|Home | About | Journals | Submit | Contact Us | Français|
The molecular mechanisms that underlie dental fluorosis are poorly understood. The retention of enamel proteins hallmarking fluorotic enamel may result from impaired hydrolysis and/or removal of enamel proteins. Previous studies have suggested that partial inhibition of Mmp20 expression is involved in the etiology of dental fluorosis. Here we ask if mice expressing only one functional Mmp20 allele are more susceptible to fluorosis. We demonstrate that Mmp20+/− mice express approximately half the amount of MMP20 as do wild-type mice. The Mmp20 heterozygous mice have normal-appearing enamel, with Vickers microhardness values similar to those of wild-type control enamel. Therefore, reduced MMP20 expression is not solely responsible for dental fluorosis. With 50-ppm-fluoride (F−) treatment ad libitum, the Mmp20+/− mice had F− tissue levels similar to those of Mmp20+/+ mice. No significant difference in enamel hardness was observed between the F−-treated heterozygous and wild-type mice. Interestingly, we did find a small but significant difference in quantitative fluorescence between these two groups, which may be attributable to slightly higher protein content in the Mmp20+/− mouse enamel. We conclude that MMP20 plays a nominal role in dental enamel fluorosis.
Matrix metalloproteinase-20 (MMP20; enamelysin) is a predominantly tooth-specific MMP (Turk et al., 2006) that is essential for proper dental enamel formation. Homozygous mutation of human MMP20 causes the non-syndromic enamel malformation termed amelogenesis imperfecta (Kim et al., 2005; Ozdemir et al., 2005; Papagerakis et al., 2008; Lee et al., 2010). Mice harboring targeted disruption of both Mmp20 alleles have severely impaired enamel matrix protein processing, leading to hypoplastic enamel with a decreased mineral content, decreased enamel hardness, and malformed enamel that fractures from the dentin surface (Caterina et al., 2002; Bartlett et al., 2004).
Proteolytic enzyme activity in dental enamel was demonstrated almost four decades ago (Suga, 1970). Rat tooth sections at the secretory stage of development were placed in a gelatin containing photographic emulsion, and proteolytic activity was observed by the presence of cleared areas where the gelatin had been digested. Hydrolysis of gelatin by the enamel layer was completely inhibited by the presence of 5 mM NaF (Suga, 1970). Thus, concurrent with the first observation of proteolytic activity within secretory-stage enamel, the first investigation of the effect of fluoride (F−) on this activity was also assessed. Subsequently, several studies have investigated fluoride’s action on enamel matrix proteolytic activity. Those studies have shown that F− inhibits maturation-stage protease activity (DenBesten and Heffernan, 1989), inhibits MMP20 activity (DenBesten et al., 2002), and inhibits Mmp20 transcription (Zhanget al., 2006, 2007). Conversely, one study demonstrated that neither proteolytic activity from the secretory stage nor that from the maturation stage was inhibited by up to 10 mM NaF exposure (Gerlach et al., 2000), and we performed enzyme kinetic analysis to demonstrate that concentrations of up to 10 mM F− did not inhibit recombinant MMP20 activity (Tye et al., 2010). We also previously demonstrated, by use of qPCR on rat secretory-stage incisor enamel organs, that Mmp20 expression was not affected after 6 wks of exposure to 0, 50, 100, or 150 ppm F− as NaF in drinking water (Sharma et al., 2010).
Here, we applied a genetic approach to assess whether reduced Mmp20 expression levels contribute to the etiology of dental fluorosis. We compared the level of fluorosis observed in F−-treated wild-type mice with that observed in F−-treated Mmp20+/− mice. We reasoned that if Mmp20 expression plays a substantial role in dental fluorosis, Mmp20+/− mice would be more susceptible to the adverse effects of F− than would wild-type mice. Prior experimental approaches were correlative in nature by asking if reduced MMP20 expression or activity correlated with severity of dental fluorosis. However, in this in vivo approach, we directly assess if reduced Mmp20 expression leads to increased fluorosis.
All handling, care, and usage of animals were approved by The Forsyth Institute. Mice were housed in an AAALAC-approved facility. Six-week-old C57BL/6 wild-type and backcrossed Mmp20+/− mice were divided into 4 groups (n = 5) and received either 0 ppm or 50 ppm F− in their drinking water for 6 wks. Fluoride content of drinking water was measured with a fluoride ion-selective electrode. Mouse chow contained an average of 1.57 ppm F−.
Quantitative real-time RT-PCR (qPCR) of first mandibular molar enamel organ RNA extracted from 7-day-old mice (6 wild-type, 9 Mmp20+/−, and 5 Mmp20−/− mice) was used to determine relative expression levels of Mmp20 as a function of the stably expressed internal reference control gene Eef1α1, as previously described (Pfaffl, 2001; Kubota et al., 2005). Primers were: Mmp20 forward (5′-GAGACACACATTTCGACAATGCTGAG-3′), Mmp20 reverse (5′-GTTGGGTACATCAGTGCTGATGGAT-3′); and Eef1α1 forward (5′-ATTCCGGCAAGTCCACCACAA-3′, Eef1α1 reverse 5′-CATCTCAGCAGCCTCCTTCTCAAA-3′).
For MMP20 protein content analysis, hard tissues from mandibular first molars from 6 mice (n = 3/genotype) were incubated in 1 mL of 0.5% formic acid at 4°C overnight, centrifuged briefly to remove residual insoluble material, transferred to an Ultracel 3K filter (3-kDa cut-off, Millipore, Billerica, MA, USA), and centrifuged. The enamel protein on the filter was eluted into sample buffer. Approximately 40 μg of protein was run on SDS-PAGE and transferred to a membrane, and MMP20 bands were visualized by Western blotting with an antiserum to MMP20 (Abcam, Cambridge, MA, USA; Ab39037, 1:2500 in TBST).
Plasma and bone ash F− concentrations were determined after overnight hexamethyldisiloxane (HMDS)-facilitated diffusion (Taves, 1968; Whitford and Reynolds, 1979). Urine was collected from anesthetized mice via bladder puncture prior to their death, and fluoride concentration was determined with a F− ion-selective electrode. Incisors were collected for assessment of dental fluorosis and microhardness values.
Two blinded examiners performed independent assessments of dental fluorosis status for each animal (inter-examiner Kappa, 0.850; p < 0.001). The determination of dental fluorosis was made clinically over the entire lower incisor tooth surfaces according to a modified Thylstrup-Fejerskov (TF) index (Thylstrup and Fejerskov, 1978; Fejerskov et al., 1994; Everett et al., 2002).
A quantitative fluorescence (QF) system was previously devised to evaluate the severity of fluorosis in mice (Everettet al., 2002). A Nikon epifluorescence microscope equipped with a Chroma Gold 11006v2 set cube (exciter D360/40x, dichroic 400DCLP, and emitter E515LPv2) was used to capture fluorescent images of teeth. The fluorescence was converted to grayscale values, and intensities were analyzed with Image J software version 1.33u (http://rsb.info/nih.gov/ij/).
Erupted portions of mandibular incisors from wild-type and Mmp20+/− littermate mice were washed and dehydrated with graded alcohol and acetone. Incisors were embedded sagittally in hard-formulation epoxy embedding medium (EpoFix, EMS, Hatfield, PA, USA). Samples were ground and polished to 0.25 μm with diamond suspensions (EMS). The polished samples were tested for enamel microhardness on an M 400 HI testing machine (Leco, St. Joseph, MI, USA). Testing was performed with a load of 25 g for 5 sec with a Vickers tip. Twenty indentations per sample were performed on 3 teeth per group and averaged.
We evaluated statistical significance by performing a two-way analysis of variance (ANOVA) with Bonferroni’s post hoc test, using GraphPad Prism 5 software (GraphPad, La Jolla, CA, USA). Inter-examiner Kappa was calculated by use of SPSS Statistics 17.0 (SPSS Inc., Chicago, IL, USA).
To determine if Mmp20 heterozygous mice have reduced Mmp20 expression, we performed qPCR on mRNA from first molars of seven-day-old wild-type, heterozygous, and homozygous null mice. Both primers annealed to an area (catalytic domain) deleted from wild-type mice to generate the Mmp20 null mouse. This precluded detection of mRNA encoding a non-functional protein. The results demonstrated that the heterozygous mice expressed half the quantity of Mmp20 transcripts, and the homozygous null mice expressed no Mmp20 transcripts, when compared with the transcript levels observed in the wild-type mice (Fig. 1, top). Western blotting confirmed that MMP20 protein levels were also reduced by approximately half in the heterozygous mouse enamel (Fig. 1, bottom). Thus, no compensatory mechanism is present to increase expression of Mmp20 in the heterozygous mice.
After 6 wks of F− treatment, the mice were weighed, and bone (femur), urine, and plasma were collected for analysis of F− levels. The mice did not significantly differ (p = 0.146) in weight among genotypes and F− treatment groups (Fig. 2A). Plasma and bone F− levels (Figs. 2B, ,2C)2C) were significantly increased in the F−-treated mice compared with those in the untreated mice (p < 0.0001). However, within each fluoride treatment group, there was no significant difference in tissue fluoride levels between genotypes. This was the same for the urine analyses (Fig. 2D), except that the significance level was lower (p < 0.05), because urine could not be obtained from all available mice. In all cases, tissue F− levels between genotypes within the same F− treatment group were not significantly different (Fig. 2). Therefore, each treatment group of mice received a similar overall F− dose during the six-week treatment time.
By use of a modified TF scale (Fig. 3A), we demonstrated that a significant increase in dental fluorosis occurred for both mouse genotypes when mice were treated with 50 ppm F− vs. the 0 ppm controls (p < 0.0001). However, for each F− treatment group, dental fluorosis between genotypes was not significantly different (TF scale). The QF method measures enamel fluorescence that correlates highly to the severity of dental fluorosis. Incisor enamel QF in the wild-type or heterozygous mice treated with 50 ppm F− was significantly higher when compared with that of the 0 ppm controls (p < 0.0001). The 0 ppm F− controls were not significantly different between genotypes. However, there was a small but statistically significant difference (p < 0.05) in QF between the F−-treated Mmp20+/+ and Mmp20+/− mice (Fig. 3B). Also shown are pictures of teeth representative of those observed in our experiments from each mouse genotype and from each F− treatment group (Fig. 3C).
We performed Vickers microhardness testing on enamel from wild-type and Mmp20 heterozygous mice with or without F− exposure to determine if Mmp20 heterozygotes have softer-than-normal dental enamel. For both genotypes, microhardness of F−-treated enamel significantly decreased (p < 0.0001) as compared with that of controls (Fig. 4). No significant differences were observed between genotypes within their respective F− treatment group. These results indicate that while F− exposure results in softer enamel, the lack of an Mmp20 allele does not by itself significantly affect enamel hardness.
Previously we demonstrated in rodents (rats) that 6 wks of exposure to 0, 50, 100, or 150 ppm F− as NaF in drinking water does not affect Mmp20 mRNA levels (Sharma et al., 2010). This study demonstrated that mice with one mutated Mmp20 allele express approximately half the normal levels of MMP20, that tissue levels of F− are unaffected by Mmp20 heterozygosity, that Mmp20+/− mice have normal-appearing enamel as determined by TF score and quantitative fluorescence, that the F−-treated Mmp20+/− mouse enamel has more fluorescence than the wild-type mouse, that F− treatment results in softer-than-normal enamel, and that no difference in enamel hardness exists between the mouse genotypes within a F− treatment group.
A dose of 10 µM NaF was previously demonstrated to reduce MMP20 protein expression by 21.0% in cultured cells and was also demonstrated to inhibit an Mmp20 promoter-reporter construct by 35% (Zhang et al., 2007). While this may occur in vitro, our results generated in vivo show that loss of an Mmp20 allele with an associated 50% reduction in MMP20 expression does not cause a fluorosis-like phenotype. Tooth color was normal, and no difference in enamel hardness was observed between the genotypes. Therefore, dental fluorosis cannot be solely attributed to partial reduction of Mmp20 expression.
The only significant phenotype associated with decreased MMP20 expression was an increased average QF value in F−-treated Mmp20+/− incisors. To determine if the increased fluorescence was caused by a rougher enamel surface in Mmp20+/− mice compared with wild-type mice, we performed atomic force microscopy in contact mode with a silicon nitride tip. However, we found no significant difference in roughness between genotypes (data not shown). Although we have not directly compared the sensitivity of QF vs. microhardness values, it is possible that QF is a more sensitive fluorosis indicator than are microhardness determinations. This may account for the discrepancy in results between these two techniques. We did look at the EDTA soluble protein-rich zone of unerupted incisor enamel in wild-type mice and observed high QF values. These QF values decreased once the enamel erupted in its virtually protein-free form (unpublished results). Thus, enamel fluorescence may positively correlate with protein content. It is possible, therefore, that the F−-treated Mmp20+/− mice retain slightly more enamel protein than do the wild-type controls. Perhaps a few enamel matrix proteins remain uncleaved by MMP20 in Mmp20+/− enamel, and these larger proteins cannot be easily re-absorbed in conditions that cause dental fluorosis. Further studies are necessary to confirm this theory.
Recently, we demonstrated that F− induces the phosphorylation of the α-subunit of eukaryotic translation initiation factor, eIF2, in both mice and rats during the maturation stage of enamel formation (Sharma et al., 2008, 2010). eIF2α phosphorylation is associated with transient attenuation of global protein translation. An overall decrease in protein translation would be expected to reduce the quantity of protein necessary for routine cell activities. Therefore, F−-mediated eIF2α phosphorylation and subsequent reduction of protein translation are consistent with the F−-induced delay (30%) in the modulation cycle of maturation-stage rat ameloblasts between their ruffle-ended and smooth-ended morphologies (Smith et al., 1993). In addition, F− also suppresses the transcription of maturation-stage genes in the rat, including Klk4 and Amtn (Sharma et al., 2010). Our current study supports the existing hypotheses stating that a delay in the removal of matrix proteins during the maturation stage is responsible for the increased protein content of fluorosed enamel (DenBesten, 1986). In any case, this study demonstrates that enamel fluorosis cannot be solely attributed to a reduction of Mmp20 expression.
We thank Dr. Henry Margolis and Amy Litman for their enamel surface roughness determinations. This work was supported by NIDCR grants DE018106 (JDB), DE16276 (JDB), and DE018104 (ETE).