Amelogenin Exon4 Forms a Novel miRNA That Directs Ameloblast and Osteoblast Differentiation.

Amelogenins constitute the major portion of secretory enamel matrix proteins and are known to be highly alternative spliced. Of all the alternatively spliced forms of amelogenins, exon4 is most commonly spliced out. Our analyses of the exon4 sequence led us to hypothesize that when spliced out, exon4 may generate a novel mature miRNA. To explore this possibility, we used in vivo mouse models (wild-type and Amel knockout mice) and in vitro cell culture to investigate the presence and function of a mature miRNA derived from exon4 (miR-exon4). When ameloblast-like cells (LS8) were transfected with an amelogenin minigene to increase amelogenin synthesis, the transfected cells synthesized miR-exon4. Introduction of a mutation in the conserved CNNC sequence required for primary miRNA recognition, downstream of the mature miR-exon4 sequence, resulted in a significantly reduced production of miR-exon4 in the transfected cells. In vivo, miR-exon4 was most highly amplified from wild-type mouse enamel organs at the secretory stage. In Amel knockout mice, an in vivo model for reduced amelogenin synthesis, we found reduced miR-exon4, with no changes in expression of enamel matrix-related genes. However, expression of Runx2 and its downstream genes Odam and Amtn were significantly downregulated. Transfection of miR-exon4 mimic to the LS8 cells also significantly upregulated Runx2. The mature miR-exon4 as well as Runx2 was also present in mouse osteoblasts with no apparent change in expression level between wild-type and Amel knockout mice. However, transfecting miR-exon4 inhibitor to the MC3T3-E1 osteoblastic cells resulted in a significant downregulation of Runx2 expression. These data indicate that when exon4 is spliced out, as occurs most of the time during alternative splicing of amelogenin pre-mRNA, a novel mature miRNA is generated from exon4. This miR-exon4 may contribute to the differentiation of ameloblasts and osteoblasts through regulation of Runx2 expression.

Exon4 amelogenin transcripts in enamel biomineralization.

Amelogenins are proteins formed by alternative splicing of the amelogenin gene, and are essential for tooth enamel formation. However, the unique functions of various alternatively spliced amelogenins in enamel formation are not well understood. In this study, we determined the spatiotemporal location of amelogenins derived from transcripts containing exon4 (AMG+4) in the enamel matrix, and the relative binding of recombinant AMG+4 to hydroxyapatite (HAP). Immunohistochemistry and mass spectrometry analyses showed that AMG+4 proteins were secreted into the enamel matrix at the early maturation stage. A stage-specific increase in the synthesis of AMG+4 was further supported by our observation that in mice overexpressing leucine-rich amelogenin peptide (TgLRAP), in which ameloblasts differentiate earlier, AMG+4 transcripts were also upregulated earlier. In vitro binding studies, supported by in silico modeling of protein binding to calcium and phosphate, showed that more recombinant AMG+4 bound to hydroxyapatite (HAP) as compared with recombinant AMG-4. The temporal and spatial localization of amelogenins containing exon4 peptide, and their functional differences in HAP binding, suggests that the unique properties of amelogenins containing exon4 cause a specific enhancement of biomineralization related to stabilization of early-formed HAP at the maturation stage.

Effects of fluoride on protein secretion and removal during enamel development in the rat.

Enamel maturation consists of a loss of the early secreted matrix proteins and an increase in mineralization. This study investigated the changes in enamel proteins of the rat incisor, caused by the ingestion of fluoride at various stages of enamel formation. Rats were given 0, 10, 25, 50, or 100 ppm fluoride in drinking water for five weeks. Changes in the protein composition of the secretory, early-maturation, and late-maturation enamel were investigated by means of gel filtration chromatography and polyacrylamide gel electrophoresis. No differences were found between fluorosed and control enamel proteins in secretory enamel. In fluorosed early-maturation enamel, amelogenins were retained in larger quantities than in control enamel in animals ingesting 25 ppm fluoride or greater. At the late-maturation stage of enamel formation, only enamel from animals ingesting 100 ppm fluoride in drinking water contained more protein, when compared with control enamel. This study suggests that fluoride ingestion levels resulting in enamel fluorosis inhibit the mechanisms involved in the removal of proteins during enamel maturation.

Mechanism and timing of fluoride effects on developing enamel.

Fluoride appears to specifically interact with mineralizing tissues, causing an alteration of the mineralization process. In enamel, fluorosis results in a subsurface hypomineralization. This hypomineralized enamel appears to be directly related to a delay in the removal of amelogenins at the early-maturation stage of enamel formation. The specific cause for this delay is not known, although existing evidence points to reduced proteolytic activity of proteinases that hydrolyze amelogenin. This delay in hydrolysis of amelogenins could be due to a direct effect of fluoride on proteinase secretion or proteolytic activity, or to a reduced effectiveness of the proteinase due to other changes in the protein or mineral of the fluorosed enamel matrix. The formation of dental fluorosis is highly dependent on the dose, duration, and timing of fluoride exposure. The early-maturation stage of enamel formation appears to be particularly sensitive to the effects of fluoride on enamel formation. Although the risk of enamel fluorosis is minimal with exposure only during the secretory stage, this risk is greatest when exposure occurs in both secretory and maturation stages of enamel formation. The risk of fluorosis appears to be best related to the total cumulative fluoride exposure to the developing dentition.

Dental fluorosis: its use as a biomarker.

Several epidemiological studies, beginning with those of Dean and co-workers in the 1940's, clearly demonstrate the relationship between dental fluorosis in humans and the level of fluoride in water supplies. These studies and others have shown that, in a population, there is a direct relationship among the degree of enamel fluorosis, plasma and bone fluoride levels, and the concentration of fluoride in drinking water. However, dental fluorosis is a reflection of fluoride exposure only during the time of enamel formation, somewhat limiting its use as a biomarker. In addition, the degree of fluorosis is dependent not only on the total fluoride dose, but also on the timing and duration of fluoride exposure. At the level of an individual response to fluoride exposure, factors such as body weight, activity level, nutritional factors, and the rate of skeletal growth and remodeling are also important. These variables, along with an individual variability in response to similar doses of fluoride, indicate that enamel fluorosis cannot be used as a biological marker of the level of fluoride exposure for an individual.

Purification and sequencing of a 21 kDa and 25 kDa bovine enamel metalloproteinase.

The developing enamel matrix contains metalloproteinases that are presumed to have a role in hydrolysis of enamel matrix proteins. Determination of the identity and function of these proteinases requires further information such as their amino acid composition and sequence. In this study, we purified the 21 kDa and 25 kDa matrix metalloproteinase from secretory stage bovine enamel matrix. After extraction, these proteinases were further purified by sequential separation by ion exchange, Con A affinity chromatography, and reversed phase HPLC. The proteinases were separated by SDS PAGE, transferred to a PVDF membrane and the N-terminus was sequenced. The N-terminal sequences of both proteinases were the same, and showed homology to the porcine enamelysin (MMP 20) cDNA sequence. A cDNA for bovine MMP 20 was isolated from a bovine enamel organ cDNA library using a probe generated by PCR amplification. The coding regions of bovine and porcine MMP 20 cDNAs were highly homologous and contained the same regions of predicted amino acid sequence homology with the proteinase N-terminal sequences. These results suggest that the 21 kDa and 25 kDa enamel matrix metalloproteinases are cleavage products of the initially secreted MMP 20, and that the sequence for MMP 20 is conserved across species.

Characterization of fluorosed human enamel by color reflectance, ultrastructure, and elemental composition.

Mature fluorosed human enamel has been described as a subsurface enamel hypomineralization, with porosity increasing relative to the degree of fluorosis. The purpose of the current study was to quantitatively measure the color of the fluorosed enamel by light reflectance, and to further characterize the enamel by scanning electron microscopy. Teeth with varying degrees of fluorosis were obtained and divided in groups of mild, moderate and severe fluorosis using Dean's index for fluorosis. The color of the labial enamel surface was measured using a Minolta Chroma Meter CR241 (Minolta, Ramsey, N.J., USA). The teeth were further characterized for elemental composition using an energy-dispersive spectrometer, and imaged in both secondary and backscattered electron modes. The results of this study showed that the moderately and severely fluorosed enamel contained an uneven distribution of areas which were more electron-absorbent with a relatively increased carbon content. The changes in the physical characteristics of the teeth could be quantitated by measurements of light reflectance. The color of the teeth was significantly different between groups, with all groups significantly different than normal.

Primary culture and characterization of enamel organ epithelial cells.

The cells of the enamel organ are programmed by signals such as growth factors and extracellular matrix components to differentiate and form dental enamel. To study how the enamel organ epithelial cells control enamel development, we have begun to characterize a primary porcine enamel organ epithelial cell culture system. The unerupted molars of 3 month old pigs were isolated, the cells were digested into a single cell suspension and grown in media either with or without serum. Expression of amelogenin and ameloblastin mRNA was monitored by RT PCR, and protein secretion was identified by immunohistochemistry. Cells grown in MEM formed a mixed cell population of epithelial- and fibroblast-like cells which grew past confluence, formed nodules, mineralized, and expressed low levels of amelogenin and ameloblastin protein. In LHC-9 media, which is selective for epithelial cells, the cells did not grow past confluence but secreted amelogenin and ameloblastin proteins more strongly. Cell viability was maintained in both serum-free and serum-containing media. However, in the serum-free media, cell proliferation proceeded slowly. Although cells grown in MEM mineralized, the mixed cell population may make studies of specific ameloblast-like cells more difficult. However, cells grown in a culture media selective for epithelial cells will require modifications such as cell immortalization to allow long term studies of cell regulation and interaction. In summary, we have established an enamel organ epithelial cell culture system which will enable us to study the role of ameloblasts in enamel matrix formation, ameloblast regulation, as well as cell-matrix interactions. Selection of specific culture conditions will depend on the questions being addressed in individual studies.