Runx2/Cbfa1 is essential for osteoblast differen-tiation and bone formation. Runx2 null mice(Runx2-/-) completely lack mineralized tissue anddie soon after birth, whereas Runx2 heterozygousknock-out mice (Runx2+/-) stay alive but showmorphological defects in the skeletal system asobserved in cleidocranial dysplasia (CCD) inhumans. The aim of this study is to elucidate therole of Runx2 in adult mineralized tissue and alsoto reveal the distinct features of heterozygousdeletion of Runx2 in response to tooth movement.Therefore, we examined the cranium, tooth and theperiodontium in adult Runx2+/- using soft X-rayand micro-CT. In addition, tooth movementinduced by mechanical loading was evaluated. Inadult Runx2+/-, crown: root ratio of the first maxillarymolar was significantly lower than that of wildtype (WT). Irregularities in root morphology wasalso observed. The cranium was narrow with thinparietal bone compared to WT. Mechanical stress-induced tooth movement was similar betweenRunx2+/- and WT in terms of movement distance.However, while rotational movement between thefirst and third week was increased in WT, it was not

altered in Runx2+/- mice. These data indicate thatRunx2 plays a role in cranium and the tooth devel-opment in adulthood.

Key words: Runx2/Cbfa1, mechanical stress,tooth, calvaria, tooth movement

Introduction

Runx2/Cbfa1, a transcription factor which binds to theosteoblast-specific cis-acting elements 2 (OSE2)1,regulates osteoblast differentiation and skeletogenesis.Runx2 is expressed in bone1, dental mesenchyme2,also in periodontal ligament and cementoblasts duringroot formation3. Homozygous Runx2 null mutant miceembryos (Runx2 -/-) almost completely lack ossificationof the membranous bone of the skull and endochondralbone and die soon after birth4,5. On the other hand,Runx2 heterozygous knock-out mice (Runx2 +/-) stayalive, but show hypoplastic clavicles and cranium,which are characteristics of human cleidocranial dys-plasia(CCD)4-6. Unlike CCD patients who have dentalanomalies such as multiple supernumerary tooth with adelay in permanent tooth eruption7-9,10 in addition tobone phenotypes, Runx2 heterozygous knock-outmice do not have supernumerary tooth nor a delay intooth eruption possibly due to the fact that mice haveonly one dentition6,11.

Mechanical stress is an important regulatory factor ofbone homeostasis which determines skeletal mor-

Original Article

Micro-CT evaluation of tooth, calvaria and mechanical stress-induced tooth movement in adult Runx2/Cbfa1 heterozygous knock-out mice

Choo-ryung J. Chung1,3), Kunikazu Tsuji1), Akira Nifuji1), Toshihisa Komori2), Kunimichi Soma3) andMasaki Noda1)

1) Department of Molecular Pharmacology, Medical Research Institute, Tokyo Medical and DentalUniversity, Tokyo, Japan2) Department of Molecular Medicine, Osaka University Medical School, Osaka, Japan3) Orthodontic Science, Tokyo Medical and Dental University, Tokyo, Japan

J Med Dent Sci 2004; 51: 105–113

Corresponding Author: Masaki NodaDepartment of Molecular Pharmacology,Medical Research Institute, Tokyo Medical and Dental University2-3-10 Kanda-surugadai, Chiyoda-ku, Tokyo 101-0062, JapanPhone: 81-3-5280-8066 Fax: 81-3-5280-8066E-mail: noda.mph@mri.tmd.ac.jpReceived November 7; Accepted December 19, 2003

phology during development and throughout life1,12,13. Inthe stomatognathic system, tooth and the periodontium,composed of periodontal ligament, cementum and thealveolar bone, are subjected to various mechanicalstimuli during mastication, speech, and breathing.The craniofacial sutures also absorb and transmitinstantaneous mechanical stress upon these activities,and therefore can modify post-natal skeletal growth ofthe cranium14. Previously, Runx2 was reported to be adirect target of mechanical stress signaling inosteoblastic cells15, indicating that Runx2 may regulatethe skeleton and the periodontium not only during earlydevelopment but also throughout the aging process.

Therefore, to elucidate the direct role of Runx2 inadulthood, we evaluated the cranium, tooth and theperiodontium of adult Runx2 heterozygous knock outmice. In addition, mechanical stress-induced toothmovement16,17 was compared between Runx2 het-erozygous knock-out and wild type mice.

Materials and methods

AnimalsRunx2 heterozygous knock-out mice were pro-

duced and provided by Komori et al 4. Total of 15 adultmice (17.8±4.6-week old) with either Runx2 het-erozygous (HT) or wild type (WT) background wereused. Tail DNA was used to genotype the mice usingPCR.

Soft X-ray analysis After anesthesia using avertin (2 wt% of 2, 2, 2-tri-

bromothanol in PBS at 0.012 ml/g body weight), ani-mals were sacrificed by cervical dislocation. Calvariawas dissected and the maxilla was cut in halves usingthe palatal suture as the reference. Soft X-ray (SOFT-EX, Tokyo, Japan using X-ray films from Fuji film, Tokyo,Japan) picture of the calvaria and hemi-maxilla wastaken at the condition of 50kV, 25mA, for two seconds.Cranial width (a) and distance of the antero-posterioraxis (b) was measured as indicated in figure 2C.

Micro-CT (ÒÒCT) analysis ÒCT (NS-ELEX, Fukuoka, Japan) with spatial reso-

lution of 5.6 Òm in X 20 magnification was used for thestudy. Sagittal sections of the maxillary first molar(M1) were taken using the apex of the mesiobuccal root(MBR) and the distobuccal root (DBR) as a reference.Crown: root ratio was measured as indicated in figure1C. Crown length (a) is defined as the length from the

cusp tip to the root bifurcation. The length from thebifurcation to MBR apex is defined as b, while thelength from the bifurcation to DBR apex is defined as c.Both b and c represent the length of the correspondingroot.

Axial sections were taken at the midplane of the DBRof M1, which was also parallel to the occlusal plane.Intrasocket proportion of the periodontal ligament(PDL; m, d) and the root (r) was measured as indicatedin figure 1F. The widths were measured along the linewhich connects the midpoint of the pulp in MBR andDBR.

To evaluate the calvarial structures 2 coronal sectionswere taken as indicated in figure 2C. Plane 1 repre-sents the coronal plane at the middle of the anteriorfontanel, while plane 2 represents the section 100 Òmposterior to plane 1. By using the coronal sections, cal-varial thickness was measured at 5 points in a 100 Òminterval starting from the middle of the sagittal suturegap as in figure 2H.

Experimental tooth movement10 g NiTi coil spring (Furukawa Co., Tokyo, Japan)

was attached in between M1 and the upper incisors(UI) using dental adhesive resin (Kuraray Co., Tokyo,Japan). M1 of the right side was moved for 7 days inthe first group of animals (HT n = 4, WT n = 3), and 21days for the second group (HT n = 4, WT n = 4). M1 ofthe left side was left unmoved to be used as an internalcontrol. Animals were fed with powder chow during theexperimental period to avoid excessive grinding force.Body weight was monitored for the second group.After the experimental period, animals were anes-thetized using avertin, and sacrificed by cervical dislo-cation. Tissues were fixed in 4% paraformaldehyde in0.1 M phosphate buffer, pH 7.4 for the first group to beused for histological evaluation, and in 70% ethanol forthe second group which were to be used for bone his-tomorphometric analysis. All animal experimentswere approved by the animal welfare committee ofTokyo Medical and Dental University.

Analysis of tooth movementParameters for tooth movement were followed as

described by Chung et al (submitted elsewhere).Briefly, incisor-molar distance (I-MD) is the distancefrom the UI to the mesial surface of M1. Molar distance(MD) is the distance between M1 and the third molar(M3). Root angle (RA) is the angle representing theMBR inclination of M1 relative to the occlusal planedecided by the unmoved second and third molars. The

C. J. CHUNG et al. J Med Dent Sci106

parameters from the unmoved left side of each animalwere used as an internal control.

Statistical evaluationThe results were presented as mean values ±

SEM. Statistical analysis was performed based onMann-Whitney’s U-test. A p value less than 0.05 wasconsidered to be significant.

Results

Crown: root ratio of the first molar is significantlylower in Runx2 heterozygous knock-out mice

Based on soft x-ray and ÒCT examinations, thetotal number of roots of each molar, crown morphology,and the vertical alveolar bone level of the adult Runx2heterozygous knock-out mice (HT) were comparable tothose of the wild type (WT). However, when crown: rootratio of M1 was evaluated using ÒCT sagittal sections(Figs.1A, B), the root length relative to the crownlength (b/a was significantly lower in HT (p＜0.05) (Fig.1G). We also observed morphological defects of theroot in HT by ÒCT axial views of M1. Especially, the dis-tobuccal root (DBR) showed distinguishing features.While the DBR of WT was circular in morphology withsmooth surface (Fig. 1D), the DBR of the HT was irreg-ular in shape with pit-like surfaces (Fig. 1E, arrowhead)in 7 out of 8 teeth evaluated. Within the alveolar sock-et of the DBR, the proportion of the periodontal liga-ment (PDL) and the root was similar in both HT and WT(Fig.1H).

Narrow cranium with thin and reduced parietalbone thickness in adult Runx2 heterozygousknock-out mice

Sagittal and frontal sutures were remained patent in22-week-old HT (Fig. 2B) compared to WT (Fig. 2A).While there was no difference in the length of antero-posterior axis (b), which was measured from thenasofrontal suture to the lamboidal suture, the lateralwidth (a) of the calvaria was mildly reduced in HT com-pared to WT (Fig. 2I). Parietal bone thickness aroundthe suture area was further examined using ÒCTfrontal sections. In WT, the medial periphery of the pari-etal bone had a bulb-like structure with thick bone vol-ume, which was fused to each other (Figs. 2D, F).However, in HT, the medial periphery of the parietalbone became thinner as it reached the midline (Figs.2E, G). As in Fig. 2J, the changes in calvarial thicknesswithin 500 Òm lateral to the midline (sagittal suture) in

WT and HT showed a reverse pattern.

Mechanical stess-induced mesial tooth movementoccurred in Runx2 heterozygous knock-out mice

Directed mechanical stimulus such as orthodonticforce applied to the tooth results in alveolar boneremodeling induced by bone resorption on the pressureside by osteoclasts, and bone formation on the tensionside by the osteoblastic cells16-18. Since Runx2 wasreported to be directly involved in mechanotransductionof the osteoblastic cells of the periodontal ligament invito15, we examined the mechanical stress-inducedtooth movement in WT and HT mice by applying coilspring to the right maxillary first molar. Body weightdecreased around 10% in both WT and HT (Fig. 3).After 7 days of coil attachment, incisor-molar distance(I-MD) decreased and intermolar distance (MD)increased similarly in both WT (Fig. 4C) and HT (Fig.4D), indicating comparable mesial tooth movement(Fig. 4G). The type of tooth movement (translationalmovement versus rotational movement) was deter-mined by the changes in the root angle (RA) before andafter tooth movement. Both WT and HT showedmesial tooth movement with 3-5 degree rotation afterseven days (Fig. 4G). On the ÒCT axial sections, peri-odontal ligament of mesial side was compressed(Figs. 4E, F arrow head), which indicates mesial rootmovement within the alveolar socket.

By extending the experimental period up to 21days, further increase in tooth movement wasobserved (Figs. 5C, D). Mesial tooth movement basedon I-MD and MD was similar between WT and HT.Mesial rotational movement based on RA was signifi-cantly increased in the long term (21 days) compared toshort term(7 days) movement in WT ( p＜0.05), but notin HT ( Fig. 5E).

Discussion

Runx2 functions in the commitment of theosteoblast lineage from the multipotential mesenchymalcells19, accelerates osteoblast differentiation in theearly stage4, inhibits osteoblast differentiation in the latestage20,21 and regulate bone matrix protein produc-tion1,22. Runx2 also plays an important role in toothdevelopment, in particular, epithelial morphogenesis,since Runx2 null mice have misshaped and hypoplas-tic enamel organs2. However, due to the lethal pheno-type of the Runx2 null mice, the role of Runx2 in theaging process of the tooth and the cranium has not

107TOOTH AND CRANIUM OF ADULT Runx2 HETEROZYGOUS MICE

been examined exclusively in vivo. Thus, by utilizing Runx2 heterozygous knock-out

mice, which have half Runx2 gene dosage defect, weanalyzed the micro-morphology of tooth, periodon-tium, and the calvaria in the adult period (17.8±4.6-week-old). Our data indicate that the molar of Runx2heterozygous knock-out mice have lower crown: rootratio. In mice, apical one third of the molar rootbecomes covered with additional cellular cementumafter the age of 35 days23. Since the cellular cementum

is mineralized, it appears radiopaque similar as the rootby conventional soft X-ray or ÒCT. Therefore, thedecrease in root length in aged Runx2 heterozygousknock-out mice may be due to the lack of cellularcementum deposition at the root apex during theaging process, which appears as shortening of the rootcompared to the WT. This is supported by the clinicalfindings that CCD patients also lack cellular cementumdeposition8. Although others reported Runx2 het-erozygous knock-out mice to have normal cementum

C. J. CHUNG et al. J Med Dent Sci108

Fig. 1. Molar crown: root ratio is lower in adult Runx2 heterozygous knock-out miceÒCT sagittal section of the maxillary first molar(M1) of wild type (WT, A) and Runx2 heterozygous mice (HT, B) C; Parameters used to evalu-ate crown (a), mesiobuccal root (b), and distobuccal root (c) length ÒCT axial section of the M1 of WT (D) and HT (E). MBR, PR, and DBR indi-cate mesiobuccal, palatal, and distobuccal root respectively. Arrowhead indicates the defective root morphology in the DBR of HT. F; Parametersused to evaluate the mesial (m) periodontal ligament, root (r), and distal (d) periodontal ligament width of the DBR. G; Crown: root ratio is sig-nificantly lower in adult HT (n=6) compared to WT (n=8). H; PDL width in HT is similar to WT. PDL: periodontal ligament, m+r+d indicate thetotal width of the alveolar socket. Data are expressed as means ± SEM. Statistically significant difference * P＜ 0.05

Sagittalview

Axialview

parametersHTWT

formation based on histological findings, the animalsused in this study were younger than 35 days. Thisperiod is before additional cellular deposition takesplace11. In addition to the decrease in root length, wewere able to detect morphological difference of the rootin axial sections using ÒCT. The roots of Runx2 het-erozygous knock-out mice were irregular in shapewith resorption pit-like structures. However, to under-

stand the mechanisms resulting in this phenotype, fur-ther evaluation of the periodontium using histologicalsection should be performed to determine whether theroot defects are due to overt root morphogenesis, ordue to an increase in external root resorption.

Periodontal ligament width is reported to decrease inhypofunctional condition of the tooth24. Since the peri-odontal ligament width of the Runx2 heterozygous

109TOOTH AND CRANIUM OF ADULT Runx2 HETEROZYGOUS MICE

Fig. 2. Narrow cranium with reduced parietal bone thickness in adult Runx2 heterozygous knock-out miceSoft X-ray of calvaria in WT (A) and HT (B). Anterior fontanel is patent in HT. C; Parameters measured for lateral width (a) and the length ofantero-posterior axis (b). Plane 1 indicates the frontal plane used for ÒCT sections D and E, and plane 2 for F and G. ÒCT frontal section of WT(D,F) and HT (E,G). Calvarial fusion and the bulb-like structures of the medial periphery of the parietal bone are missing in HT. Arrow indicatesthe midline. H; Parietal bone thickness was measured in a 100 Òm interval along sections from plane 1. I; Cranial width was significantly reducedin adult HT (n=4) compared to WT (n=4). J; Parietal bone thickness was significantly reduced in HT (n=4) compared to WT (n=4). Data areexpressed as means ± SEM. Statistically significant difference * P＜ 0.05, ** P＜ 0.01.

knock-out mice is similar to that of the wild type, the oralmasticating function seems to be unaffected. In additionto the dental anomalies, the lateral growth of the crani-um was reduced in Runx2 heterozygous knock-outmice, which may be due to the defects in fusion of thesagittal suture and the underdevelopment of the parietalbone.

Runx2 was shown to be a direct target of mechan-otransduction in periodontal ligament osteoblasticcells through the MAPK pathway in vitro15, and to indi-rectly regulate osteoclast differentiation throughRANKL production of osteoblasts25. These observ-ations suggest a possibility that Runx2 is a regulator of

C. J. CHUNG et al. J Med Dent Sci110

Fig. 3. Time course changes in body weight during mechanicalstress-induced tooth movement. There was no significant differencebetween WT and HT.

Fig. 4. Mechanical stress induced tooth movement occurred in both Runx2 heterozygous knock-out and wild type mice Soft X-rays of the left side in WT (A), and HT (B). Space is not detectable between the first and the second molar. Soft X-rays of the right sidein WT (C), and HT (D) after 7 days of tooth movement. Space is induced due to the mesial movement of the first molar. Arrow indicates the direc-tion of tooth movement. ÒCT axial sections of the right side first molar of WT (E), and HT (F). The mesial periodontal ligament of the distobuccalroot is compressed by mesial movement (arrow head). G; Tooth movement occurred in both HT (n=4) and WT (n=3). I-MD, distance form theincisors to the first molar; MD, distance in between the first and third molar; RA, an angle indicating the inclination of the mesiobuccal root tothe occlusal plane of the second and third molar; RT, right side (loaded); LT, left side (control). Data are expressed as means ± SEM.

Control(LT. side)

Loaded(RT. side)

Loaded(ÒCT)

WT HT

bone remodeling to mechanical stress. Orthodontictooth movement is based on the remodeling of the sur-rounding alveolar bone in response to orthodonticforce. By induction of bone-resorbing osteoclasts on thepressure side and new bone-forming osteoblasts on thetension side of the root in response to mechanical load-ing26-28,16, tooth is relocated within the alveolar bone.Therefore, we hypothesized that the half gene dosageof Runx2 may cause a delay in mechanical stress-induced tooth movement due to the insufficiency ofalveolar bone remodeling in response to mechanicalloading. However, the amount of tooth movement inRunx2 heterozygous knock-out mice was similar to thatof wild type, indicating that heterozygous deletion of the

Runx2 gene is sufficient to maintain mechanotransdu-cation during tooth movement. Recent study usingRunx2 heterozygous knock-out mice show that fullRunx2 dosage is required for osteoblastic differentiationof the bone marrow cell in vitro, and also for cancellousbone regeneration following bone marrow ablation invivo, only in aged mice (over 20-week-old)29. Since ourstudy deals with aged animals near the age of 20weeks, further investigations using dynamic bone his-tomorphometry may reveal different mechanisms inmechanical stress-induced tooth movement, whichmay have been overlooked. Even though the amount oftooth movement was similar between wild type andRunx2 heterozygous knock-out mice, the nature of

111TOOTH AND CRANIUM OF ADULT Runx2 HETEROZYGOUS MICE

Fig. 5. Increase in rotational movement did not occur in Runx2 heterozygous knock-out mice in the long termSoft X-rays of the left side in WT (A), and HT (B). Soft X-rays of the right side in WT (C), and HT (D) after 21 days of tooth movement. Largergap is induced in compared to 7 days of tooth movement (Fig. 4). Arrow indicates the direction of tooth movement. E; There was a significantincrease in the root angle (RA) in WT as the experimental period was extended to 21 days, indicating rotational movement took place. Theincrease in RA was not detected in HT. Data are expressed as means ± SEM. Statistically significant difference * P＜ 0.05

tooth movement seems to differ between the twogroups in the long term. In wild type mice, molar rota-tion was significantly increased as tooth movementperiod was extended, whereas in Runx2 heterozygousknock-out mice tooth movement was mainly transition-al. In case of mice maxillary fist molar, the center ofrotation exists in the apical one third of the palatal rootwhen force is applied to the molar crown18. The shorterroot length in Runx2 heterozygous mice may contributeto the different type of tooth movement. By redistribut-ing the center of rotation more close to the toothcrown, rotational movement could have been preventedin Runx2 heterozygous knock-out mice.

Cleidocranial dysplasia (CCD) is an autosomaldominant skeletal dysplasia characterized by abnormalclavicles, patent sutures and fontanelles, and supernu-merary tooth. These phenotypes are associated withheterozygous loss or loss of function mutations ofRunx2 30. The phenotype of Runx2 heterozygousknock-out mice is consistent with the phenotype of thehuman CCD syndrome, including abnormal clavicles,patent sutures and fontanelles4,5. However, involvementof multiple supernumerary tooth and noneruption of thepermanent dentition is distinct between the twospecies. In humans, the primary dentition and the firstmolar erupt normally, whereas the eruption of the per-manent dentition is delayed or nonerupted7. In addition,multiple supernumerary teeth and malocclusion isfound in significant number of CCD affectedpatients7,31,32. In Runx2 heterozygous knock-out mice,tooth erupts normally11, and supernumerary tooth is notpresent5. The absence of these tooth phenotypes inRunx2 heterozygous mice may be due to the fact thatmice have only one dentition, which may represent theprimary dentition in humans5,11. Because of the highincidence in malocclusion, CCD patients requireorthodontic/orthognathic treatment by late adoles-cence or young adulthood7,31. However, reports onstress-induced tooth movement in CCD patients arelacking. Although some discrepancies could existbetween the dental phenotype of CCD in human com-pared to Runx2 heterozygous knock-out mice, utiliza-tion of the Runx2 heterozygous knock-out mice can bea valuable tool in providing information on the role ofRunx2 gene in mechanical stress-induced toothmovement, and the underlying bone remodeling in vivo.Lower crown : root ratio and defects in root morphologycan also be a concerning factor during dental andorthodontic treatment. Therefore, careful observation ofCCD patients would be required before and duringtreatment based on this study.

Acknowledgments

This research was supported by the grant-in-aidreceived from the Japanese Ministry of Education(15659352, 14207056, 14034214, 14028022), Grantsfrom Japan Space Forum, NASDA, Japan Society forPromotion of Science (JSPS, Research for the FutureProgram, Genome Science), and a JSPS researchgrant.

References1. Ducy, P., Zhang, R., Geoffroy, V., et al. Osf2/Cbfa1: a tran-

scriptional activator of osteoblast differentiation. Cell1997;89:747-54.

2. D’Souza, R. N., Aberg, T., Gaikwad, J., et al. Cbfa1 isrequired for epithelial-mesenchymal interactions regulatingtooth development in mice. Development 1999;126:2911-20.

3. Bronckers, A. L., Engelse, M. A., Cavender, A., et al. Cell-spe-cific patterns of Cbfa1 mRNA and protein expression in post-natal murine dental tissues. Mech Dev 2001;101:255-8.

4. Komori, T., Yagi, H., Nomura, S., et al. Targeted disruption ofCbfa1 results in a complete lack of bone formation owing tomaturational arrest of osteoblasts. Cell 1997;89:755-64.

5. Otto, F., Thornell, A. P., Crompton, T., et al. Cbfa1, a candidategene for cleidocranial dysplasia syndrome, is essential forosteoblast differentiation and bone development. Cell1997;89:765-71.

6. Mundlos, S., Huang, L. F., Selby, P., et al. Cleidocranial dys-plasia in mice. Ann N Y Acad Sci 1996;785:301-2.

7. Becker, A., Lustmann, J., and Shteyer, A. Cleidocranial dys-plasia: Part 1—General principles of the orthodontic and sur-gical treatment modality. Am J Orthod Dentofacial Orthop1997;111:28-33.

8. Lukinmaa, P. L., Jensen, B. L., Thesleff, I., et al. Histologicalobservations of teeth and peridental tissues in cleidocranialdysplasia imply increased activity of odontogenic epitheliumand abnormal bone remodeling. J Craniofac Genet Dev Biol1995;15:212-21.

9. Kreiborg, S., Jensen, B. L., Larsen, P., et al. Anomalies ofcraniofacial skeleton and teeth in cleidocranial dysplasia. JCraniofac Genet Dev Biol 1999;19:75-9.

10. Jensen, B. L., and Kreiborg, S. Development of the dentition incleidocranial dysplasia. J Oral Pathol Med 1990;19:89-93.

11. Zou, S. J., D’Souza, R. N., Ahlberg, T., et al. Tooth eruptionand cementum formation in the Runx2/Cbfa1 heterozygousmouse. Arch Oral Biol 2003;48:673-7.

12. Mullender, M. G., and Huiskes, R. Proposal for the regulatorymechanism of Wolff’s law. J Orthop Res 1995;13:503-12.

13. Nomura, S., and Takano-Yamamoto, T. Molecular eventscaused by mechanical stress in bone. Matrix Biol2000;19:91-6.

14. Mao, J. J. Mechanobiology of craniofacial sutures. J Dent Res2002;81:810-6.

15. Ziros, P. G., Gil, A. P., Georgakopoulos, T., et al. The bone-spe-cific transcriptional regulator Cbfa1 is a target of mechanicalsignals in osteoblastic cells. J Biol Chem 2002;277:23934-41.

16. King, G. J., Keeling, S. D., and Wronski, T. J.Histomorphometric study of alveolar bone turnover in ortho-

C. J. CHUNG et al. J Med Dent Sci112

dontic tooth movement. Bone 1991;12:401-9.17. Basdra, E. K. Biological reactions to orthodontic tooth move-

ment. J Orofac Orthop 1997;58:2-15.18. Pavlin, D., Zadro, R., and Gluhak-Heinrich, J. Temporal pattern

of stimulation of osteoblast-associated genes duringmechanically-induced osteogenesis in vivo: early responses ofosteocalcin and type I collagen. Connect Tissue Res2001;42:135-48.

19. Kobayashi, H., Gao, Y., Ueta, C., et al. Multilineage differenti-ation of Cbfa1-deficient calvarial cells in vitro. BiochemBiophys Res Commun 2000;273:630-6.

20. Liu, W., Toyosawa, S., Furuichi, T., et al. Overexpression ofCbfa1 in osteoblasts inhibits osteoblast maturation and caus-es osteopenia with multiple fractures. J Cell Biol2001;155:157-66.

21. Komori, T. Requisite roles of Runx2 and Cbfb in skeletal devel-opment. J Bone Miner Metab 2003;21:193-7.

22. Kern, B., Shen, J., Starbuck, M., et al. Cbfa1 contributes to theosteoblast-specific expression of type I collagen genes. J BiolChem 2001;276:7101-7.

23. Cohn, S. A. Development of the molar teeth in the albinomouse. Am J Anat 1957; 101:295-319.

24. Kaneko, S., Ohashi, K., Soma, K., et al. Occlusal hypofunctioncauses changes of proteoglycan content in the rat periodontalligament. J Periodontal Res 2001;36:9-17.

25. Geoffroy, V., Kneissel, M., Fournier, B., et al. High bone resorp-tion in adult aging transgenic mice overexpressingcbfa1/runx2 in cells of the osteoblastic lineage. Mol Cell Biol2002;22:6222-33.

26. Rody, W. J., Jr., King, G. J., and Gu, G. Osteoclast recruitmentto sites of compression in orthodontic tooth movement. Am JOrthod Dentofacial Orthop 2001;120:477-89.

27. Zhou, D., Hughes, B., and King, G. J. Histomorphometric andbiochemical study of osteoclasts at orthodontic compressionsites in the rat during indomethacin inhibition. Arch Oral Biol1997;42:717-26.

28. Keeling, S. D., King, G. J., McCoy, E. A., et al. Serum and alve-olar bone phosphatase changes reflect bone turnover duringorthodontic tooth movement. Am J Orthod DentofacialOrthop 1993;103:320-6.

29. Tsuji, K., Komori, T., and Noda, M. Loss of a half cbfa1 genedosage sharply blocks cancellous bone regeneration afterbone marrow ablation in aged adult mice due to the reductionin osteoblast cell population in the stromal cells. J Bone MinerRes 2002;17:S151.

30. Mundlos, S., Otto, F., Mundlos, C., et al. Mutations involvingthe transcription factor CBFA1 cause cleidocranial dysplasia.Cell 1997;89:773-9.

31. Cooper, S. C., Flaitz, C. M., Johnston, D. A., et al. A naturalhistory of cleidocranial dysplasia. Am J Med Genet2001;104:1-6.

32. Becker, A., Shteyer, A., Bimstein, E., et al. Cleidocranial dys-plasia: Part 2—Treatment protocol for the orthodontic and sur-gical modality. Am J Orthod Dentofacial Orthop1997;111:173-83.

113TOOTH AND CRANIUM OF ADULT Runx2 HETEROZYGOUS MICE