EBPβ transgenic mice: Reduced alveolar bone mass and site-specific dentin dysplasia

Bone 39 (2006) 552 – 564 www.elsevier.com/locate/bone

Mandibular phenotype of p20C/EBPβ transgenic mice: Reduced alveolar bone mass and site-specific dentin dysplasia T. Savage a,1 , T. Bennett a,1 , Y.-F. Huang b , P.L. Kelly a , N.E. Durant, D.J. Adams c , M. Mina a , J.R. Harrison a,⁎ a

c

Department of Craniofacial Sciences, Pediatric Dentistry and Advanced Education in General Dentistry, University of Connecticut Health Center, Farmington, CT 06030, USA b Department of Medicine, University of Connecticut Health Center, Farmington, CT 06030, USA Department of Orthopaedic Surgery, University of Connecticut Health Center, Farmington, CT 06030, USA Received 11 September 2005; revised 4 January 2006; accepted 11 January 2006 Available online 6 May 2006

Abstract CCAAT enhancer binding proteins (C/EBP) comprise a family of basic-leucine zipper transcription factors that regulate cellular differentiation and function. To determine the role of C/EBP transcription factors in osteoblasts and odontoblasts, we generated a transgenic (TG) mouse model with Co1a1 (pOBCol3.6) promoter-targeted expression of a FLAG-tagged dominant negative C/EBP isoform, p20C/EBPβ (previously LIP). Two of the four transgenic lines presented with abnormalities in the developing incisors, including breakage, overgrowth, and malocclusion. Histological examination revealed that the amount of alveolar bone was reduced in TG compared to wild-type (WT) mice. By microcomputed tomography (microCT), the bone volume fraction of the mandible was reduced at the level of the first and third molars, demonstrating a severe mandibular osteopenia. The lingual dentin morphology of TG incisors differed dramatically from WT. Labial dentin (enamel side) showed normal thickness and tubular dentin structure, whereas the lingual dentin was thinner (25–30% of WT at the alveolar crest) with an amorphous globular structure characteristic of dentin dysplasia. FLAG immunostaining was seen in both lingual and labial odontoblasts, indicating that the sitespecific defect was not due to a lack of labial transgene expression. Northern blot analysis demonstrated reduced osteocalcin expression in TG mandibles, while bone sialoprotein was increased, consistent with prior results in calvariae and long bones. Dental sialophosphoprotein, a marker of the odontoblast lineage whose absence causes dentin dysplasia, was modestly reduced in TG mice by Northern blot and in situ hybridization analysis. By fluorescence microscopy, pOBCol2.3-GFP, a marker of the odontoblast lineage, was expressed in both labial and lingual odontoblasts, although GFP-marked lingual odontoblasts were more flattened than WT cells. Moreover, GFP-positive processes in the lingual dentin tubules were truncated and less organized than those in WT dentin. MicroCT analysis showed reduced tissue density in the lingual dentin. These data suggest that C/EBP transcription factors may be involved in the regulation of odontoblast polarization and dentin matrix production. © 2006 Elsevier Inc. All rights reserved. Keywords: C/EBP; Transcription factor; Incisor; Tooth; Dentin; Osteopenia

Introduction Tooth development requires reciprocal signaling between adjacent epithelial and mesenchymal structures, resulting in the differentiation of ameloblasts and odontoblasts and ultimately in the production of mature enamel and dentin matrices [29,31]. In

⁎ Corresponding author. Fax: +1 860 679 1258. E-mail address: [email protected] (J.R. Harrison). 1 Drs. Savage and Bennett contributed equally to this work. 8756-3282/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2006.01.164

the mouse incisor, which undergoes continuous eruption throughout life, only the labial dentin is covered with enamel, with the lingual portion of the tooth analogous to the molar roots [1]. This asymmetric patterning of the mouse incisor has recently been demonstrated to be due to a follistatin-mediated inhibition of BMP signaling and amelogenesis in the lingual portion of the incisor [35]. The differentiation of ameloblasts and odontoblasts in the labial portion of the incisor is regulated in a spatial and temporal manner. Epithelial and mesenchymal stem cell populations present in the cervical loop region at the posterior end of the incisor give rise to polarizing and differentiating

T. Savage et al. / Bone 39 (2006) 552–564

ameloblasts and odontoblasts [16]. The degree of maturation of enamel and dentin therefore increases progressively in a posterior-to-anterior direction along the longitudinal axis of the tooth. Although many of the extracellular signals and growth factors regulating dentinogenesis have now been elucidated [30], the transcription factors regulating the differentiation of these cell types remain poorly understood. CCAAT enhancer binding proteins (C/EBP) comprise a subfamily of six known transcription factors belonging to the basic-leucine zipper family. C/EBP transcription factors are known to play a role in the regulation of genes involved in energy metabolism [11] as well as in immune and inflammatory responses [36]. C/EBP have emerged as key regulators of cellular differentiation in a variety of cell types including adipocytes [10], hepatocytes [9], granulosa cells [28], mammary epithelium [25], granulocytes [24], and macrophages [38]. We have developed a transgenic mouse model (TG) in which a naturally occurring amino-truncated isoform of C/EBPβ, called p20C/EBPβ (originally liver-enriched inhibitory protein, or LIP [12]), is expressed using a 3.6-kb Col1a1 promoter construct. Studies using various reporter genes, most recently green fluorescent protein (GFP), have demonstrated expression of this promoter in both osteoblasts and odontoblasts [3,4,18]. Since p20C/EBPβ contains the DNA binding and dimerization domains but lacks transactivation domains, it can act as a dominant negative inhibitor of C/EBP function. TG mice exhibit osteopenia secondary to reduced bone formation and altered levels of osteoblast marker gene expression, suggesting a role for C/EBP transcription factors in osteoblast differentiation or function [17]. The odontoblast is closely related to the osteoblast and shares with it the expression of a common subset of genes, including runx2, col1a1, osteocalcin (OC), bone sialoprotein (BSP) and dentin matrix protein-1 (DMP-1) [5,7,8,14]. Other genes, such as dentin sialophosphoprotein (DSPP), appear to be more specific to the odontoblast lineage [27]. A recent report has suggested that C/EBPβ may regulate the expression of DSPP in vitro, indicating a role for these transcription factors in the control of odontoblast gene expression [21]. We therefore hypothesized that C/EBP transcription factors may play a key role in the regulation of odontoblast differentiation and dentinogenesis. In the present study, we demonstrate expression of endogenous C/EBP transcription factors in odontoblasts in vivo. Further, we report that p20C/EBPβ TG mice exhibit a tooth phenotype characterized by a site-specific dentin dysplasia in the lingual dentin of the mandibular incisor, possibly implicating C/EBP transcription factors in the process of root odontoblast differentiation. Materials and methods Generation of pOBCol3.6-p20C/EBPβ transgenic mice The generation of the p20C/EBPβ expression construct and TG mouse lines was described previously [17]. Briefly, expression of p20C/EBPβ was targeted to osteoblasts and odontoblasts by cloning an amino-terminal FLAG-tagged p20C/EBPβ cassette downstream of the 3.6-kb Col1a1 promoter and first intron (pOBCol3.6) and upstream of the bovine growth hormone polyadenylation sequence. Following microinjection into mouse oocytes, a total of four TG

553

founders (2 male, 2 female) were obtained and bred with wild-type CD-1 mice to establish TG lines, designated lines 50-9, 50-10, 63, and 65.

Histology and immunohistochemistry Mandibles harvested from WT and TG mice at 6 to 8 weeks of age were fixed in 4% paraformaldehyde, decalcified in EDTA, and embedded in paraffin. Thin sections were prepared and stained with hematoxylin and eosin for routine histology. For immunohistochemistry, mandibles were fixed in 4% paraformaldehyde, decalcified overnight with 5% nitric acid, and embedded in paraffin. Five-micrometer sections were cut and mounted on charged, precleaned slides. The slides were deparaffinized in xylene, and rehydrated in decreasing concentrations of ethanol. Endogenous peroxidases were quenched by incubating slides in 3% H2O2. Normal goat serum in PBS containing 0.1% BSA was used to prevent non-specific staining. Tissues were incubated overnight at 4°C with rabbit anti-C/EBPα and β primary antibodies (Santa Cruz Biotechnology), at a dilution of 1:200. This was followed by incubation with biotinylated goat–anti-rabbit IgG secondary antibody for 30 min, then incubation in ABC Reagent for 30 min at room temperature. For FLAG immunohistochemistry, sections were incubated with the monoclonal M2 antiFLAG antibody and a rabbit anti-mouse secondary antibody. Slides were developed in DAB-H2O2 for 5 min, counterstained briefly, dehydrated, and mounted.

Preparation of cryosections Mandibles were dissected free of adherent tissues, bisected at the symphysis and fixed in fresh 4% paraformaldehyde at 4°C under constant agitation for 3 days. Mandibles were then decalcified in 14% EDTA for 3–4 days, washed in PBS, soaked in 30% sucrose in PBS overnight, and embedded in OCT compound. Five-micrometer sections were cut on a cryotome, captured using the Cryo-Jane tape-transfer system (Instrumedics, Inc.), and transferred to charged, precleaned slides for analysis by fluorescence microscopy and in situ hybridization. Localization of the p20C/EBPβ transgene was detected by fluorescence immunostaining in cryosections using an anti-C/EBPβ antibody. Slides were washed with 0.5% BSA in PBS and blocked for 1 h using 5% normal donkey serum in 0.1% BSA in PBS. Tissues were incubated overnight at 4°C with a rabbit C/EBPβ primary antibody at a dilution of 1:200. This was followed by incubation with donkey anti-rabbit IgG secondary antibody at a dilution of 1:100 for 1 h at room temperature. Slides were rinsed in PBS and mounted under 50% glycerol in 1 mM MgCl.

pOBCol2.3-GFP expression Transgenic mice have been generated that contain a GFP reporter construct under the control of the rat 2.3-kb Col1a1 promoter and a portion of the Col1a1 first intron (pOBCol2.3). Hemizygous p20C/EBPβ transgenic mice were crossed with mice homozygous for pOBCol2.3-GFP mice so that progeny were either WT or TG with respect to the p20C/EBPβ transgene on a hemizygous pOBCol2.3-GFP background. Mandibles of 6- to 8-week-old WT and TG mice were dissected and processed for frozen sections. Cryosections were analyzed for GFP activity by fluorescence microscopy and images acquired with a Spot camera.

In situ hybridization Expression of DSPP mRNA was examined by in situ hybridization in frozen tissue sections using a 32P-labeled DSPP anti-sense RNA probe as described previously [4].

Microradiography WT and TG mice were sacrificed, and the heads were positioned laterally in a Faxitron MX-20 Specimen Radiography System (Faxitron X-ray Corp., Wheeling, IL), then radiographed at an energy of 15 kV on Kodak X-OMAT film.

554

T. Savage et al. / Bone 39 (2006) 552–564

Microcomputed tomography A qualitative and quantitative 3D analysis of the teeth and surrounding alveolar bone was performed using micro-CT imaging (μCT20 and μCT40, Scanco Medical AG, Bassersdorf, Switzerland) at the University of Connecticut Health Center MicroCT Facility. Mandibles were dissected from 15-day or 6-week-old WT or TG mice, bisected at the symphysis and fixed, and scanned in 70% ethanol. Serial images were acquired transverse to the longitudinal axis of the mandible at 55 kVP (145 μA), employing 1000 conebeam projections per revolution at an integration time of 300 μs. Digital images were reconstructed at a spatial resolution of 18 μm (171,468 isometric voxels/mm3) and 12 μm (578,704 isometric voxels/mm3) for the 6-week and 15-day-old mice, respectively. In 6-week-old mice, sections at the level of the alveolar crest were used to measure enamel thickness and area, as well as labial and lingual dentin thickness and total dentin area. In 15-day-old mice, three locations along the anteroposterior length of the incisor were selected for morphometric and densitometric analysis: at the level of the (1) alveolar crest (first section in which alveolar bone appeared), (2) first molar (center of the mesial root), and (3) third molar. At each of these locations, the fractional bone volume was defined and measured within the osseous margins, excluding the teeth. To quantify degree of mineralization of incisors, tissue density of the enamel, labial dentin, and lingual dentin were measured volumetrically through 10 serial sections (120 μm span) after calibrating Xray attenuation to a hydroxyapatite phantom (mg HA/cm3). In addition, the tissue density of the coronal and developing radicular dentin was determined in the mesial root of the first molar. Data were analyzed statistically using t tests with a significance level of P < 0.05.

RNA extraction and Northern blot analysis Mandibles were harvested and bisected, then further divided into three segments; the most anterior segment containing the anterior portion of the incisor (I), the middle segment containing the molars and posterior incisor (M),

and the posterior segment containing the mandibular condyles (C). These segments were snap frozen in liquid nitrogen and extracted in TRIzol reagent (Invitrogen, Carlsbad, CA), followed by chloroform extraction and precipitation of RNA with isopropanol. RNA was resuspended in GTC, re-precipitated with ethanol and washed. Quantitation of RNA was determined by measurement of A260/A280. RNA was fractionated on a 1% agarose gel in MOPS buffer containing 2.2 M formaldehyde and transferred to GeneScreen (New England Nuclear, Boston, MA) by capillary blotting. Filters were hybridized with random-primed 32P-labeled cDNA probes at 42°C.

Results Endogenous C/EBP expression in incisors To date, the expression of C/EBP transcription factors in teeth has not been examined. We therefore looked for expression of C/EBPα and β in the incisors of 6-week-old WT mice by immunohistochemistry. We found evidence for expression of both C/EBPα and β in ameloblasts and in both the labial and lingual odontoblasts of the incisor (Fig. 1). Although C/EBPβ, and to a lesser extent C/EBPα, expression has been reported in numerous tissues and cell types, expression appeared to be relatively specific to these cell types in the tooth, since the dental pulp adjacent to the odontoblasts showed very weak C/EBP immunoreactivity. We also noted that labial and lingual odontoblasts showed differences in the subcellular localization of endogenous C/ EBP, with lingual odontoblasts exhibiting more intense nuclear staining of C/EBPα and β.

Fig. 1. Endogenous C/EBPα and C/EBPβ transcription factors are expressed in odontoblasts of the mouse incisor. Low power images show expression of C/EBPα and C/EBPβ in lingual (LiO) and labial (LaO) odontoblasts, with weak staining in the intervening dental pulp (P) cells. Ameloblasts (Am) are also visible in low power images and stain strongly for both transcription factors. At high power, a difference in the intracellular distribution is visible between labial (largely cytoplasmic) and lingual (primarily nuclear) odontoblasts. Control sections were probed with non-immune IgG as the primary antibody. Yellow arrows indicate the odontoblast layers. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

T. Savage et al. / Bone 39 (2006) 552–564

Gross dental phenotype of pOBCol3.6-p20C/EBPβ mice Two lines of TG mice (lines 50-10 and 65) showed decreased body weight and gross manifestations of a dental phenotype in the hemizygous state. Line 65 appears to have the most severe phenotype and the highest level of transgene mRNA expression [17]. In these two lines, we have observed misalignment of upper and lower incisors, breakage of both upper and lower incisors, and overgrowth of the mandibular incisors (Figs. 2A–F). The latter example provides the most dramatic phenotype, resulting in a tusklike appearance of the teeth. Microradiography demonstrated a more pronounced curvature of the TG incisors, contributing to the malocclusion of the upper and lower teeth (Figs. 2G, H). Since mouse incisors grow rapidly (2.8 mm/week for the lower incisors), the malocclusion phenotype has necessitated trimming the teeth in affected animals on a weekly basis. All transgenic lines with malocclusion also received supplemental granulated diet to ensure adequate nutrition. Two additional TG lines (50-9 and 63) that show lower levels of transgene expression in bone [17] do not show gross tooth abnormalities in hemizygous animals. However, crossing hemizygous male and female mice of these lines has produced offspring with a similar tooth

555

phenotype to that seen in line 50-10 and 65 hemizygotes, suggesting a gene dosage-dependent effect. Histological analysis As in the long bones and calvariae, we found clear histological evidence of osteopenia in the mandible, manifested by sparse alveolar bone surrounding the incisors and the molars in TG mice (Figs. 3A, B). A sagittal section through the molars showed robust alveolar bone forming the tooth socket in WT mice, while TG animals showed a lack of alveolar bone surrounding the root and in the inter-radicular spaces. Examination of the mandibular incisors in sagittal section revealed a striking asymmetry in the structure of the dentin (Figs. 3C, D). The labial dentin (enamel side) in TG mice appeared to be histologically normal, although in line 50-10, it was somewhat wider than in WT controls. The lingual dentin, on the other hand, was thinner in TG mice and appeared to be structurally abnormal along the entire anteroposterior length of the tooth. At high power, it was evident that the normal tubular dentin structure was absent, replaced by dentin matrix with a globular morphology (Figs. 3E, F). Globular dentin has been reported to be associated

Fig. 2. Transgenic mice from lines 50-10 (B, C) and 65 (E, F) show evidence of malocclusion, overgrowth, and breakage of incisors compared to their respective wildtype littermates (A and D). Radiographic images show an increased curvature of the incisors in a transgenic mouse (H) relative to its wild-type littermate (G).

556

T. Savage et al. / Bone 39 (2006) 552–564

Fig. 3. Transgenic mice show reduced mandibular bone and reduced thickness of the labial dentin of the incisor. Low (A, B), medium (C, D) and high power (E, F) views of the mandible of a wild-type (A, C, E) and a transgenic mouse (B, D, F). Note the lack of alveolar bone (b) surrounding the incisor and first molar in the transgenic mouse, comparing panels A and B. Panels C and D demonstrate the thin lingual dentin (Li) along with the increase in the thickness of the labial dentin (La) seen only in line 50-10. Panels E and F contrast the morphology of the lingual dentin at high power, with the globular dentin matrix apparent in the transgenic mouse (F).

with defective mineralization in other genetic models of dentin dysplasia [26,37]. MicroCT analysis of incisor morphology We used a novel approach to quantify this structural defect based on a microCT analysis of the tooth microarchitecture. Incisors were scanned transversely, and the section at the level of the alveolar crest (i.e., the first section to contain bone) was used for analysis of dentin and enamel thickness and area (Fig. 4A). The enamel is highly mineralized and therefore very opaque, providing a landmark reference for the labial aspect of the tooth. Both TG lines showed a significant decrease in the lingual dentin thickness, which gave the transgenic incisors a characteristic asymmetric appearance in cross-section (Figs. 4B and C). The labial dentin thickness was increased in line 50-10 but was not significantly different in line 65. Both lines showed a significant reduction of the total dentin area. While not measured, the size of the pulp chamber was substantially increased in TG incisors. No significant changes in enamel thickness or area were observed in either TG line.

Dentin tissue density Since the dentin matrix morphology suggested a possible hypo-mineralization phenotype, we examined the inherent tissue density of both the labial and lingual dentin in 15-dayold mice from line 50-10 using a microCT-based approach. As shown in Fig. 5, the incisor lingual dentin tissue density was reduced substantially at both the level of the first molar and at the alveolar crest (Fig. 5A). The tissue density of the labial dentin, on the other hand, was not as profoundly affected, showing a slight but significant increase at the level of the first molar and a slight decrease at the alveolar crest location. There were no changes in the tissue density of the incisor enamel measured at the level of the first molar (Fig. 5B). We found no evidence for alteration of the dentin tissue density in the molars, either in the crown or in the developing root (Fig. 5C). The decrease in the amount of mandibular bone observed in histological sections was quantitatively confirmed by microCT. Significant reductions in the bone volume fraction of 38% and 47% were seen in TG mandibles at the level of the third and first molar, respectively.

T. Savage et al. / Bone 39 (2006) 552–564

557

Fig. 4. Quantitative microCT analysis of mouse incisor morphology confirms the reduction in the labial dentin thickness. (A) Incisors were scanned transversely and the first section containing bone (B) was analyzed. Linear measurements of enamel (E) and labial and lingual dentin (D) thickness as well as enamel and dentin area were made. (B) Representative images of line 50-10 incisors showing enlarged pulp chambers and asymmetric dentin distribution in transgenic mice. (C) Quantitative analysis of linear and area measurements in lines 50-10 and 65. Values represent means ± SEM; n = 4–10; *P < 0.05.

Immunohistochemical transgene localization A trivial explanation for the selectivity of the dentin defect in this model would be if expression of the p20C/EBPβ transgene were restricted to the lingual odontoblasts. Such a finding was not anticipated, since prior studies of pOBCol3.6 promoter expression in teeth using GFP reporter constructs demonstrated expression in odontoblasts lining both the labial and lingual surfaces [3,4]. Fig. 6A shows diffuse, mainly cytoplasmic expression of endogenous C/EBPβ in the labial odontoblasts of a WT mouse, whereas Fig. 6B shows an intense nuclear signal in the labial odontoblasts of a TG animal. No staining was seen

in the non-immune control (Fig. 6C). Fig. 6D represents a tangential section of the incisor showing the transition from lingual to labial odontoblasts, both of which show intense nuclear staining that, based on comparison to WT mice (panel A), is due to overexpression of the p20C/EBPβ transgene. This interpretation was confirmed by immunohistochemistry using the anti-FLAG antibody, which showed intense nuclear expression of the transgene in both lingual and labial odontoblasts (Fig. 6E). Not surprisingly, the expression pattern seen with the anti-FLAG antibody is very similar to that observed previously for pOBCol3.6-GFP. For example, expression of the p20C/EBP transgene in odontoblasts near the

558

T. Savage et al. / Bone 39 (2006) 552–564

Fig. 5. MicroCT analysis of tissue density shows hypomineralization of incisor lingual dentin in 15-day old WT and TG line 50-10 mice. (A) Incisor labial and lingual tissue density measured at two sites; the level of the mesial root of the first molar and the alveolar crest. (B) Incisor enamel density at the level of the first molar. (C) Molar dentin tissue density in the crown and in the developing root. (D) Fractional bone volume (BV/TV) of the incisor at the level of the third and first molar, showing the reduction in bone volume at both locations. Values represent means ± SEM; n = 12; *P < 0.05, **P < 0.01.

cervical loop region at the apical end of the incisor is strikingly reminiscent of the spatial pattern of pOBCol3.6-GFP expression, showing increased expression during the process of labial odontoblast polarization and differentiation (Fig. 6G). The only other cells showing high levels of FLAG immunoreactivity in the mandible sections were osteoblasts; no readily apparent staining was seen in ameloblasts or in dental pulp cells. Thus, the apparent lack of effect of the transgene on labial odontoblast function and dentin morphology is not due to the absence of transgene expression in these cells. pOBCol2.3-GFP expression in transgenic incisors In mouse incisors, expression of both pOBCol3.6- and pOBCol2.3-GFP is restricted to mature odontoblasts [3,4], with increased fluorescence intensity observed as cells undergo polarization and differentiation anterior to the cervical loop. We therefore examined the pattern and intensity of pOBCol2.3-GFP expression as an established marker of the differentiated odontoblast. GFP was strongly expressed in the odontoblasts lining the both the labial and lingual dentin of WT mice (Figs. 7A, C). In TG mice, GFP was also expressed in both labial and lingual odontoblasts (Figs. 7B, D) and at a level comparable to that seen in WT mice. This suggests that odontoblast differentiation, by the criterion of pOBCol2.3-GFP expression, was not inhibited in lingual odontoblasts. However, visualiza-

tion of GFP fluorescence clearly delineated the shape of odontoblasts lining the lingual surface of the dentin, revealing a more flattened shape than that seen in lingual odontoblasts of WT mice (Figs. 7B, D). Moreover, GFP fluorescence was clearly visible in odontoblast processes within the dentin tubules. In WT mice, in both labial and lingual dentin, and in the labial odontoblasts of TG mice, odontoblast processes were organized with the main trunks in fairly regular parallel arrays, extending across the entire width of the dentin matrix (Fig. 7E). In contrast, TG lingual odontoblast processes were irregular and truncated, with more extensive branching and a failure to extend across the full width of the dentin matrix (Fig. 7F). It therefore appears that odontoblast polarization, including attainment of normal cell height and the elaboration of cell processes, was inhibited by p20C/EBPβ overexpression in lingual odontoblasts. Gene expression in mandible WT and TG line 65 mandibles were divided into 3 regions – incisal (I), molar (M) and condylar (C) – prior to extraction of total RNA and Northern blot analysis of gene expression (Fig. 8). Northern analysis showed a decrease in DSPP expression in segments I and M. As expected, there was little or no DSPP or ameloblastin expression in segment C, which lacks dentition. All segments showed an increase in BSP and a decrease in OC

T. Savage et al. / Bone 39 (2006) 552–564

559

both TG lines (Figs. 9A, B, D, F). The DSPP hybridization signal seen on the lingual surface of TG line 50-10 mice was thinner than that on the labial side corresponding to the change in cell shape but appeared to be comparable in terms of intensity (compare Figs. 9C and E). In line 65 (Fig. 9G), in addition to the reduction in the domain of expression, the signal intensity also appeared to be reduced relative to WT controls. Histological analysis of molars Since the lingual dentin represents an analogous structure to the molar roots, we have examined the molars for signs of radicular dentin dysplasia. As shown in Fig. 5, we did not find evidence for decreased tissue density or severe architectural defects at post-natal day 15, during development of the molar roots [20]. We have, however, occasionally seen morphological evidence of dentin dysplasia in the molar roots of older mice. As shown in Fig. 10, radicular dentin in cross-section from 6-week-

Fig. 6. The p20C/EBPβ transgene (line 50-10) is expressed in both labial and lingual odontoblasts. (A–D) Immunofluorescence using an anti-C/EBPβ primary antibody in cryosections. There is diffuse, primarily cytoplasmic, staining of endogenous C/EBPβ in wild-type labial odontoblasts (A). In transgenic mice, intense nuclear staining is seen in both labial (B, D) and lingual (D) odontoblasts. Panel C represents a non-immune control. (E–G) Specific transgene immunostaining in paraffin sections probed with the M2 anti-FLAG antibody. Panel E shows transgene immunostaining in labial (bottom) and lingual (top) odontoblasts in paraffin sections probed with the M2 anti-FLAG antibody. Panel F shows the lack of M2 staining in a wild-type mouse. In panel G, labial odontoblasts near the cervical loop show increased expression of the transgene following polarization and differentiation.

expression, reflecting changes observed previously in the long bones and calvariae of these mice. Expression of ameloblastin, primarily a marker of the ameloblast [2,19], was not affected. To examine DSPP expression in specific odontoblast populations, we performed in situ hybridization analysis. In situ hybridization of mandible cryosections revealed DSPP expression in the labial and lingual odontoblasts of WT and TG mice (Fig. 9). As expected, odontoblasts were the only cells in the mandible that expressed high levels of DSPP mRNA. The spatiotemporal pattern of DSPP expression in odontoblasts was similar to that of pOBCol2.3-GFP expression, confirming the previously reported co-expression of these markers [4]. Labial odontoblasts showed an intense DSPP signal in WT mice and in

Fig. 7. pOBCol2.3-GFP expression, cell shape and dentin tubule processes in transgenic mice. (A–B) Low power sagittal images showing strong Col2.3-GFP expression in odontoblasts of wild-type (A) and line 50-10 transgenic (B) incisors. (C–F) High power images showing pOBCol2.3-GFP expression in wild-type (C, E) and transgenic (D, F) labial (C, D) and lingual (E, F) odontoblasts. Note the loss of cell height and polarity in the transgenic lingual odontoblasts (F). Panels G and H are longer exposures showing the structure of the labial (G) and lingual (H) odontoblast processes within the dentin matrix of a transgenic mouse.

560

T. Savage et al. / Bone 39 (2006) 552–564

odontoblasts of the mouse incisor. The more intense nuclear staining of endogenous C/EBPα and C/EBPβ seen in lingual compared to labial odontoblasts indicates a possible difference in the intracellular localization of these transcription factors between the labial and lingual compartments. The significance of this is not clear, but it may be associated with differential transcriptional regulation by C/EBP in these odontoblast populations. It seems reasonable to postulate that the more nuclear distribution in lingual cells might be associated with a greater level of C/EBP-dependent transcriptional activity. This might account, in part, for the preferential inhibitory effects of

Fig. 8. DSPP expression is modestly reduced by Northern blot analysis. WT and TG line 65 mandibles were divided into three segments (incisal, I; molar, M; condylar, C) and total RNA extracted. DSPP expression was reduced in the segments containing dentition, while the ameloblast marker ameloblastin was unaffected. Bone sialoprotein (BSP) mRNA was increased in all segments, while osteocalcin mRNA was reduced.

old mice was sometimes thinner, presenting a globular morphology along with cellular inclusions (Figs. 10A, D). Tooth crowns appeared to be normal in these instances, with a transition from normal to dysplastic dentin morphology at the level of the cemento-enamel junction (Figs. 10C, F). Discussion In the present study, we report that TG mice overexpressing p20C/EBPβ have an overt dental phenotype grossly characterized by malocclusion, overgrowth and fragility of the incisors. Histology and microCT analysis revealed enlarged pulp chambers in the incisor and a site-specific dentin dysplasia in which the lingual dentin matrix of the incisor is thinner and displays a globular morphology. This defect was associated with odontoblasts showing an abnormal morphology, including reduced cell height and absent or truncated cellular processes. Little is known about the role of C/EBP transcription factors in the differentiation or function of tooth cells and there have been no reports to date on in vivo expression patterns of endogenous C/EBP transcription factors in teeth. We found direct evidence for in vivo expression of endogenous C/EBPα and β transcription factors in both the labial and lingual

Fig. 9. In situ analysis of DSPP expression. (A) Low power comparison of DSPP expression in sagittal sections of a wild-type and line 50-10 and 65 transgenic mice. Mandibles are oriented with the lingual dentin toward the top of the image. (B–G) High power images of in situ DSPP mRNA expression in wild-type (B, C) and lines 50-10 (D, E) and 65 (F, G) transgenic labial (B, D, F) and lingual (C, E, G) odontoblasts. Note the reduced domain of DSPP expression in transgenic lingual odontoblasts (E, G), and the reduced signal intensity in transgenic line 65 (G).

T. Savage et al. / Bone 39 (2006) 552–564

561

Fig. 10. Histology of wild-type (A–C) and transgenic (D–F) molars in 6-week-old WT and TG line 50-10 mice. TG molars occasionally displayed a globular dentin matrix with cellular inclusions (*) (D, E). In some cases, the radicular dentin was found to be thin, with a sharp transition occurring at the cemento-enamel junction (J) (compare panels C and F).

the p20C/EBPβ transgene on the lingual aspect of the incisor. Evidence for endogenous C/EBP protein expression in odontoblasts coupled with the dentin defect in transgenic mice provides support for a role of these transcription factors in regulating odontoblast differentiation and/or function. The expression of C/EBPα and β we observed in ameloblasts is consistent with a report that the LS8 ameloblast cell line expresses C/EBPα [39]. Moreover, a C/EBP binding site in the amelogenin promoter was required for both basal and C/EBPαstimulated promoter activity, suggesting a role for C/EBP transcription factors in ameloblast differentiation and enamel production. Since pOBCol2.3-GFP has been established as a marker for osteoblast and odontoblast differentiation, use of this reporter allowed us to analyze whether the dentin defect in TG mice is due to the transgene interfering with terminal odontoblast differentiation. Previous work demonstrated that the spatiotemporal pattern of pOBCol2.3-GFP in odontoblasts correlated tightly with odontoblast differentiation, being restricted to polarized dentin-secreting cells [4]. We therefore wanted to assess whether the pattern or intensity of GFP expression was altered in the TG mice, particularly in regions exhibiting dentin dysplasia. Furthermore, recent work in our laboratory demonstrated a dramatic decrease in pOBCol2.3-GFP expression in femoral osteoblasts of TG mice and in osteoblasts differentiated in ex vivo primary bone marrow stromal or calvarial cell cultures [17]. Given that osteoblasts and odontoblasts are closely related cells that express a battery of overlapping genes, we predicted that a similar effect on pOBCol2.3-GFP expression might be seen in the dentin of transgenic mice. Contrary to the findings in osteoblasts, however, no alterations in the intensity of GFP expression were apparent in the odontoblasts of TG incisors. pOBCol2.3-GFP was strongly expressed in both labial and lingual odontoblasts of these mice, with the intensity of expression comparable to that observed in their WT littermates. These data indicate that odontoblast

differentiation, at least as assessed by pOBCol2.3-GFP expression, is not preferentially inhibited in lingual odontoblasts of TG mice. Moreover, it suggests that there is differential cell context-dependent regulation of the pOBCol2.3 promoter by C/EBP transcription factors in osteoblasts and odontoblasts. While the intensity of pOBCol2.3-GFP expression was not inhibited, the GFP analysis did reveal a striking change in odontoblast morphology in regions associated with dysplastic dentin. TG lingual odontoblasts did not appear to achieve their normal cell height, columnar appearance, or polarity. The cell bodies appeared flattened, and odontoblast processes were missing, truncated, or disorganized, exhibiting less parallelism within the dentin matrix than those in WT mice or in labial TG odontoblasts present in the same section. These observations suggest that p20C/EBPβ expression may affect the ability of these cells to polarize and produce a normal dentin matrix. Unda et al. [33,34] examined the odontoblastic differentiation of dental papilla cells in vitro and were able to define combinations of growth factors that differentially regulated odontoblast polarization and extracellular matrix production, thereby dissecting these processes. In those studies, a combination of TGFβ and FGF-1 promoted extensive extracellular matrix production, while TGFβ plus FGF-2 stimulated greater odontoblast polarization and the development of cell processes. It is therefore interesting to speculate that FGF-2 and TGFβ signaling might regulate odontoblast polarization via a C/EBPdependent pathway. Lastly, since pOBCol2.3-GFP was expressed at a high level in TG lingual odontoblasts, it appears that although cell polarization normally precedes pOBCol2.3GFP expression in vivo, it is not a strict requirement for pOBCol2.3 promoter activation. In vitro studies using odontoblast cell lines have demonstrated the presence of a C/EBP binding site within the DSPP promoter and postulated a possible role for C/EBP transcription factors in regulating the tissue-specific expression of DSPP in the odontoblast lineage [21]. Moreover, dspp null mice exhibit

562

T. Savage et al. / Bone 39 (2006) 552–564

dentin dysplasia, hypomineralization, decreased dentin width, a widened zone of predentin, and enlarged pulp chambers [26]. Finally, in the present study, Northern blot analysis using mandible segments showed a modest reduction in the levels of DSPP mRNA in line 65 TG mice. Taken together, these findings suggested that the dentin defect in p20C/EBPβ overexpressing mice might be associated with altered DSPP expression. We therefore anticipated that DSPP expression would be downregulated specifically in lingual odontoblasts, whereas labial dentin odontoblasts would demonstrate normal levels of DSPP expression. From our in situ hybridization analysis, it is evident that both labial and lingual odontoblasts in TG mice expressed at least some level of DSPP mRNA. The more flattened morphology seen in TG lingual odontoblasts makes interpretation of the in situ hybridization signal somewhat difficult since the DSPP hybridization signal was confined to a smaller domain in these cells. This was particularly true for line 50-10, in which the DSPP signal appeared to be of a similar intensity to that seen in WT odontoblasts, though restricted to a narrow band corresponding to the flattened cell layer. In the more severely affected TG line 65, however, both the domain and the DSPP signal intensity were reduced in TG lingual odontoblasts, suggesting a reduction in DSPP mRNA levels and a likely inhibitory effect of p20C/EBPβ on dspp gene transcription. Given the important role of DSPP in dentin mineralization established in dspp null mice, we expected that the inhibition of DSPP expression would cause reduced dentin mineralization in the affected region. We therefore used a microCT approach to make quantitative measurements of the local tissue density, reflecting a combination of matrix mineralization and porosity within the sampled region. By microCT, we were able to detect the expected posterior-to-anterior gradient in dentin and enamel tissue density, i.e., increasing matrix density proceeding from the apex toward the cutting surface of the tooth. This pattern is consistent with the known spatial and temporal maturation of the mouse incisor dentin and enamel matrices that occurs during continuous incisor eruption. Similarly, a significant difference was seen between the tissue density of mature coronal dentin and that of the developing radicular dentin in the molars of 15day-old mice. As anticipated, the tissue density of the enamel was not significantly different between WT and TG mice. Unexpectedly, we did find modest changes in the labial dentin tissue density in the incisors of TG mice, with a slight but significant increase at the level of the first molar and a slight decrease at the alveolar crest location. On the other hand, the larger decrease seen in lingual dentin tissue density of TG mice was not surprising given the hypomineralization defect seen in other transgenic models exhibiting dysplastic globular dentin, including DSPP null mice [26], TGFβ overexpressing mice [32], and DMP-1 knockout mice [37]. However, additional studies will be required to distinguish between an inherent mineralization defect in the dentin matrix and a possible increase in dentin porosity at a level below the resolution of the microCT analysis. MicroCT also revealed dramatic differences in the quantity of alveolar bone in TG mice. Consistent with the osteopenic phenotype seen in long bones and vertebrae of transgenic mice

[17], the mandibles showed significant reductions in bone volume. A 38% and 47% reduction in the bone volume fraction was observed at the levels of the third molar and first molar, respectively, in 15-day-old TG mice. In the long bones, dynamic histomorphometry indicated a significant inhibition of bone formation, but osteoclast number was unchanged [17]. Preliminary observations in mandibles, on the other hand, have shown evidence of surface pitting, a loss of alveolar height surrounding the molars, and increased TRAP staining in frozen sections, suggesting that the reduced bone mass seen in the mandibles may result from effects on both bone formation and resorption. Northern blot analysis demonstrated reduced osteocalcin expression in the mandibles of TG mice, as we reported previously in the long bones and calvariae of these animals [17]. It is clear from our earlier work and from examination of the edentulous portion of the mandible in the present study that OC mRNA levels are reduced in bone. This is consistent with a report that C/EBPβ and Runx2 act in synergy to enhance OC gene transcription in osteoblasts [15]. Since OC null mice show increased bone mass and a higher rate of bone formation [13], consistent with an inhibitory action of OC on osteoblast activity, the relative lack of OC expression is unlikely to account for the TG bone phenotype. Rather, decreased OC coupled with reduced expression of pOBCol2.3-GFP, another late osteoblast marker, and a decrease in the rate of bone formation suggest an inhibition of terminal osteoblast differentiation by p20C/EBPβ, resulting in the severe osteopenia. Interestingly, BSP expression is enhanced, suggesting a possible accumulation of less differentiated osteoprogenitor cells. It is not clear from the present study whether OC expression is similarly affected in odontoblasts. However, it has been reported that dentin thickness and mineralization was largely unaffected in OC null mice [6], suggesting that OC insufficiency in odontoblasts would not contribute to the observed dentin phenotype in our model. The precise molecular mechanisms underlying the effect of p20C/EBPβ on DSPP expression and dentinogenesis are likely to be complex, since C/EBPβ isoforms could play roles in both activation and repression of the dspp promoter. Narayanan et al. [21] have reported that C/EBPβ represses DSPP expression in undifferentiated odontoblasts via a direct association with another bZip transcription factor, Nrf1. The C/EBPβ-Nrf-1 complex then binds to juxtaposed C/EBP and Nrf1 binding sites in the dspp promoter to synergistically repress transcription. C/ EBPβ and Nrf1 continue to be expressed during odontoblast differentiation and mineralization, but no longer bind to the dspp promoter, indicating a loss of binding capacity or derepression by formation of a higher order complex. Overexpression of p20C/EBPβ might interfere with this derepression mechanism by occupying the C/EBP binding site following odontoblast differentiation, leading to a partial repression of dspp transcription. Alternatively, p20C/EBPβ might compete for binding of other corepressors to the C/EBPβNrf1 complex, allowing prolonged occupancy of the site by this complex and preventing de-repression. We have also found immunohistochemical evidence for C/EBPα expression in

T. Savage et al. / Bone 39 (2006) 552–564

odontoblasts. Although a role for C/EBPα has not been established in odontoblasts, in other systems, C/EBPα has been shown to induce growth arrest and promote terminal differentiation [23]. It remains unclear why the dentin defect is restricted to the lingual dentin, but it likely involves differential C/EBP function between these heterogeneous odontoblast populations. Interestingly, based on observations in the fragilitas ossium (fro/fro) mouse and in the TGFβ1 overexpressing transgenic mouse models, both of which exhibit hypoplasia of the lingual incisor, Opsahl et al. [22] have proposed that the lingual-forming part of the incisor may comprise a structural entity. Our data provide additional support for this hypothesis and suggest that C/EBP transcription factors could represent a transcriptional pathway integrating signals that regulate the development and maintenance of this region. Further, the alterations in dentin and odontoblast morphology in this TG model suggest a possible role for the C/EBP family of transcription factors in the elongation, polarization, and final differentiation of odontoblasts in enamel-free areas of the developing dentition.

[15]

[16]

[17]

[18]

[19]

[20]

[21]

References [1] Amar S, Karcher-Djuricic V, Meyer JM, Ruch JV. The lingual (root analogue) and the labial (crown analogue) mouse incisor dentin promotes ameloblast differentiation. Arch Anat Microsc Morphol Exp 1986;75:229–39. [2] Begue-Kirn C, Krebsbach PH, Bartlett JD, Butler WT. Dentin sialoprotein, dentin phosphoprotein, enamelysin and ameloblastin: tooth-specific molecules that are distinctively expressed during murine dental differentiation. Eur J Oral Sci 1998;106:963–70. [3] Braut A, Kalajzic I, Kalajzic Z, Rowe DW, Kollar EJ, Mina M. Col1a1GFP transgene expression in developing incisors. Connect Tissue Res 2002;43:216–9. [4] Braut A, Kollar EJ, Mina M. Analysis of the odontogenic and osteogenic potentials of dental pulp in vivo using a Col1a1-2.3-GFP transgene. Int J Dev Biol 2003;47:281–92. [5] Bronckers AL, Lyaruu DM, Woltgens JH. Immunohistochemistry of extracellular matrix proteins during various stages of dentinogenesis. Connect Tissue Res 1989;22:65–70. [6] Bronckers AL, Price PA, Schrijvers A, Bervoets TJ, Karsenty G. Studies of osteocalcin function in dentin formation in rodent teeth. Eur J Oral Sci 1998;106:795–807. [7] Chen J, Shapiro HS, Sodek J. Development expression of bone sialoprotein mRNA in rat mineralized connective tissues. J Bone Miner Res 1992;7:987–97. [8] Chen S, Gu TT, Sreenath T, Kulkarni AB, Karsenty G, MacDougall M. Spatial expression of Cbfa1/Runx2 isoforms in teeth and characterization of binding sites in the DSPP gene. Connect Tissue Res 2002;43:338–44. [9] Darlington GJ. Molecular mechanisms of liver development and differentiation. Curr Opin Cell Biol 1999;11:678–82. [10] Darlington GJ, Ross SE, MacDougald OA. The role of C/EBP genes in adipocyte differentiation. J Biol Chem 1998;273:30057–60. [11] Darlington GJ, Wang N, Hanson RW. C/EBP alpha: a critical regulator of genes governing integrative metabolic processes. Curr Opin Genet Dev 1995;5:565–70. [12] Descombes P, Schibler U. A liver-enriched transcriptional activator protein, LAP, and a transcriptional inhibitory protein, LIP, are translated from the same mRNA. Cell 1991;67:569–79. [13] Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, et al. Increased bone formation in osteocalcin-deficient mice. Nature 1996; 382:448–52. [14] Gaikwad JS, Hoffmann M, Cavender A, Bronckers AL, D'Souza RN.

[22]

[23]

[24]

[25]

[26]

[27]

[28] [29] [30] [31] [32]

[33]

563

Molecular insights into the lineage-specific determination of odontoblasts: the role of Cbfa1. Adv Dent Res 2001;15:19–24. Gutierrez S, Javed A, Tennant DK, Rees van M, Montecino M, Stein GS, et al. CCAAT/enhancer-binding proteins (C/EBP) beta and delta activate osteocalcin gene transcription and synergize with Runx2 at the C/EBP element to regulate bone-specific expression. J Biol Chem 2002; 277: 1316–23. Harada H, Kettunen P, Jung HS, Mustonen T, Wang YA, Thesleff I. Localization of putative stem cells in dental epithelium and their association with Notch and FGF signaling. J Cell Biol 1999;147:105–20. Harrison JR, Huang YF, Wilson KA, Kelly PL, Adams DJ, Gronowicz GA, et al. Col1a1 promoter-targeted expression of p20 CCAAT enhancerbinding protein beta (C/EBPbeta), a truncated C/EBPbeta isoform, causes osteopenia in transgenic mice. J Biol Chem 2005;280:8117–24. Kalajzic Z, Liu P, Kalajzic I, Du Z, Braut A, Mina M, et al. Directing the expression of a green fluorescent protein transgene in differentiated osteoblasts: comparison between rat type I collagen and rat osteocalcin promoters. Bone 2002;31:654–60. Krebsbach PH, Lee SK, Matsuki Y, Kozak CA, Yamada KM, Yamada Y. Full-length sequence, localization, and chromosomal mapping of ameloblastin. A novel tooth-specific gene. J Biol Chem 1996;271:4431–5. MacNeil RL, Berry JE, Strayhorn CL, Shigeyama Y, Somerman MJ. Expression of type I and XII collagen during development of the periodontal ligament in the mouse. c 1998;43:779–87. Narayanan K, Ramachandran A, Peterson MC, Hao J, Kolsto AB, Friedman AD, et al. The CCAAT enhancer-binding protein (C/EBP)beta and Nrf1 interact to regulate dentin sialophosphoprotein (DSPP) gene expression during odontoblast differentiation. J Biol Chem 2004;279: 45423–32. Opsahl S, Septier D, Aubin I, Guenet JL, Sreenath T, Kulkarni A, et al. Is the lingual forming part of the incisor a structural entity? Evidences from the fragilitas ossium (fro/fro) mouse mutation and the TGFbeta1 overexpressing transgenic strain. Arch Oral Biol 2005;50:279–86. Porse BT, Pedersen TA, Xu X, Lindberg B, Wewer UM, Friis-Hansen L, et al. E2F repression by C/EBPalpha is required for adipogenesis and granulopoiesis in vivo. Cell 2001;107:247–58. Radomska HS, Huettner CS, Zhang P, Cheng T, Scadden DT, Tenen DG. CCAAT/enhancer binding protein alpha is a regulatory switch sufficient for induction of granulocytic development from bipotential myeloid progenitors. Mol Cell Biol 1998;18:4301–14. Seagroves TN, Krnacik S, Raught B, Gay J, Burgess-Beusse B, Darlington GJ, et al. C/EBPbeta, but not C/EBPalpha, is essential for ductal morphogenesis, lobuloalveolar proliferation, and functional differentiation in the mouse mammary gland. Genes Dev 1998;12:1917–28. Sreenath T, Thyagarajan T, Hall B, Longenecker G, D'Souza R, Hong S, et al. Dentin sialophosphoprotein knockout mouse teeth display widened predentin zone and develop defective dentin mineralization similar to human dentinogenesis imperfecta type III. J Biol Chem 2003;278: 24874–80. Sreenath TL, Cho A, MacDougall M, Kulkarni AB. Spatial and temporal activity of the dentin sialophosphoprotein gene promoter: differential regulation in odontoblasts and ameloblasts. Int J Dev Biol 1999;43: 509–516. Sterneck E, Tessarollo L, Johnson PF. An essential role for C/EBPbeta in female reproduction. Genes Dev 1997;11:2153–62. Thesleff I. Epithelial-mesenchymal signalling regulating tooth morphogenesis. J Cell Sci 2003;116:1647–8. Thesleff I, Mikkola M. The role of growth factors in tooth development. Int Rev Cytol 2002;217:93–135. Thesleff I, Sharpe P. Signalling networks regulating dental development. Mech Dev 1997;67:111–23. Thyagarajan T, Sreenath T, Cho A, Wright JT, Kulkarni AB. Reduced expression of dentin sialophosphoprotein is associated with dysplastic dentin in mice overexpressing transforming growth factor-beta 1 in teeth. J Biol Chem 2001;276:11016–20. Unda FJ, Martin A, Hernandez C, Perez-Nanclares G, Hilario E, Arechaga J. FGFs-1 and-2, and TGF beta 1 as inductive signals modulating in vitro odontoblast differentiation. Adv Dent Res 2001;15:34–7.

564

T. Savage et al. / Bone 39 (2006) 552–564

[34] Unda FJ, Martin A, Hilario E, Begue-Kirn C, Ruch JV, Arechaga J. Dissection of the odontoblast differentiation process in vitro by a combination of FGF1, FGF2, and TGFbeta1. Dev Dyn 2000; 218: 480–9. [35] Wang XP, Suomalainen M, Jorgez CJ, Matzuk MM, Werner S, Thesleff I. Follistatin regulates enamel patterning in mouse incisors by asymmetrically inhibiting BMP signaling and ameloblast differentiation. Dev Cell 2004;7:719–30. [36] Wedel A, Ziegler-Heitbrock HW. The C/EBP family of transcription factors. Immunobiology 1995;193:171–85. [37] Ye L, MacDougall M, Zhang S, Xie Y, Zhang J, Li Z, et al. Deletion of

dentin matrix protein-1 leads to a partial failure of maturation of predentin into dentin, hypomineralization, and expanded cavities of pulp and root canal during postnatal tooth development. J Biol Chem 2004;279:19141–8. [38] Zhang DE, Zhang P, Wang ND, Hetherington CJ, Darlington GJ, Tenen DG. Absence of granulocyte colony-stimulating factor signaling and neutrophil development in CCAAT enhancer binding protein alphadeficient mice. Proc Natl Acad Sci U S A 1997;94:569–74. [39] Zhou YL, Snead ML. Identification of CCAAT/enhancer-binding protein alpha as a transactivator of the mouse amelogenin gene. J Biol Chem 2000;275:12273–80.