Rescue of odontogenesis in Dmp1-deficient mice by targeted re-expression of DMP1 reveals roles for DMP1 in early odontogenesis and dentin apposition in vivo

Abstract

Dentin matrix protein 1 (DMP1) is expressed in both pulp and odontoblast cells and deletion of the Dmp1 gene leads to defects in odontogenesis and mineralization. The goals of this study were to examine how DMP1 controls dentin mineralization and odontogenesis in vivo. Fluorochrome labeling of dentin in Dmp1-null mice showed a diffuse labeling pattern with a 3-fold reduction in dentin appositional rate compared to controls. Deletion of DMP1 was also associated with abnormalities in the dentinal tubule system and delayed formation of the third molar. Unlike the mineralization defect in Vitamin D receptor-null mice, the mineralization defect in Dmp1-null mice was not rescued by a high calcium and phosphate diet, suggesting a different effect of DMP1 on mineralization. Re-expression of Dmp1 in early and late odontoblasts under control of the Col1a1 promoter rescued the defects in mineralization as well as the defects in the dentinal tubules and third molar development. In contrast, re-expression of Dmp1 in mature odontoblasts, using the Dspp promoter, produced only a partial rescue of the mineralization defects. These data suggest that DMP1 is a key regulator of odontoblast differentiation, formation of the dentin tubular system and mineralization and its expression is required in both early and late odontoblasts for normal odontogenesis to proceed.

Fig. 1. Dentin mineral apposition rate is reduced in Dmp1-null mice

Panel A, fluorochrome labeled sections of the lower first molar from heterozygous control (HET, left, and inset) and Dmp1 knockout mice (KO, right, and inset) reveal a reduction of the dentin mineralization rate in the knockout mice . Bar = 10μm. Panel B, quantitative analysis shows a significant difference in the dentin appositional rate between the control and the KO in the first molar (data are mean ± SEM from n=4 replicates, *** = p< 0.001 by Student’s t-test)

Fig. 2. Abnormal morphology of the dentinal tubules in Dmp1-null mice

Panel A, shows resin-casted SEM images of the dentinal tubules in heterozygous control (HET, upper panel) and Dmp1 knockout mice (KO, middle panel). Note that the tubules in the KO mice appear more disorganized and show less branching that in controls. Bar = 10μM. Panel B, shows a higher magnification view in which it is clear that the dentin tubules in Dmp1-null mice (KO, lower right panel) are flatter, wider and less branched compared to controls (HET, lower left panel). Bar = 5μM.

Fig. 3. The Role of DMP1 in odontogenesis appears to be local and not systemic

Panel A, representative radiographs demonstrate a similar thin dentin wall (arrowhead) in the molars from VDR KO (middle) and Dmp1 KO mice (right) compared to wild type controls (WT, left). Panel B, the procion red dye, which diffuses through the dentinal tubule system, shows no gross abnormalities in the appearance of dentin tubules in the VDR KO mice (middle). In contrast, dramatic abnormalities in the dentin tubular morphology are observed in the Dmp1 KO mice (right) compared to controls (left). Bar = 10μm Panel C, a high calcium diet, fed to the mice from weaning until 2 months of age, fails to rescue the reduced dentin thickness (arrow) observed in the DMP1-KO mice (right) compared to controls (left).

Fig. 4. Rescue strategy and targeted re-expression of transgenic DMP1 in Dmp1 KO Mice

Panel A, shows a schematic diagram depicting the key steps in odontoblast differentiation. The arrows corresponding to the Col1a1 and Dspp promoters indicate when these promoters are active during the odontogenesis pathway. Transgenic mice were generated in which the coding region of the murine Dmp1 cDNA was expressed under control of the Col1a1 promoter (for early expression in pulp cells and odontoblasts) or under control of the Dspp promoter (for expression in mature odontoblasts). These transgenic mice were crossed onto the Dmp1-null background for re-expression of DMP1 at specific stages during odontogenesis. Panels B–E show in situ hybridization (red staining with green nuclear counterstain) and immunostaining (brown staining with green nuclear counterstain) for DMP1 in the lower 3rd molar in 10 day old mice. In Dmp1-KO mice (B), no DMP1 was detected, either at the mRNA (left) or protein (right) level. In control mice (C), endogenous expression of DMP1 mRNA (left) and protein (right) was observed in the early odontoblasts (black arrows). In Dmp1-KO mice expressing the Col 1a1-DMP1 transgene (D), strong expression of DMP1 RNA and protein was observed in both early (arrowhead) and mature (arrow) odontoblasts. Expression was also seen in the pulp cells (inset). In Dmp1-KO mice expressing the Dspp-DMP1 transgene (E) expression of DMP1 was detected in mature odontoblasts (arrow). Unlike the Col1a1 transgene, there was no expression of the Dspp-DMP1 transgene in the pulp cells (inset). (Bar = 25μm, inset bar = 5μm).

Fig. 5. Full or partial rescue of the tooth phenotype in Dmp1-KO mice by re-expression of transgenic DMP1 under control of the 3.6 kb Col1a1 (Early Rescue) or Dspp (Late Rescue) promoters

Panel A shows radiographs of the lower first molar from one month old control (HET) and Dmp1 KO (KO) mice as well as KO mice re-expressing DMP1 under control of the Col1a1 promoter (early rescue) and Dspp promoter (late rescue). Note the reduced thickness of the dentin wall (arrow) in the KO mice that is rescued by the Col1a1-DMP1 (early) transgene but not by the Dspp-DMP1 (late) transgene. Panel B shows H & E staining of the first molars from the HET control (left), Dmp1 KO (left center), early rescue (right center), and late rescue mice (right). Enlarged images of the dentin and predentin are shown below. Bar = 50μm. Panel C shows quantitative analysis of the dentin thickness and dentin/predentin ratio. Note that the Col 1a1-DMP1 transgene completely rescues the dentin phenotype in these mice, whereas the Dspp-DMP1 transgene only restores the predentin thickness (data are mean ±SEM from n=4 replicates, significance was determined by ANOVA, followed by Tukey-Kramer multiple comparisons, *= p<0.05, **=p<0.01). Panel D shows TEM images of dentinal tubules after immunogold staining for DMP1 using 14 nm gold labeled antibodies. Note the absence of labeling in the Dmp1-null mice (left), together with the irregular surface of the tubule wall. Re-expression of DMP1 under control of the Col1a1 promoter (Early RES) is shown on the right. Note the accumulation of gold labeling on the tubular wall, which also has a smooth morphology.

Fig. 6. Reduced osterix expression in the tooth in Dmp1-KO mice and its restoration by re-expression of DMP1 under control of the Col1a1 promoter

Panel A. In situ hybridization in the lower first molar of newborn (left) and 10-day-old-mice shows that in control (HET) mice, the transcription factor, osterix, is expressed in both pulp cells (P) and odontoblasts (Od) (red staining indicates RNA expression and nuclei are counterstained in green). Panel B. At this time, in the Dmp1 KO mice (KO) osterix was undetectable in the newborn molar (left) with a low level in 10-day-old molar (right). Panel C. Re-expression of DMP1 under control of the Col 1a1 promoter in mice lacking endogenous DMP1 (Early RES) restored osterix expression in both pulp and odontoblast cells from newborns (left) and 10-day-old molars (right). The insert showed an enlargement of the 10-day-old root dentin area. The expression level in the rescued mice was actually higher than in control animals. Bar = 100μm.

Fig. 7. Rescue of delayed development of the third molars in Dmp1-KO mice by re-expression of DMP1 under control of the 3.6 kb Col1a1 but not the Dspp promoter

Panel A shows radiographs of the mandible of ten day old mice. Note that in the control mice (HET), the third molar is evident on the radiograph. However, in Dmp1-KO mice (KO), the development of the third molar is delayed. Re-expression of DMP1 under control of the Col1a1 transgene (Early RES) rescues the development of the third molar but the Dspp-DMP1 transgene fails to rescue third molar development. Panel B shows H&E staining of sections through the mandible, confirming the delayed development of the third molar in Dmp1-KO mice and its rescue by the Col1a1-DMP1 transgene (Early RES) but not the Dspp-DMP1 transgene (Late RES) (Bar = 200μm). Panel C shows radiographs of the mandible from one month old mice. Note that in these one month old animals, the third molar appears similar in control, Dmp1-KO and the early and late rescued animals suggesting that in Dmp1-KO mice there is only a delay, but not a failure in development of the third molar.