| Hypophosphatemic rickets (HR) is characterized by hypomineralization of bone, rickets and osteomalacia due to abnormal Pi reabsorption in renal tubules. The clinical features of HR include hypophosphatemia, phosphaturia, rickets and osteomalacia, with an incidence of 1 in 20,000 newborns. Hypophosphatemic rickets can be inherited in autosomal dominant, autosomal recessive, or X-linked pattern. X-linked hypophosphatemic rickets (XLHR, MIM 307800) is caused by mutation in the PHEX gene (phosphate regulating endopeptidase homolog, X-linked). Autosomal dominant hypophosphatemic rickets (ADHR, MIM 193100) is caused by mutation in the FGF23 gene (Fibroblast Growth Factor 23). Autosomal recessive hypophosphatemic rickets (ARHR, MIM 241520 and 613312) is caused by mutation in the DMP1 gene (Dentin Matrix Protein 1) or in the ENPP1 gene (ectonucleotide pyrophosphatase/phosphodiesterase 1). Hereditary hypophosphatemic rickets with hypercalciuria (HHRH, MIM 241530) is caused by mutation in the SLC34A3 gene (solute carrier family 34, member 3).DMP1 gene encodes an acidic non-collagen extracellular matrix protein, a member of the SIBLING (Small Integrin-Binding LIgand N-linked Glycoprotein) family whose members share similar features in both biological properties and genomic structure. DMP1 gene was mapped to human chromosome 4q21-22 and mouse chromosome 5q21, respectively. DMP1 is highly expressed in bone and dentin, and lower expressed in non-mineralized tissues. DMP1 is cleaved into the 37 kDa N-terminal and the 57 kDa C-terminal fragments, but it is not clear which fragment is functionally critical for bone and tooth formation both in vivo and in vitro.Dmpl null embryos and newborns display no apparent gross abnormal phenotype, suggesting that Dmpl is not essential for early mouse skeletal or dental development. Dmpl null mice postnatally developed a profound tooth phenotype characterized by partial failure of maturation of predentin into dentin, enlarged pulp chambers and root canals, increased width of predentin zone with reduced dentin wall, and hypomineralization. Dmpl null mice also showed absence or delayed development of the third molar and periodontal breakdown, including porous alveolar and defective cementum. Dmpl null mice manifested decreased bone mineralization and osteomalacia, as well as rickets, which was characterized by shorter and wider vertebrae and long bones, delayed and malformed secondary ossification centers, irregular and highly expanded growth plate (due to increased cell proliferation in proliferating zone and reduced apoptosis in the hypertrophic zone) and increased epiphyses area (due to impaired blood vessel invasion). The rickets and osteomalacia phenotypes of Dmpl null mice were associated with isolated renal phosphate-wasting, elevated Fgf23 levels and normal calciuria.In the patients of one of our previously reported ARHR kindreds, a homozygous deletion of nucleotides 1484-1490 in DMP1 gene was detected. The deletion of the 7 nucleotides resulted in a frameshift that replaced the conserved C-terminal 18 residues with 33 novel residues. In the present study, we characterized the tooth phenotypes of the patients, and analyzed the function of mutant protein in vitro. To further understand the molecular and pathophysiologic mechanism of this DMP1 mutation, we generated and characterized a mouse model with this specific mutation.Firstly, we examined the tooth phenotype of the patients with DMP1 C-terminal mutation, and found that the patients displayed obvious abnormalities during tooth development:the radiographs from the patient teeth (both primary and permanent teeth) showed significantly enlarged pulps and root canals with extremely thin dentin and loss of enamel. Secondly, in vitro function assays suggested that the mutant protein maintained partial function of DMP1:ARHR patients with DMP1 mutation and Dmpl null mice share some common features, such as hypophosphatemia. However, minor differences exist between the mouse model and the human ARHR patients. These differences could be due to species-specificity of human versus mouse, or that the mutant DMP1 in human may maintain partial function of DMP1. To determine the function of the C-terminal mutant DMP1 and the mechanism in which it causes ARHR, we examined the cleavage and secretion of the C-terminal mutant DMP1, and its function on ERK phosphorylation. The C-terminal mutant DMP1 can be secreted and cleaved normally and maintained partial function on ERK phosphorylation. These data suggested that the mutant protein did not completely lose its function.Thirdly, we genetated a mouse model harboring the human DMP1 1481-1490 deletion mutation. To study the mechanism in which DMP1 C-terminal mutation cause ARHR, we generated transgenic mice harboring the full-length Dmpl cDNA with the same deletion mutation as observed in ARHR patients under control of the 3.6 kb Collal promoter. Eleven founder mice were got and 3 independent lines were used for expression analyses by RT-PCR and in situ hybridization. The expression pattern of the transgene in bone and teeth is the same as endogenous Dmpl gene, and the expression level is higher than endogenous Dmpl gene. Radiographs showed that the transgenic mice expressing the mutant Dmpl gene did not show an apparent bone and teeth phenotype compared to the WT control mice. Then, these transgenic lines were crossed to Dmpl null mice, and the homozygous mice with Dmpl mutation but without endogenous Dmpl gene were obtained.Fourthly, the Dmpl mutant mice recaptured the human ARHR patients' phenotypes, including bone, tooth and serum markers:The phenotypes of Dmpl mutant mice were characterized and compared with those of ARHR patients and Dmpl null mice using several assays including histology, radiography, Micro-CT, SEM, calcein/alizarin red double labeling, and serum biochemistry.1) Bone phenotype:As showed in radiographs of the whole bodies and hind limbs, Dmpl mutant mice displayed rickets phenotype, including shorter femur and tibia, expanded metaphyses, enlarged growth plates and malformed bone shape. Quantitative analyses of tibia length confirmed that the long bones in Dmpl mutant mice are shorter compared to the control mice, but are not as severe as Dmpl null mice. Histological data of Safranin-O staining further confirmed the rickets phenotype such as striking expansion of the growth plate, including resting zone, proliferation zone and hypertrophic zone with the last one expanded most. Goldner assay showed abundant osteoid in the mineral of cortical bone in Dmpl mutant mice, a significant sign of hypo-mineralization and osteomalacia phenotype. Double fluorochrome labeling assay showed slow rate and disruption of bone mineralization in Dmpl mutant mice. Dmpl mutant mice also showed abnormalities in the osteocyte lacuna-canalicular system by SEM, such as enlarged size and buckled surface of the osteocytes, and fewer and shorter synapses. These phenotypes of Dmpl mutant mice were consistent with observations in samples from ARHR patients, but milder than those in Dmpl null mice, suggesting that the Dmpl mutant mice can recapture the human ARHR bone phenotype.2) Tooth phenotype:Dmpl mutant mice showed thin dentin and enlarged pulp cavities and root canals with radiograph and Micro-CT analyses, but the phenotype is milder than that in Dmpl null mice. Histological data of H&E staining further confirmed the dentin phenotype in Dmpl mutant mice:the expanded predentin and the thin dentin. The double labeling assay also showed reduction of the dentin formation rate and diffused labeling lines in Dmpl mutant mice. Dmpl mutant mice also showed defects in condyle formation. However, all these analyses as well as backscattered SEM results did not show any enamel defects in Dmpl mutant mice, which is different from the enamel defects in ARHR patients. These data suggested that the Dmpl mutant mice can recapture the human ARHR tooth phenotype, and the enamel defects in ARHR patients may be due to reasons other than DMP1 mutation.3) Serum biochemistry:we also showed a mild increase of serum Fgf23 in the Dmpl mutant mice compared to the control. Similarly, the decrease in serum Pi level was also mild in all three age groups tested. The serum calcium level is largely unchanged in the Dmpl mutant mice. These data of Dmpl mutant mice were consistent with those in ARHR patients, but milder than those of Dmpl null mice.In summary, Dmpl mutant mice recaptured the phenotype of the human ARHR patients in all 3 aspects mentioned above. However, the phenotype of Dmpl mutant mice is milder that that of Dmpl null mice, suggesting that the mutant DMP1 molecule maintained a partial function of this protein.Successful generation and characterization of the DMP1 C-terminal mutant mice model will provide a foundation for studies on the mechanism in which DMP1 mutation causes ARHR, a powerful tool for studies of clinical therapy of this disease, and an important clue for the studies on other mutations of DMP1 gene. |
PDF Full Text Request