-kilodalton laminin receptorwith alpha-actinin and vinculin. J BiolChem. 1993;268:8053-8059.33. Healy KE, Rezania A, Stile RA.Designing biomaterials to direct biologicalresponses. Ann N Y Acad Sci.1999;875:24-35.34. Rezania A, Healy KE. Biomimeticpeptide surfaces that regulateadhesion, spreading, cytoskeletal organization,and mineralization of thematrix deposited by osteoblast-likecells. Biotechnol Prog. 1999;15:19-32.35. Rezania A, Healy KE. Integrinsubunits responsible for adhesion ofhuman osteoblast-like cells to biomimeticpeptide surfaces. J Orthop Res.1999;17:615-623.36. Puleo DA, Kissling RA, SheuMS. A technique to immobilize bioac-ORAL IMPLANTOLOGYMonday Mar 24 2003 02:19 PMAllen Press x DTPro Systemtive proteins, including bone morphogeneticprotein-4 (BMP-4), on titaniumalloy. Biomaterials. 2002;23:2079-2087.37. Griffith DL, Keck PC, SampathTK, Rueger DC, Carlson WD. Threedimensionalstructure of recombinanthuman osteogenic protein 1: structuralparadigm for the transforming growthfactor beta superfamily. Proc Natl AcadSci U S A. 1996;93:878-883.orim 29_205 Mp_65File # 05emKeith Kirkwood et al38. Scheufler C, Sebald W, HulsmeyerM. Crystal structure of humanbone morphogenetic protein-2 at 2.7 Aresolution. J Mol Biol. 1999;287:103-115.Journal of Oral Implantology 65