Basic Research—Biology

EphB–EphrinB Interaction Controls Odontogenic/OsteogenicDifferentiation with Calcium HydroxideXiaozhe Wang, DDS, PhD,*‡ George Jong, DDS,† Louis M. Lin, BDS, DDS, PhD,†

and Emi Shimizu, DDS, PhD*†

Abstract

Introduction: Calcium hydroxide is used in direct pulpcapping of uncontaminated exposed vital pulps causedby mechanical or traumatic injury. Calcium hydroxidecreates a high alkaline pH environment and initiatesa mineralized tissue formation in the pulp. The exactmechanism by which calcium hydroxide induces thereparative dentin formation is unknown. Because Ephreceptors and ephrin ligands play a role in pulp stemcell migration and proliferation, our hypothesis isthat calcium hydroxide–related odontogenic/osteogenicdifferentiation may be associated with Eph-ephrin inter-action. The aim of this study was to investigate whetherEph-ephrin interaction regulates odontogenic/osteogenicdifferentiation with calcium hydroxide. Methods:Primary pulp cells were harvested from the molars ofC57BL/6 mice. The cells were treated with calciumhydroxide. Immunofluorescence was used to detectprotein expression. A knockout of the ephrinB1 orEphB2 gene was performed with short hairpin RNAs.Cell migration, proliferation, and gene expression werethen analyzed. Results: Calcium hydroxide stimulatedEphB2 gene expression but suppressed ephrinB1gene expression at the proliferation stage. However,calcium hydroxide stimulated both ephrinB1 andEphB2 gene expression at the differentiation stage. Inaddition, EphB2 localized at ephrinB1–positive cells atthe area of Dentin sialoprotein (DSP) staining, whichincreased with calcium hydroxide treatment. Knockdownof ephrinB1–EphB2 significantly suppressed cell prolifer-ation. Additionally, knockdown of the ephrinB1gene caused cell migration, whereas a lack of the

From the *Departments of Basic Science and CraniofacialBiology and †Endodontics, New York University College ofDentistry, New York, New York; and ‡Department of PreventiveDentistry, Peking University School and Hospital of Stomatol-ogy, Beijing, China.

Xiaozhe Wang and George Jong contributed equally to thiswork.

Supported by an American Association of EndodontistsFoundation Research Grant.

Address requests for reprints to Dr Emi Shimizu, Depart-ment of Basic Science and Craniofacial Biology and Departmentof Endodontics, New York University College of Dentistry, 345East 24th Street, New York, NY 10010-4086. E-mail address:es152@nyu.edu0099-2399/$ - see front matter

Copyright ª 2013 American Association of Endodontists.http://dx.doi.org/10.1016/j.joen.2013.06.016

1256 Wang et al.

EphB2 gene suppressed calcium hydroxide–induced mineralization from primary pulpcells. Conclusions: EphrinB1–EphB2 interaction contributes to calcium hydroxide–induced odontogenic/osteogenic differentiation. This observation is the first finding ofthe mechanism of calcium hydroxide–induced odontogenic/osteogenic differentiation.(J Endod 2013;39:1256–1260)

Key WordsCalcium hydroxide, Eph, ephrin, primary pulp cells, pulp capping

Calcium hydroxide has been extensively used for direct pulp capping in clinicaldentistry. It is known that calcium hydroxide initiates the formation of reparative

dentin at the site of pulp exposure because of its high alkaline pH. The reparativeprocess that follows involves cell migration and the proliferation of pulp mesenchymalstem cells (1, 2). However, the exact mechanism by which calcium hydroxide inducesthe reparative dentin formation remains unclear.

The Eph/ephrin family is well known for its role in mediating inhibitory or repul-sive cellular responses and has been shown to be expressed during tooth development(3). Eph receptors are the largest subgroup of the receptor tyrosine kinase family. Theircell surface ligands, known as ephrins, are involved in a variety of cell communicationsthat affect processes such as cell proliferation, chemotaxis, angiogenesis, extracellularmatrix remodeling, and cell differentiation (4–8). It is known that forward signaling ofephrinB1 inhibits the migration of dental pulp stem cells, while permitting mobilizationof dental pulp stem cells after injury through the down-regulation of ephrinB1 geneexpression (7). In addition, forward signaling of ephrinB1 increases the number ofbromodeoxyuridine-positive pulp stem cells and CD146 expression in the proliferatingcells (9). Thus, a forward signal of ephrinB1 inhibits the migration of dental pulp stemcells while stimulating the proliferation of pulp stem cells. With respect to the Ephreceptors, a reverse signal of EphB2 in pulp stem cells has been shown to stimulatedentin marker genes (9).

Our hypothesis is that calcium hydroxide–induced odontogenic/osteogenic differ-entiation may be associated with Eph-ephrin interaction. The aim of this study was toinvestigate how Eph-ephrin interaction regulates odontogenic/osteogenic differentia-tion with calcium hydroxide.

Material and MethodsAnimals and Cell Culture of Pulp Stem Cells

Primary pulp cells were harvested from 1- to 2-month old male C57BL/6 mice(a total of 6 mice), minced in phosphate-buffered saline (PBS), and incubated inPBS containing 3 mg/mL collagenase type I (Sigma-Aldrich, St Louis, MO) and4 mg/mL dispase (Sigma-Aldrich) for 30 minutes at 37�C (10, 11). The cells werecultured on plates with proliferation condition media (alpha-Minimum EssentialMedium [alpha-MEM] supplemented with 20% fetal bovine serum, 2 mmol/LL-glutamine, and 100 U/mL penicillin-streptomycin; Life Technologies, Grand Island,NY). Fifty micrograms per milliliter of L-ascorbic acid (Sigma-Aldrich) was added tothe media on day 7 and every 2–3 days afterward. For the last 4 days of the cultureperiod, 5 mmol/L beta-glycerol phosphate (EMD Millipore, Billerica, MA) was addedto the media for mineralization. All primary cells were used in their second or third

JOE — Volume 39, Number 10, October 2013

Basic Research—Biology

passage. Cells were treated with calcium hydroxide as described by Jiet al (2). The final concentration of 1 mg/mL was made by adding1 mL of 1 mg/mL calcium hydroxide stock into the 12-well plate with1 mL media. Primary pulp cells were proliferating at 7 days (prolifera-tion stage) until they became confluent at 8–9 days. After confluence,ascorbic acid was added, and cells began differentiating in the plates.At 14 days (differentiation stage), cells were accumulated to createnodule formation.

Migration AssayMouse primary pulp cells were placed in an upper chamber of

QCM Chemotaxis Cell Migration Assay (EMD Millipore). Twenty-four-well flat-bottom plates were incubated with or without calciumhydroxide for 72 hours at 37�C in 5% CO2. The interior of the transwellwas cleaned to remove cells that did not migrate through themembrane.Migrated cells were quantitated using a cell-permeable fluorescent dye(excitation maximum: �485 nm; emission maximum: �520 nm).

Proliferation AssayMouse primary pulp cells were placed in 96-well or 12-well plates

and then incubated with or without calcium hydroxide for at 37�C in 5%CO2. Cells in 12-well plates were corrected on day 7 and counted usinga hemocytometer. Cells in 96-well plates were incubated for 72 hourswith or without calcium hydroxide. tetrazolium dye (MTT) solution(Biotium Inc, Hayward, CA) was added to the 100 mL medium ineach well according to the manufacturer’s instructions. Cells were incu-bated at 37�C for 4 hours. Dimethyl sulfoxide (DMSO) was directlyadded into the medium in each well and pipetted in succession todissolve the formazan salt. The absorbance signal was measured ona spectrophotometer at 570 nm. The background absorbance wasmeasured at 630 nm. Fold changes in the normalized absorbance values(background absorbance subtracted from signal absorbance) relativeto the negative control were calculated.

Quantitative Real-time Reverse-transcriptionPolymerase Chain Reaction

Total RNA from primary pulp cells was isolated with TRIzol reagent(Sigma-Aldrich) and reverse transcribed to complementary DNA withTaqMan Reverse Transcription Reagents (Life Technologies) accordingto the manufacturer’s instructions. The sequences were amplified byadding 2.5 mL complementary DNA to the polymerase chain reaction(PCR) mixture (22.5 mL) containing each primer (0.2 mm) and12.5 mL Platinum SYBR Green qPCR SuperMix uracil-DNA glycosylase(UDG) (Life Technologies). The reactions were preincubated at 50�Cfor 2 minutes for decontamination of deoxyuridine (dU)-containingDNA by UDG and then incubated at 95�C for 2 minutes to inactivateUDG and activate Taq. The PCR program continued 49 cycles of dena-turation at 95�C for 15 seconds with the annealing and elongation of theprimers at 60�C for 30 seconds. Fold changes in gene expression rela-tive to control samples were calculated using the following formula:2(DCt control or negative control – DCt calcium hydroxide or shRNA). All of thesamples were normalized to beta-actin. The primer sequences wereas follows: alkaline phosphatase (ALP), forward: 50-tcccacgttttca-cattcgg-30, reverse: 50-cccgttaccatataggatggcc-30; osterix (Osx),forward: 50-agaggttcactcgctctgacga-30, reverse: 50-ttgctcaagtggtcgcttctg-30; bone sialoprotein (BSP), forward: 50-acaccccaagcacagacttttg-30,reverse: 50-tcctcgtcgctttccttcact-30; dentin matrix protein-1 (DMP-1),forward:50- ctcaggacagtagccgatcc-30, reverse: 50 tgggtttgttgtggtaagca-30;dentin sialophosphoprotein (DSPP), forward 50-ctgggaagagccaagatcag-30, reverse: 50-ctccactccttggtgtccat-30; and beta-actin, forward:

JOE — Volume 39, Number 10, October 2013

50-tcctcctgagcgcaagtactct-30, reverse: 50-cggactcatcgtactcctgctt-30. ThePCR reactions and program are described in a previous publication (12).

Plasmids, Short Hairpin shRNA TransfectionVectors that express ephrinB1 and EphB2 shRNAs under the

control of the U6 promoter were constructed by inserting pairs of an-nealed DNA oligonucleotides into the lentivirus-shRNA expressionvector (GeneCopeia, Rockville, MD). Supernatants containing infec-tious lentivirus were amplified by HEK293 cells. Mouse primary pulpcells were plated into 24- or 96-well plates and infected with 20 multi-plicity of infection (MOI) of viruses for 6 hours. Freshmediumwas usedafter the incubation period, and the cells were cultured for an additional3 days to examine silencing effects.

Immunofluorescence AssayMouse primary pulp cells were grown in 24-well plates with or

without calcium hydroxide and then fixed in 4% paraformaldehyde inPBS. Primary antibodies for anti–ephrinB1, anti- EphB2 (R&D SystemsInc, Minneapolis, MN), anti-Oct3/4, and anti-DSP (Santa Cruz Biotech-nology, Dallas, TX) were used. Secondary antibodies conjugated togreen-fluorescent Alexa Fluor 488 dye (Life Technologies) or red-fluorescent Texas Red dye (Santa Cruz Biotechnology) were used ata dilution of 1:300. The nuclei were stained with 40,6-diamidino-2-phe-nylindole (DAPI) (Sigma-Aldrich).

Statistical AnalysisAll results are expressed as mean � standard deviation of tripli-

cate measurements with all experiments being repeated a minimumof 3 times. Statistical analyses were performed using the Student t test.

ResultsCalcium Hydroxide Affected EphrinB1 and EphB2Interaction

To investigate the effect of calcium hydroxide on ephrinBs andEphBs, we examined the gene expression of ephrinBs and EphBswith or without calcium hydroxide treatment in primary pulp cells.Calcium hydroxide stimulated EphB2 gene expression at the prolifera-tion stage while inhibiting ephrinB1 gene expression (Fig. 1A).However, calcium hydroxide stimulated the ephrinB1 gene and theEphB2 gene at the differentiation stage (Fig. 1B). Calcium hydroxidehad no effect on ephrinB2, EphB1, EphB4, and EphB6 (Fig. 1A and B).

Next, we performed an immunofluorescence assay to examinewhere ephrinB1 and EphB2 were expressed. Calcium hydroxide causedcell aggregation and nodule formation during high levels of ephrinB1and EphB2 expression. EphB2 localized ephrinB1–positive cells withor without calcium hydroxide (Fig. 1C). To examine if ephrinB1 is asso-ciated with stem cells, we used stem cell markers such as Oct3/4, whichpartially localized to ephrinB1–positive cells (Fig. 1D). DSP expression,which increased with calcium hydroxide, was shown to localizesurrounding or overlapping ephrinB1–positive cells (Fig. 1E). Calciumhydroxide significantly stimulated cell numbers on day 7 (Fig. 1F).

Calcium Hydroxide Stimulated Cell Migrationthrough EphrinB1

To assess if ephrinB1 and EphB2 genes were associated withcalcium hydroxide–induced cell migration, we performed knockdownof ephrinB1 and EphB2 genes. As shown in Figure 2A, shRNA of eph-rinB1 and EphB2 caused a 50%–60% reduction of genes in primarypulp cells. A high concentration (1mg/mL) of calcium hydroxide signif-icantly increased cell migration in the negative control. Knockdown ofephrinB1 induced cell migration with calcium hydroxide compared

EphB–EphrinB Interaction 1257

Figure 1. (A) Total RNA was extracted from mouse primary pulp cells cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 7 days. RNA was measuredusing real-time reverse-transcription PCR. The relative levels of messenger RNA was normalized to beta-actin and then expressed as fold stimulation over control.*P < .05 compared with the control. (B) Total RNA was extracted from primary pulp cells cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 14 days withascorbic acid. RNA was measured using real-time reverse-transcription PCR. The relative levels of messenger RNA were normalized to beta-actin and then expressedas fold stimulation over the control. *P < .05 compared with the control. (C) Mouse pulp cells cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 14days with ascorbic acid. Primary antibodies against ephrinB1 (red) or EphB2 (green) were used at a dilution of 1:100. Primary antibodies were conjugated to eithergreen-fluorescent Alexa Fluor 488 dye or red-fluorescent Texas Red dye (magnification �100). DAPI was used to stain the nuclei (blue). (D) Mouse pulp cellswere cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 14 days with ascorbic acid. Primary antibodies against ephrinB1 (red) or Oct3/4 (green) wereused at a dilution of 1:100. Primary antibodies were conjugated to either green-fluorescent Alexa Fluor 488 dye or red-fluorescent Texas Red dye (magnification�100). DAPI was used to stain the nuclei (blue). (E) Mouse pulp cells were cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 14 days with ascorbicacid. Primary antibodies against ephrinB1 (green) or DSP (red) were used at a dilution of 1:100. Primary antibodies were conjugated to either green-fluorescentAlexa Fluor 488 dye or red-fluorescent Texas Red dye (magnification�100). DAPI was used to stain the nuclei (blue). (F) Primary pulp cells were cultured in 12-well plates for 7 days with or without calcium hydroxide. Cells were corrected and counted by a hemocytometer.

Basic Research—Biology

with the negative control (Fig. 2B). Interestingly, knockdown of EphB2did not affect cell migration. Knockdown of ephrinB1 and EphB2 signif-icantly suppressed cell proliferation with or without calcium hydroxide(Fig. 2C). However, calcium hydroxide did not stimulate cell prolifer-ation in the negative control (Fig. 2C).

EphrinB1– EphB2 Interaction Controlled CalciumHydroxide–induced Mineralization

To examine whether ephrinB1– EphB2 interaction controlledcalcium hydroxide–induced odontogenic/osteogenic differentiation,we performed real-time reverse-transcription PCR using bone anddentin marker genes. Calcium hydroxide increased gene expressionof Alp, Osx, Bsp, andDmp-1 (Fig. 3A). The dentin sialophosphoprotein(Dspp) gene was only detected in the negative control and during treat-ment with calcium hydroxide (0.1 mg/mL). Although a low concentra-tion of calcium hydroxide significantly enhanced Dspp gene expression(data not shown), the basal expression of Dspp was very low. Knock-down of ephrinB1 and EphB2 significantly decreased calciumhydroxide–induced gene expression of Alp, Osx and Bsp (Fig. 3A).

1258 Wang et al.

EphB2 knockdown did not prevent calcium hydroxide–induced geneexpression of Dmp-1, whereas ephrinB1 knockdown completely elim-inated expression (Fig. 3A). In our results, we speculate that ephrinB1controls odontogenic/osteogenic differentiation from pulp cells,whereas EphB2 regulates only osteogenic differentiation.

DiscussionIn this study, we investigated the effect of ephrinB1– EphB2 inter-

action on odontogenic/osteogenic differentiation with calciumhydroxide. Direct pulp capping is defined as a dental material placeddirectly on a mechanical or traumatic vital pulp exposure (13). Thetreatment of exposed vital pulp is accomplished by sealing the pulpalwound with calcium hydroxide ormineral trioxide aggregate to facilitatereparative dentin formation (13–15). In addition, Ji et al (2) showedthat calcium hydroxide induces the activation of phospho-extracel-lular-signal-regulated kinase (p-ERK) in the cultured pulp stem cells.

The Eph-ephrin signaling molecules have been shown to be ex-pressed during tooth development, indicating that Eph-ephrin interac-tion is associated with tooth development (3). Our data show that

JOE — Volume 39, Number 10, October 2013

Figure 2. (A) Mouse pulp cells were infected with the negative control (scramble shRNA), ephrinB1 shRNA, EphB2 shRNA, or both ephrinB1 and EphB2 shRNA.Infected cells were cultured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 7 days. RNA was measured using real-time reverse-transcription PCR. The relativelevels of messenger RNA were normalized to beta-actin and then expressed as fold stimulation over the control. (B) Infected primary pulp cells were cultured withor without calcium hydroxide for 24 hours. Migrated cells were quantitated using a cell-permeable fluorescent dye (485/520). *P < .05 compared with the negativecontrol (1 mg/mL calcium hydroxide). #P < .05 compared with the negative control (vehicle). &P < .05 compared with the negative control (0.1 mg/mL calciumhydroxide). (C) Infected primary pulp cells were cultured with or without calcium hydroxide for 72 hours. Proliferating cells were quantitated using the absorbancesignal (570/630). Fold changes in the normalized absorbance values relative to the negative control were calculated.

Basic Research—Biology

calcium hydroxide stimulates EphB2 gene expression but inhibits eph-rinB1 gene expression of primary pulp cells at the proliferation stage.Interestingly, calcium hydroxide stimulates ephrinB1 and EphB2 geneexpression at the differentiation stage, which contains ascorbic acid. As-corbic acid stimulates differentiation and mineralization of osteoblastsand odontoblasts (16, 17).

Forward signaling of ephrinB1 inhibits the migration of pulp cellsthrough the extracellular signal-regulated kinase/Mitogen-ActivatedProtein Kinase (ERK/MAPK) signaling pathway, whereas ephrinB1gene expression is suppressed in injured pulp tissues (7). Thus,a forward signal of ephrinB1 inhibits the migration of primary pulpcells, whereas it stimulates the proliferation of primary pulp cells(9). Our data also show that knockdown of ephrinB1 increases cellmigration, suggesting that calcium hydroxide stimulates cell migrationthrough ephrinB1 of primary pulp cells. In addition, knockdown ofephrinB1 and EphB2 significantly suppresses cell proliferation withor without calcium hydroxide, indicating that ephrinB1– EphB2 inter-action plays a role for cell proliferation in primary pulp cells.

Figure 3. (A) The total RNA was extracted from infected mouse pulp stem cells cumeasured using real-time reverse-transcription PCR. The relative levels of messengerthe control. *P < .05 compared with the control. (B) A schematic representation oprimary pulp cells.

JOE — Volume 39, Number 10, October 2013

We have shown that ephrinB1 colocalizes EphB2 with or withoutcalcium hydroxide. Interestingly, ephrinB1–expressing cells overlapwith DSP expression, indicating ephrinB1–positive cells may be similarto odontoblast-like cells. In addition, Oct3/4 expression partially local-izes ephrinB1–positive cells. This observationmeans that ephrinB1mayinteract with stem cells.

Arthur et al (9) showed that a reverse signal of EphB2 stimulatesmineral production and dentin marker genes as well dentin sialopro-tein, dentin phosphoprotein, and DMP. We showed that knockdownof ephrinB1 prevents calcium hydroxide–induced bone and dentinmarkers. Although a lack of EphB2 suppresses only bonemarker genes,especially Alp, it did not block calcium hydroxide–inducedDmp-1 geneexpression. Therefore, a reverse signal of EphB2 may control the differ-entiation of osteogenesis from primary pulp cells. Furthermore, knock-down of ephrinB significantly inhibits the gene expression of Bsp andDmp-1, indicating that ephrinB1 plays an important role in the differ-entiation of osteoblasts and odontoblasts from primary pulp cells withcalcium hydroxide. In contrast, we report that ephrinB1 in

ltured with calcium hydroxide (0.1 mg/mL or 1 mg/mL) for 21 days. RNA wasRNA were normalized to beta-actin and then expressed as fold stimulation overf ephrinB1– EphB2 interactions and the effect of calcium hydroxide on mouse

EphB–EphrinB Interaction 1259

Basic Research—Biology

preosteoclasts/monocytes suppresses osteoblast differentiation throughthe interaction of ephrinB1 and EphB1/B3 (18). Osteoblast-specificephrinB1 knockout mice show abnormal skeletal development, espe-cially lessened osteoblast differentiation through ephrinB1 and EphB2(19). Thus, it is important to know where ephrinB1 is expressed andwhich receptor activates through ephrinB1.

Calcium hydroxide activates the release of transforming growthfactor beta (20, 21). The stimulation of dentinogenesis has beenobserved after the implantation of the basic fibroblast growth factorand transforming growth factor beta, bone morphogenetic proteins,and insulin growth factor I (22–25). Such growth factors may beinvolved in ephrinB1–EphB2 interaction or recruit adoptermolecules. Based on our data, we could not directly confirm howcalcium hydroxide induced ephrinB1–EphB interaction. Furtherstudies regarding the association between ephrinB1 and EphB2 andgrowth factors are needed.

In conclusion, calcium hydroxide stimulates gene expression ofephrinB1 and EphB2. EphB2–ephrinB1 interaction controls migration,proliferation, differentiation, and mineralization from primary pulp cellswith calcium hydroxide (Fig. 3B). This observation is the first finding ofthe mechanism of calcium hydroxide–induced odontogenic/osteogenicdifferentiation through the interaction of ephrinB1 and EphB2.

AcknowledgmentsThe authors deny any conflicts of interest related to this study.

References1. Guven EP, Yalvac ME, Sahin F, et al. Effect of dental materials calcium hydroxide-

containing cement, mineral trioxide aggregate, and enamel matrix derivative onproliferation and differentiation of human tooth germ stem cells. J Endod 2011;37:650–6.

2. Ji YM, Jeon SH, Park JY, et al. Dental stem cell therapy with calcium hydroxide indental pulp capping. Tissue Eng Part A 2010;16:1823–33.

3. Luukko K, Loes S, Kvinnsland IH, et al. Expression of ephrin-A ligands and EphAreceptors in the developing mouse tooth and its supporting tissues. Cell TissueRes 2005;319:143–52.

4. Salvucci O, de la Luz Sierra M, Martina JA, et al. EphB2 and EphB4 receptorsforward signaling promotes SDF-1-induced endothelial cell chemotaxis and branch-ing remodeling. Blood 2006;108:2914–22.

5. Yu G, Luo H, Wu Y, et al. Ephrin B2 induces T cell costimulation. J Immunol 2003;171:106–14.

1260 Wang et al.

6. Gerety SS, Anderson DJ. Cardiovascular ephrinB2 function is essential for embry-onic angiogenesis. Development 2002;129:1397–410.

7. Stokowski A, Shi S, Sun T, et al. EphB/ephrin-B interaction mediates adult stem cellattachment, spreading, and migration: implications for dental tissue repair. StemCells 2007;25:156–64.

8. Klein R. Bidirectional modulation of synaptic functions by Eph/ephrin signaling. NatNeurosci 2009;12:15–20.

9. Arthur A, Koblar S, Shi S, et al. Eph/ephrinB mediate dental pulp stem cell mobili-zation and function. J Dent Res 2009;88:829–34.

10. Balic A, Mina M. Characterization of progenitor cells in pulps of murine incisors.J Dent Res 2010;89:1287–92.

11. Balic A, Aguila HL, Caimano MJ, et al. Characterization of stem and progenitor cellsin the dental pulp of erupted and unerupted murine molars. Bone 2010;46:1639–51.

12. Shimizu E, Selvamurugan N, Westendorf JJ, et al. HDAC4 represses matrixmetalloproteinase-13 transcription in osteoblastic cells, and parathyroid hormonecontrols this repression. J Biol Chem 2010;285:9616–26.

13. American Association of Endodontists. Glossary of Endodontic Terms, 7th ed.American Association of Endodontics; 2003:41.

14. Lin LM, Rosenberg PA. Repair and regeneration in endodontics. Int Endod J 2011;44:889–906.

15. Cvek M. A clinical report on partial pulpotomy and capping with calciumhydroxide in permanent incisors with complicated crown fracture. J Endod1978;4:232–7.

16. Franceschi RT, Iyer BS, Cui Y. Effects of ascorbic acid on collagen matrix formationand osteoblast differentiation in murine MC3T3-E1 cells. J Bone Miner Res 1994;9:843–54.

17. Levenson GE. Effect of ascorbic acid deficiency on mouse second molar tooth germscultivated in vitro. J Embryol Exp Morphol 1976;36:73–85.

18. Shimizu E, Tamasi J, Partridge NC. Alendronate affects osteoblast functions by cross-talk through EphrinB1-EphB. J Dent Res 2012;91:268–74.

19. Xing W, Kim J, Wergedal J, et al. Ephrin B1 regulates bone marrow stromal celldifferentiation and bone formation by influencing TAZ transactivation via complexformation with NHERF1. Mol Cell Biol 2010;30:711–21.

20. Laurent P, Camps J, About I. Biodentine(TM) induces TGF-beta1 release fromhuman pulp cells and early dental pulp mineralization. Int Endod J 2012;45:439–48.

21. Graham L, Cooper PR, Cassidy N, et al. The effect of calcium hydroxide on solubi-lisation of bio-active dentine matrix components. Biomaterials 2006;27:2865–73.

22. Tziafas D. Reparative dentinogenesis [thesis]. Thessaloniki, Greece: UniversityStudio Press; 1997.

23. Nakashima M. Induction of dentin formation on canine amputated pulp by recombi-nant human bone morphogenetic proteins (BMP)-2 and -4. J Dent Res 1994;73:1515–22.

24. Lovschall H, Fejerskov O, Flyvbjerg A. Pulp-capping with recombinant humaninsulin-like growth factor I (rhIGF-I) in rat molars. Adv Dent Res 2001;15:108–12.

25. Jepsen S, Albers HK, Fleiner B, et al. Recombinant human osteogenic protein-1induces dentin formation: an experimental study in miniature swine. J Endod1997;23:378–82.

JOE — Volume 39, Number 10, October 2013