Root Cementum Modulates PeriodontalRegeneration in Class III FurcationDefects Treated by the Guided TissueRegeneration Technique: A HistometricStudy in DogsPatricia F. Goncxalves,* Bruno C.V. Gurgel,* Suzana P. Pimentel,* Enilson A. Sallum,*Antonio W. Sallum,* Marcio Z. Casati,* and Francisco H. Nociti Jr.*

Background: Because the possibility of root cementumpreservation as an alternative approach for the treatment ofperiodontal disease has been demonstrated, this study aimedto histometrically evaluate the effect of root cementum on peri-odontal regeneration.

Methods: Bilateral Class III furcation defects were created indogs, and each dog was randomly assigned to receive one ofthe following treatments: control (group A): scaling and rootplaning with the removal of root cementum; or test (groupB): removal of soft microbial deposits by polishing the rootsurface with rubber cups and polishing paste, aiming at max-imum cementum preservation. Guided tissue regeneration(GTR) was applied to both groups.

Results: Four months after treatment, a superior length ofnew cementum (3.59 – 1.67 mm versus 6.20 – 2.26 mm; P =0.008) and new bone (1.86 – 1.76 mm versus 4.62 – 3.01 mm;P = 0.002) and less soft tissue along the root surface (2.77 –0.79 mm versus 1.10 – 1.48 mm; P = 0.020) was observedfor group B. Additionally, group B presented a larger areaof new bone (P = 0.004) and a smaller area of soft tissue(P = 0.008).

Conclusion: Within the limits of this study, root cementummay modulate the healing pattern obtained by guided tissueregeneration in Class III furcation defects. J Periodontol2006;77:976-982.

KEY WORDS

Cementum; dogs; regeneration.

During the progression of peri-odontal disease, root cementumbecomes exposed to the subgin-

gival and/or oral environment as attach-ment loss occurs and progresses. Anumber of changes affecting the ex-posed root cementum have been de-scribed,1-4 and these have formed thebasis for mechanical treatment of theroot surface. This treatment should re-move bacteria and calculus from thesubgingival root surface and attemptto remove the contaminated part of theroot cementum. Because there are nomeans by which the clinicians can de-termine whether all contaminated rootcementum has been removed, extensiveand aggressive scaling and root planinghas been advocated.2,3,5,6

However, a gentler approach has beenproposed based on the observation thatendotoxin does not deeply penetrate intothe root cementum.6,7 It has been dem-onstrated that periodontal health can beachieved by polishing the root surface,8,9

meaning that the removal of the root ce-mentum for the purpose of eliminatingendotoxins may not be necessary. In fact,there is accumulating evidence that ce-mentum may be critical for appropriatematuration of the periodontium during

* Division of Periodontics, School of Dentistry at Piracicaba, State University of Campinas,Sao Paulo, Brazil.

doi: 10.1902/jop.2006.050243

Volume 77 • Number 6

976

formation and regeneration.10 In dogs, it has been ob-served that periodontal regeneration (with architec-ture and function completely renewed) occurs onlyin parts of the root where some original cementum re-mains after scaling and root planing.11 In light of theemerging evidence, the present study aimed to histo-metrically evaluate the impact of the root cementumon periodontal regeneration in Class III furcation de-fects treated by guided tissue regeneration (GTR).

MATERIALS AND METHODS

AnimalsNine adult male mongrel dogs (15.5 – 3.1 kg) were in-cluded in this blinded split-mouth study. All surgicalprocedures were performed under general anesthesiawith intravenous injection of sodium pentobarbital(30 mg/kg). The Institutional Committee on AnimalResearch, State University of Campinas, approved thestudy protocol.

Teeth Extraction and Periodontitis InductionInitially, the second and fourth mandibular premolars(P2 and P4) were bilaterally extracted to provide a largeridge for future flap management as recommendedby Lindhe et al.,12 and both third premolars (P3)were assigned as experimental teeth. After 2 months,a cotton ligature was placed around P3 to induce peri-odontitis (Class III furcation defects). At the time thatclinical examinations and radiographs showed a boneloss of half of the root length and that ‘‘through andthrough’’ furcation defects were produced (4 to 5months), the ligatures were removed, and a 2-weekperiod of supragingival plaque control was initiated(daily toothbrushing and topical application of 0.12%chlorhexidine).

Treatment ProcedureFollowing the plaque control phase, the furcationdefects were surgically approached. Incisions weremade, buccal and lingual mucoperiosteal flaps raised,and a notch was placed at the alveolar bone crest levelon the mesial and distal roots. The defects were ran-domly assigned to one of the following treatments:control (group A): root surface was scaled and planedwith curets and diamond-coated flame-formed bursand polished with the use of rubber cups and a polish-ing paste to remove all plaque and calculus, aiming attotal removal of cementum previously exposed to thebiofilm; or test (group B): the roots were not scaled butpolished as described above so that only soft bacterialdeposits, but not the root cementum, were removed.All the defects were treated by the GTR technique us-ing bioabsorbable membranes.† Two membranes wereprepared with identical configuration and adaptedto cover the buccal and lingual aspects of thedefect, both extending to the surrounding bone (2 to3 mm) and sutured around the tooth with bioabsorbable

sutures. The flaps were coronally positioned and ex-panded polytetrafluoroethylene (ePTFE) interruptedsutures were applied.‡ Figure 1 illustrates the surgicalprocedure. After the surgery, the animals were givenan intramuscular injection of penicillin (1:50000 IU),and the same dose was repeated after 5 days. The su-tures were removed 14 days after the surgery, and thepostoperative plaque control was performed by dailytopical application of 0.12% chlorexidine solutionduring the experimental period.

HistologyFour months after treatment, the animals were sacri-ficed, jaws were dissected, and the blocks containingteeth and surrounding bone were fixed in a 4% neu-tral buffered formalin solution for 1 week. Specimenswere decalcified in a solution of equal parts of 50% for-mic acid and 20% sodium citrate for 4 months. The de-calcified specimens were washed in running water,dehydrated, and embedded in paraffin. Serial me-sio-distal sections (7 mm thick) were obtained andstained with hematoxylin and eosin (H&E).

Histometric AnalysisTen sections per site representing the middle portionof the defect (30 mm apart) were used to obtain themean for each parameter (Fig. 2) in each dog andmeasured with an image analysis system.§

Linear MeasurementsThe following linear measurements (mm) were ob-tained in the mesial and distal roots under the furca-tion: 1) total defect length (TDL): total length of theroot surface between notches on the mesial and distalroots; 2) tissue-free defect length (TFL): portion of theTFL with the absence of any new tissue formation; 3)new cementum (NC): linear extension of the root sur-face covered by new cementum; 4) periodontal re-generation (R): linear extension of the root surfacecovered by NC adjacent to newly formed bone withfunctionally oriented collagen fibers; and 5) epithe-lium/connective tissue extension (ECT): extensionof the root surface covered by epithelium and connec-tive tissue.

Area MeasurementsA line connecting both notches defined the apical limitof the defect and the following parameters (mm2)were obtained by the point counting technique: 1)total defect area (TDA): area limited by the apical lineand the root surface in the furcation region; 2) non-filled area (NFA): portion of the TDA not filled withany tissue; 3) filled area (FA): portion of the TDA filledwith epithelium, connective tissue, and new bone; 4)soft tissue area (STA): portion of the FA filled with

† RESOLUT XT, Gore-Tex, W.L. Gore & Associates, Flagstaff, AZ.‡ Gore-Tex sutures, W.L. Gore & Associates.§ Image Pro Plus 3.0, Media Cybernetics, Silver Spring, MD.

J Periodontol • June 2006 Goncxalves, Gurgel, Pimentel, et al.

977

epithelium and connective tissue; and 5) new bonearea (NBA): portion of the FA filled with new bone.

Statistical AnalysisThe data were averaged, and the hypothesis that therewas no difference between the groups regarding theevaluated parameters was tested by an intergroupanalysis (e.g., control versus test) using the non-parametric Wilcoxon test (a = 0.05).

RESULTS

Clinical ObservationsDuring the healing period, no suppuration or abscessformation was observed. Gingival recession (2 to 3mm) with exposure of the coronal part of the mem-brane was observed in some sites (three and two sitesin the control and test groups, respectively). Thesecases were controlled with topical application of1% chlorhexidine gel during the whole experimentalperiod, and they showed no further complications.At sacrifice, soft tissues had totally healed and ex-hibited no clinical signs of inflammation (Fig. 1F).

Histologic ObservationsNo evidence of acute inflammation was found. Preex-isting cementum was present along the root surface ofthe furcation defect in test sites and completely absentin control sites. Different stages of bone regrowth andperiodontal ligament organization were observed inthe healed furcation sites. Bone formation was fre-quently incomplete, and large bone marrow spaceswere often present. The presence of functionally ori-ented collagen fibers appeared to be closely relatedto the presence or absence of newly formed alveolarbone adjacent to new cementum. A consistent con-nection seemed to exist between the initially depos-ited new cementum and the old cementum, with thenew cementum being of a cellular type. Approxi-mately 55% (five sites) and 22% (two sites) of the de-fects in the test and control groups, respectively,presented a complete filling of the furcation, and inthese cases, cementum with inserting collagen fibershad formed along the entire root surface of the defect.

Histometric AnalysisData analysis showed no significant differences be-tween the groups regarding the initial defect area(11.81 – 3.81 mm and 11.28 – 3.97 mm, for groupsA and B, respectively; P = 0.43) and extension(7.36 – 1.47 mm and 7.77 – 1.31 mm for groups Aand B, respectively; P = 0.50). In addition, intergroupanalysis demonstrated that group B (cementum pre-servation) presented a superior length of new cemen-tum (3.59 – 1.67 mm and 6.20 – 2.26 mm forgroups A and B, respectively; P = 0.008) and newbone (1.86 – 1.76 mm and 4.62 – 3.01 mm, respec-tively; P = 0.002), and a lower formation of connective

Figure 2.Schematic drawing illustrating the histometric parameters evaluated.A) Mesio-distal view of the experimental tooth showing initial defectarea (1) and initial defect length (2). B) Magnification of thefurcation zone showing the following parameters: non-filled area (3),soft tissue area (4), new bone area (5), new cementum (6) extension,and periodontal regeneration (7) extension. The filled area wasmeasured by subtracting the non-filled area of the initial defect area.Note that the line connecting the notches (N) defines the apical limitof the defect.

Figure 1.Treatment procedure. A) Presurgical clinical aspect of P3. B) Aspectof the Class III furcation defect after the removal of the granulationtissue. C) Vestibular aspect of the membrane positioned andsutured around the P3, covering the defect after treatment A (control)or B (test). D) Occlusal view of P3, showing the membranes coveringvestibular and lingual aspects of the defect. E) Flaps coronallypositioned and sutured. F) Clinical aspect of P3 4 months later, atthe day of sacrifice.

Root Cementum and Periodontal Regeneration Volume 77 • Number 6

978

tissue/epithelium along the root surface (2.77 – 0.79mm and 1.10 – 1.48 mm, respectively; P = 0.020). Dataanalysis additionally demonstrated that the extensionof periodontal regeneration, e.g., new cementumsurface adjacent to new bone with inserting collagenfibers, was greater for group B (Table 1). The propor-tional distribution of the different healing patternsalong the root surface is presented in Figure 3. Thegroup in which root cementum was present alongthe root surface showed a higher percentage of peri-odontal regeneration and lower proportion of softtissue adaptation (P <0.05). By comparing the two ex-perimental groups with respect to the area of the de-fect, data analysis showed a larger area of new boneand a smaller area of connective tissue/epitheliumfor group B compared to group A (Table 2). Finally,a higher percentage of defect filling was found forgroup B (80.95 – 17.46 mm and 94.50 – 10.27 mmfor groups A and B, respectively; P <0.05), with ahigher proportion of new bone and a lower proportionof soft tissue and non-filled area (P <0.05) (Fig. 4).Figure 5 illustrates the histometric results.

DISCUSSION

Cementum was first described in 1835,13 and re-search has demonstrated that it is a unique tissue his-tologically but shares many properties with othermineralizedtissues,particularlybone.14Despitemanyyears of research, very little is known about the impor-tance of cementum in the reparative process followingperiodontal disease. The results of the present studysupport previous reports8,9 showing that cementumremoval may not be required for the resolution ofthe inflammation but that the elimination of soft bac-terial deposits, rather than the removal of the cemen-tum, is essential for accomplishing periodontal healthfollowing therapy. Additionally, the present study in-dicates for the first time, to our knowledge, that rootcementum may modulate periodontal regenerationin vivo.

A number of changes affecting the contaminated(‘‘diseased’’) root cementum have been described,such as areas of hypermineralization and demineral-ization, progressive loss of proteins contained in theroot cementum, as well as loss of collagen matrix,1

adsorption of endotoxins and other mediators ofinflammation,2 formation of localized cell-mediatedresorption lacunae, and invasion of bacteria in theroot cementum and radicular dentin.4 All thesechanges formed the basis for cementum removal dur-ing mechanical therapy. However, the increasinglyquestioned significance of root cementum con-tamination6,15 and the poor predictability of total ce-mentum removal16 support the gentler treatmentapproach tested in this study, which leaves in placemost of the cementum but, at the same time, removesand disturbs the bacterial biofilm attached to the rootsurface.

At this point, there is no evidence to directly explainthe mechanisms by which the root cementum favorsperiodontal regeneration. However, the preservationof root cementum as a goal in periodontal therapymay be an important factor to avoid root structure lossand dentin hypersensitivity in maintenance patients17

Table 1.

Extension (mean – SD, mm) and Range of Linear Parameters in Control and Test Groups

Control (N = 9) (range) Test (N = 9) (range) P Value

Initial defect length 7.36 – 1.47 (7.32-9.64) 7.77 – 1.31 (7.35-9.69) 0.500

Tissue-free defect length 2.67 – 1.25 (0.99-4.39) 1.58 – 1.98 (0.00-4.46) 0.023*

New cementum 3.59 – 1.67 (0.91-5.91) 6.20 – 2.26 (3.25-9.69) 0.008*

Periodontal regeneration 1.91 – 1.80 (0.00-5.21) 5.08 – 3.36 (0.46-9.69) 0.004*

Epithelium/connective tissue extension 2.77 – 0.79 (1.36-3.62) 1.10 – 1.48 (0.00-3.47) 0.020*

* Statistically significant (P <0.05; Wilcoxon test).

Figure 3.The proportion (%) of the different healing patterns along the rootsurface of the furcation defects in the control and test groups.

J Periodontol • June 2006 Goncxalves, Gurgel, Pimentel, et al.

979

and to prevent root resorption18 and may also favorperiodontal regeneration, providing an appropriatemicroenvironment for cementum formation.11,19 Ce-mentum is the site where soft-tissue attachment has tobe reestablished, and surface cementum seems topresent a high chemotatic activity compared to dentinor deep cementum.20 Moreover, the cementum ma-trix is a rich source of growth factors, which influencethe activities of various periodontal cell types.19

Additionally, recent studies have shown that dentin-derived bioactive factors can modulate the inflamma-tory response.21 Dentin extracts trigger leukocytemigration in vivo in a time and dose-dependent man-ner22 via the synthesis of interleukin-1 beta (IL-1b),tumor necrosis factor-alpha (TNF-a), and chemo-kines23 and the in vitro expression of nitric oxide,TNF-a, and IL-1b by macrophages22 and osteo-blasts.24 Thus, dentin-derived factors seem to presentproinflammatory properties that may alter the peri-odontal tissues, and the presence of root cementumfollowing periodontal therapy may be a relevant issue

Figure 4.The proportion (%) of soft tissue (epithelium and connective tissue),new bone, and non-filled area in the initial defect area.

Figure 5.Mesio-distal sections of the furcation defect. A) Test group. The totaldefect area is filled in its majority with new bone (NB), whichdeveloped in continuity with the preexistent bone, and the wholeextension of the defect (extension of the root surface between thetwo notches [N]) is covered by the preexisting root cementum (*).B) Higher magnification of A showing periodontal regeneration, e.g.,the layer of new cementum (thick arrows), which covers thepreexistent cementum (*), with periodontal ligament (thin arrows)connecting this newly formed cementum to the new bone (NB).C) Control group. In contrast to the scenario found in A, the furcationdefect is not completely filled (non-filled area [NFA]). Furthermore,in the filled area, new bone (NB), epithelium (E), and connectivetissue (CT) can be observed at the coronal part of the defect. Newcementum formation is restricted to the apical portion of the defect,with the junctional epithelium reaching its coronal part. D) Highermagnification of C showing the connective tissue adaptationover the new cementum (arrows) adapted to the dentin surface (D).Furthermore, epithelium (E) is observed at the coronal part of thedefect. (H&E staining; original magnification: A and C, ·12.5; Band D, ·50; bar: A and C, 0.5 mm; B and D, 0.25 mm.)

Table 2.

Area (mean – SD, mm2) and Range of Evaluated Parameters in Control and Test Groups

Control (N = 9) (range) Test (N = 9) (range) P Value

Initial defect area 11.81 – 3.81 (8.57-16.54) 11.28 – 3.97 (8.67-18.00) 0.430

Non-filled area 2.25 – 1.98 (0.00-5.08) 0.63 – 1.00 (0.00-2.48) 0.008*

Filled area 9.56 – 4.70 (4.07-16.54) 10.65 – 4.19 (5.38-18.00) 0.430

Soft tissue area 6.57 – 1.06 (5.35-7.98) 4.20 – 0.74 (3.23-5.78) 0.008*

New bone area 2.98 – 0.99 (1.57-4.74) 6.45 – 0.80 (5.26-6.94) 0.004*

* Statistically significant (P <0.05; Wilcoxon test).

Root Cementum and Periodontal Regeneration Volume 77 • Number 6

980

with this regard. Finally, the availability of animal andhuman cementoblastic cells in culture25,26 has en-abled us to further investigate the significance of thesecells in modulating periodontal regeneration. In vitrostudies have shown that cementoblasts retain expres-sion of genes associated with mineralized tissues(such as bone sialoprotein and osteocalcin), promotemineralization (as measured by von Kossa staining),and respond to growth factors by eliciting changesin gene profile and mitogenesis and to osteotropichormones by evoking changes in gene profile andthe ability to induce mineral nodule formation.10,27,28

Additionally, in vivo studies have demonstrated thatcementoblasts have a marked ability to induce miner-alization in periodontal wounds when delivered viapolymer sponges.29,30 In summary, it can be specu-lated that the root cementum may act in three direc-tions, associated or not, as 1) a source of growthfactors from its matrix, 2) as a barrier, avoiding the un-desirable interaction of dentin matrix proteins with thehealing site, and 3) by cementoblast modulation ofcementum regeneration.

CONCLUSIONS

The results of the present study should not be inter-preted to signify that root planing should not be per-formed before GTR therapy. For practical reasons,removal of hard deposits such as calculus and miner-alized surface coatings must involve removal of thecementum. However, our results support the view thatthe intentional excessive removal of cementum forthe purpose of eliminating endotoxins may not be jus-tified even if a regenerative technique, such as GTR, isapplied. Nevertheless, further studies should be con-sidered to investigate the role of root cementum inperiodontal regeneration regarding the mechanismsinvolved and to elucidate the clinical predictability ofthis gentler approach on a long-term basis. The sup-portive periodontal therapy should subsequently aimat reducing the recontamination of the treated root sur-face to levels that are compatible with a healthy andstable periodontal condition. Within the limits of thepresent study, root cementum may modulate the heal-ingpatternobtainedbyGTRinClass III furcationdefects.

ACKNOWLEDGMENTS

This research was supported by the State of Sao PauloResearch Foundation (FAPESP; grants 02/09244-5and 02/09245-1). The authors thank Mariana Piove-zan Fugolin Lazarin, Division of Periodontics, Schoolof Dentistry at Piracicaba, for preparing the histologicmaterial.

REFERENCES1. Selvig KA. Ultrastructural changes in cementum and

adjacent connective tissue in periodontal disease. ActaOdontol Scand 1966;24:459-600.

2. Aleo JJ, De Renzis FA, Farber PA, Varboncoeur AP.The presence and biologic activity of cementum-bound endotoxin. J Periodontol 1974;45:672-675.

3. Aleo JJ, De Renzis FA, Farber PA. In vitro attachmentof human gingival fibroblasts to root surfaces. J Peri-odontol 1975;46:639-645.

4. Adriaens PA, Edwards CA, De Boever JA, Loesche WJ.Ultrastructural observations on bacterial invasion incementum and radicular dentin of periodontally dis-eased human teeth. J Periodontol 1988;59:493-503.

5. Adelson LJ, Hanks CT, Ramfjord SJ, Caffesse RG.Cytotoxicity of periodontally diseased root surfaces.J Periodontol 1980;51:700-704.

6. Cheetham WA, Wilson M, Kieser JB. Root surfacedebridement. An in vitro assessment. J Clin Periodon-tol 1988;15:288-292.

7. Smart GJ, Wilson M, Davies EH, Kieser JB. Theassessment of ultrasonic root surface debridement bydetermination of residual endotoxin levels. J ClinPeriodontol 1990;17:174-178.

8. Nyman S, Sarhed G, Ericsson I, Gottlow J, Karring T.Role of diseased root cementum in healing followingtreatment of periodontal disease. An experimentalstudy in the dog. J Periodontal Res 1986;21:496-503.

9. Nyman S, Westfelt E, Sarhed G, Karring T. Role ofdiseased root cementum in healing following treat-ment of periodontal disease. A clinical study. J ClinPeriodontol 1988;15:464-468.

10. Saygin NE, Giannobile WV, Somerman MJ. Molecularand cell biology of cementum. Periodontol 2000 2000;24:73-98.

11. Schupbach P, Gaberthuel T, Lutz F, Guggenheim B.Periodontal repair or regeneration: Structures of differenttypes of new attachment. J Periodontal Res 1993;28:281-293.

12. Lindhe J, Pontoriero R, Berglundh T, Araujo M. Theeffect of flap management and bioresorbable occlu-sive devices in GTR treatment of degree III furcationdefects. An experimental study in dogs. J Clin Peri-odontol 1995;22:276-283.

13. Denton GB. The discovery of cementum. J Dent Res1939;18:239-242.

14. Tenorio D, Crushley A, Hughes FJ. Immunocytochem-ical investigation of the rat cementoblast phenotype.J Periodontal Res 1993;28:411-419.

15. Blomlof L, Lindskog S, Appelgren R, Jonsson B,Weintraub A, Hammarstrom L. New attachment inmonkeys with experimental periodontitis with andwithout removal of the cementum. J Clin Periodontol1987;14:136-143.

16. O’Leary TJ, Kafrawy AH. Total cementum removal: Arealistic objective? J Periodontol 1983;54:221-226.

17. Wallace JA, Bissada NF. Pulpal and root sensitivityrated to periodontal therapy. Oral Surg Oral Med OralPathol 1990;69:743-747.

18. Lindskog S, Pierce AM, Blomlof L, Hammarstrom L.The role of the necrotic periodontal membrane in ce-mentum resorption and ankylosis. Endod Dent Trau-matol 1985;1:96-101.

19. Grzesik WJ, Narayanan AS. Cementum and periodon-tal wound healing and regeneration. Crit Rev OralBiol Med 2002;13:474-484.

20. Nishimura K, Hayashi M, Matsuda K, Shigeyama Y,Yamasaki A, Yamaoka A. The chemoattractive po-tency of periodontal ligament, cementum and dentinfor human gingival fibroblasts. J Periodontal Res1989;24:146-148.

J Periodontol • June 2006 Goncxalves, Gurgel, Pimentel, et al.

981

21. Silva TA, Rosa AL, Lara VS. Dentin matrix proteinsand soluble factors: Intrinsic regulatory signals forhealing and resorption of dental and periodontal tis-sues? Oral Dis 2004;10:63-74.

22. Lara VS, Figueiredo F, da Silva TA, Cunha FQ. Dentin-induced in vivo inflammatory response and in vitroactivation of murine macrophages. J Dent Res 2003;82:460-465.

23. Silva TA, Lara VS, Silva JS, Garlet GP, Butler WT,Cunha FQ. Dentin sialoprotein and phosphoproteininduce neutrophil recruitment: A mechanism depen-dent on IL-1b, TNF-a, and CXC chemokines. CalcifTissue Int 2004;74:532-541.

24. Silva TA, Lara VS, Rosa AL, Cunha FQ. Cytokine andchemokine response of bone cells after dentin chal-lenge in vitro. Oral Dis 2004;10:258-264.

25. D’Errico JA, Berry JE, Ouyang H, Strayhorn CL,Windle JJ, Somerman MJ. Employing a transgenicanimal model to obtain cementoblasts in vitro.J Periodontol 2000;71:63-72.

26. Saito M, Handa K, Kiyono T, et al. Immortalizationof cementoblast progenitor cells with Bmi-1 andTERT. J Bone Miner Res 2005;20:50-57.

27. Giannobile WV, Lee CS, Tomala MP, Tejeda KM, ZhuZ. Platelet-derived growth factor (PDGF) gene deliveryfor application in periodontal tissue engineering.J Periodontol 2001;72:815-823.

28. Saygin NE, Tokiyasu Y, Giannobile WV, SomermanMJ. Growth factors regulate expression of mineral as-sociated genes in cementoblasts. J Periodontol 2000;71:1591-1600.

29. Zhao M, Jin Q, Berry JE, Nociti FH Jr., Giannobile WV,Somerman MJ. Cementoblast delivery for periodontaltissue engineering. J Periodontol 2004;75:154-161.

30. Jin QM, Zhao M, Economides AN, Somerman MJ,Giannobile WV. Noggin gene delivery inhibits cement-oblast-induced mineralization. Connect Tissue Res 2004;45:50-59.

Correspondence: Dr. Francisco H. Nociti Jr., Faculty ofOdontology of Piracicaba, State University of Campinas,Avenida Limeira, 901, Areiao, Piracicaba, Sao Paulo13414-903, Brazil. Fax: 55-19-34125301; e-mail: nociti@fop.unicamp.br.

Accepted for publication December 9, 2005.

Root Cementum and Periodontal Regeneration Volume 77 • Number 6

982