|Home | About | Journals | Submit | Contact Us | Français|
Cementogenesis is sensitive to altered local phosphate levels; thus, we hypothesized a cementum phenotype, likely of decreased formation, would be present in the teeth of X-linked hypophosphatemic (Hyp) mice. Mutations in the phosphate-regulating gene with homologies to endopeptidases on the X chromosome (Phex) cause X-linked hypophosphatemia, characterized by rickets, osteomalacia, and hypomineralized dentin formation, a phenotype recapitulated in the Hyp mouse homolog. Here, we report a developmental study of tooth root formation in Hyp mouse molars, focusing on dentin and cementum.
Light and transmission electron microscopy were used to study molar tissues from wild-type (WT) and Hyp mice. Demineralized and hematoxylin and eosin–stained tissues at developmental stages 23 to 96 days postcoital (dpc) were examined by light microscopy. Immunohistochemistry methods were used to detect bone sialoprotein (BSP) distribution in Hyp and WT mouse molar tissues, and transmission electron microscopy was used to study similar molar tissues in the non-demineralized state.
Dentin in Hyp mice exhibited mineralization defects by 33 dpc, as expected, but this defect was partially corrected by 96 dpc. In support of our hypothesis, a cementum phenotype was detected using a combination of immunohistochemistry and transmission electron microscopy, which included thinner BSP-positive staining within the cementum, discontinuous mineralization, and a globular appearance compared to WT controls.
Mutations in the phosphate-regulating Phex gene of the Hyp mouse resulted in defective cementum development.
X-linked hypophosphatemia (XLH), or vitamin D-resistant rickets, is a disorder from inactivating mutations in Phex and is the most common non-nutritional form of rickets in developed countries.1 PHEX, an endopeptidase, is expressed in osteoblasts, osteocytes, and odontoblasts, suggesting a role in biomineralization.2 Similar to patients with XLH, the X-linked hypophosphatemic (Hyp) mouse homolog features shortened bones with rickets and osteomalacia.3 Transgenic expression of Phex in the Hyp mouse restored the bone phenotype without affecting phosphate (Pi) metabolism deficiencies.4 Pi wasting in XLH and Hyp mice is associated with increased expression of fibroblast growth factor-23 (Fgf-23), a phosphatonin that increases renal Pi excretion. 5 Hyp mice are also characterized by reduced 1, 25-(OH)2D3 (Vit D3) levels, increased parathyroid hormone levels, and normal calcium levels.4
In addition to skeletal defects, tooth mineralization in XLH and Hyp mice is adversely affected, with thin dentin, widened predentin, and “interglobular” dentin indicating hypomineralization.6–8 Autosomal recessive hypophosphatemic rickets is another phosphate-wasting disorder featuring tooth defects; it results from mutations in dentin matrix protein 1 (Dmp1) and features high Fgf- 23.9 The mouse homolog is the Dmp1 knock-out (KO) mouse.9,10 Periodontal breakdown and altered cementum are reported in Dmp1 KO mice.11
Cementogenesis is sensitive to altered regulation of local Pi and pyrophosphate (PPi); i.e., loss of tissue non-specific alkaline phosphatase (TNAP)12,13 in mice and humans, and lack of progressive ankylosis protein (ANK) and plasma membrane glycoprotein 1/nucleotide pyrophosphatase phosphodiesterase 1 (PC1/NPP1) in mice14 results in decreased and increased cementum, respectively. Dentin seemed to develop normally, whereas acellular cementum was profoundly affected in these cases. These models suggest local changes in Pi/PPi affect cementum development and prompted studies of tooth root formation under conditions of altered systemic Pi. We hypothesized that acellular cementum in the Hyp mouse would be similar to that observed in teeth obtained from the TNAP mutant mouse, i.e., limited or absent.12 We report, as expected, defective dentin mineralization, now quantified and, for the first time, a cementum phenotype (decreased formation of acellular cementum), as well as partial correction of the dentin mineralization defect over time in teeth from Hyp mice.
Hyp mouse heterozygote breeding pairs were provided by Dr. Beate Lanske, Department of Developmental Biology, Harvard School of Dental Medicine, Boston, Massochusetts. Hyp males and females and wild-type (WT) littermates were sacrificed at 23, 27, 33, 45, and 96 days postcoital (dpc); birth was day 19. Animal studies were approved by the University of Washington Institutional Animal Care and Use Committee.
Mandibles were dissected, fixed in Bouin’s fixative overnight, and stored in 70% ethanol. Tissues from 27 dpc and later were demineralized in acetic acid, neutral buffered formalin, and sodium chloride. Tissues were processed and paraffin embedded for the preparation of 5-μm buccal–lingual sections of the mandibular first molar. Standard hematoxylin and eosin (H&E) staining was used for histologic observation by light microscopy.|| Acellular cementum, dentin, and predentin in the first molars were measured at a site 75 μmapical to the cemento-enamel junction (CEJ), which was selected to ensure a region limited to acellular cementum.
All tissue measurements were obtained from three samples and compared using the Student t test, with significance indicated by P <0.05.
Standard immunohistochemistry methods were used to detect the bone sialoprotein (BSP) distribution in 45 dpc Hyp and WT mouse molar tissues.15 Rabbit anti-mouse BSP antibody was a gift from Dr. Renny Franceschi, Department of Periodontics and Oral Medicine, University of Michigan, Ann Arbor, Michigan. Positive staining was developed using a kit¶ per the manufacturer’s instructions. Negative control samples used normal serum in place of antibody, whereas WT acellular cementum samples served as the positive control for BSP staining.
Mandibles from 45 dpc Hyp and WT mice were dehydrated via graded ethanol series, mounted, and cured in room-temperature-cure epoxy.# Molars were ground with 1,500 grit SiC paper at mesial surfaces, revealing the molar interior. Ultrasectioning was performed using a diamond knife** mounted on an ultramicrotome.†† Ultrathin sections were collected onto lacey carbon–coated Cu grids.‡‡ Transmission electron microscopy§§ (TEM) was performed with a tungsten filament at 100 keV.
H&E-stained molar tissues from WT and Hyp mice were examined by light microscopy at developmental stages 23 to 96 dpc. Dentin was dramatically thinner, the zone of predentin was expanded, and pulp chambers were enlarged in Hyp mice as early as 33 dpc (Fig. 1). This dentin phenotype in Hyp mouse molars is consistent with previous descriptions of advanced developmental ages.6 The ratio of predentin/dentin was higher in Hyp molars compared to WT molars at 33 and 45 dpc, indicating a mineralization lag resulting from the Hyp mutation (Fig. 2). This proportional difference between Hyp and WT molars diminished considerably by 96 dpc, although the overall width of dentin in Hyp mice remained thinner (Figs. 1E and 1F; Fig. 2). Also seen was a markedly expanded and osteomalacic appearance of the surrounding alveolar bone in Hyp mice, starting from 33 dpc (Figs. 1A through 1D), which was still apparent at 96 dpc (Figs. 1E and 1F).
In addition to widened predentin in Hyp mice, an uneven mineralization front in dentin was observed at 33 and 45 dpc. This pattern was most apparent at 45 dpc, where areas of the mineralization front appeared globular (Fig. 1D, black arrows), resembling what was called “globular/interglobular dentin” in previous articles.6–8
This study adds new information on dentin defects, quantifying the extent of the defect and reporting a less severe phenotype with age. Histologically, no apparent differences were noted inWT versus Hyp acellular cementum (Fig. 1), cellular cementum (data not shown), or periodontal ligament (PDL). Based on the observation of a marked dentin phenotype at 45 dpc, this time point was selected for further analyses of cementogenesis.
Immunohistochemistry for BSP was used as an unambiguous means of visualizing cementum because BSP is a well-established marker for cementum formation, with negligible expression in dentin and PDL.15 Low magnification showed the expected BSP distribution in the cementum and alveolar bone of WT and Hyp mice (Figs. 3A and 3B). Higher magnification revealed that the width of BSP staining was less in Hyp acellular cementum (Figs. 3C and 3D); i.e., the width of BSP-stained cementum measured 75 μm apically from the CEJ was 4.3 ± 0.5 μm in WT samples and 2.3 ± 0.3 μm in Hyp samples (n = 4).
The difference in cementum width between WT and Hyp mice revealed by BSP staining prompted investigation into cementum structure using TEM on non-demineralized tissues from 45 dpc mice. TEM analysis focused on acellular cementum (Figs. 4A and 4B, boxes). The difference in acellular cementum morphology in WT versus Hyp molars was dramatic (Figs. 4C and 4D). WT cementum was uniform in width and distinguishable from adjacent dentin as a less dense tissue with a clear demarcation at the cementum –dentin junction (Fig. 4C). In contrast, Hyp mouse acellular cementum appeared uneven, with discontinuous globular features (Fig. 4D). Furthermore, in Hyp mice, the transition zone between dentin and cementum was indistinguishable because of comparable contrast in the two tissues (Fig. 4C versus Fig. 4D). Another finding not reported previously for dentin was the uneven mineralization represented by areas of non-uniform contrast noted throughout Hyp dentin (Fig. 4F). More highly mineralized globular dentin was distinguishable by darker regions.
The results from this study are consistent with our hypothesis that a decrease in systemic Pi would lead to a decrease in the formation of acellular cementum. Further, we provide new data on the root dentin phenotype, including quantification of the mineralization defect (widened predentin) and the fact that the mineralization lag is largely resolved by 96 dpc. Although histologic sections did not point to abnormalities in Hyp mouse cementum, immunohistochemistry for BSP revealed an altered pattern of localization of BSP in Hyp mouse cementum compared to WT at 45 dpc. TEM examination showed that Hyp mouse cementum featured discontinuous mineralization and a globular appearance. The area of cellular cementum in murine teeth is not uniform, which impedes the preparation of consistent histologic samples; thus, we chose to focus on acellular cementum.
The TEM appearance of Hyp mouse acellular cementum is evocative of the dentin phenotypes of hypophosphatemic mice described here and elsewhere, 6,16–18 e.g., discontinuous mineralized pockets indicating a dentin hypomineralization phenotype, which were referred to as “interglobular” and considered a hallmark for the defective dentin associated with Pi and Vit D3 deficiency.19 These findings suggest that pathologic reductions in serum Pi and possibly alterations in other factors may have induced a developmental cementum phenotype, although not as striking as that reported from the loss of TNAP, ANK, or PC-1/NPP1 in mice. Loss of TNAP function results in disrupted and decreased cementum formation, loss of PDL attachment, and exfoliation of teeth.12,13,20 Mice with loss of function mutations in Ank or ectonucleotide pyrophosphatase phosphodiesterase 1 (Enpp1) genes exhibit a profound increase in cementum formation.14 These proteins regulate microenvironmental Pi/PPi ratios, whose effects may differ from systemic Pi changes. It is unclear if the Hyp cementum aberration is a direct or indirect effect of serum Pi levels; other mineralization-regulating factors must be accounted for to fully understand the cause(s) of the observed cementum phenotype, which is an area of active investigation in our laboratory.
Complex interactions govern the regulation of serum Pi, making it difficult to identify proximate and compounding causes of Hyp mineralization defects. Hyp mice exhibit dramatic increases in FGF23 in osteocytes, and FGF23 is higher in developing ameloblasts and odontoblasts.21,22 Overexpression of FGF23 was shown to be responsible for much of the pathology in the Hyp mouse.3,23 Osteocalcin, a mineralization regulator, was increased in odontoblasts,24 and the expression of Pi transporter NaPi2b was reduced in Hyp mice.25 Small integrin-binding ligand N-linked glycoprotein family proteins, matrix extracellular phosphoglycoprotein (MEPE), and dentin matrix protein 1 (DMP1) contain acidic serine-aspartate–rich MEPE–associated motifs (ASARM peptides), cleavage products that are mineralization inhibitors. ASARM is reported to be increased in Hyp mice as a result of increased cleavage of MEPE and DMP1. Increases in ASARM have been linked to low bone mineral density.26 Although ASARM levels have not been measured in cementum, this peptide may be present in increased concentrations in the cementum of Hyp mice and may contribute to disrupted mineralization. In vitro experiments to understand the mechanisms for the observed cementum and dentin phenotypes due to local and systemic phosphate level alterations are being conducted in our laboratory.
The literature describes dentin/pulp-related disorders in patients with XLH, including increased predentin, interglobular dentin, and enlarged pulp chambers,17,27–29 similar to reports of the Hyp mouse tooth phenotype described here and previously.6–8,29–31 Further, periodontal involvement, including cementum abnormalities, periodontal abscesses, and loss of lamina dura indicating alveolar bone disruption, has been noted in the teeth of patients with XLH,17,28 and now we report a tooth root cementum phenotype in Hyp mice.
In conjunction with previous reports, our findings suggest that hypophosphatemic individuals are susceptible to periodontal disease and should be monitored carefully to detect early signs of diseased periodontium and to administer timely treatment. Administration of a high-Pi diet normalized serum Pi and rescued bone mineralization in Hyp mice.32,33 Additionally, correction of the dentin phenotype in the permanent teeth of patients with XLH was achieved by modulation of Pi and Vit D3,19 suggesting that such interventions may be successful. Although further research is needed, results from ongoing studies strongly suggest that appropriate levels of Pi and PPi and related regulatory factors are critical for the maintenance of healthy teeth as well as bones.
The authors thank Dr. Bruce Rutherford, Department of Oral Biology, University of Washington, for his insightful critical comments. The electron microscopy work was supported by the Shared Experimental Facilities of Genetically Engineered Materials Science and Engineering Center, University of Washington, a National Science Foundation–funded Materials Research Science and Engineering Center program. This research was supported by the following sources: National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland (grant RO1 DE15109 to MJS) and National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (grant RO1 073944 to BL). The authors report no conflicts of interest related to this study.
¶Vector Laboratories, Burlingame, CA.
#Allied High Tech Products, Rancho Dominguez, CA.
**Diatome, Hatfield, PA.
††MT 6000-XL, RMC, Tucson, AZ.
‡‡Ted Pella, Redding, CA.
§§Philips EM420, FEI, Hillsboro, OR.