Calcif Tissue Int (2005) 76:17–25 DOI: 10.1007/s00223-004-0075-3

Tooth Quality in Dental Fluorosis: Genetic and Environmental Factors A. P. G. F. Vieira,1,2 R. Hanocock,3 H. Eggertsson,4 E. T. Everett,4 M. D. Grynpas1,2 1

Faculty of Dentistry, University of Toronto, 124 Edward St., Toronto-ON, Canada M5G 1G6 Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Ave., Room 840, Toronto-ON, Canada M5G JX5 Department of Chemical Engineering and Applied Chemistry, University of Toronto, 200 College St., Toronto, ON, Canada MS5 3E5 4 Indiana University School of Dentistry, 1121 West Michigan St., Indianapolis, IN 46202, USA 2 3

Received: 19 April 2004 / Accepted: 20 July 2004 / Online publication: 14 October 2004

Abstract. Dental fluorosis (DF) affects the appearance and structure of tooth enamel and can occur following ingestion of excess fluoride during critical periods of amelogenesis. This tooth malformation may, depending on its severity, influence enamel and dentin microhardness and dentin mineralization. Poor correlation between tooth fluoride (F) concentration and DF severity was shown in some studies, but even when a correlation was present, tooth fluoride concentration explained very little of DF severity. This fact calls into question the generally accepted hypothesis that the main factor responsible for DF severity is tooth fluoride concentration. It has been shown previously that genetic factors (susceptibility to DF) play an important role in DF severity although DF severity relates to individual susceptibility to fluoride exposure (genetics), tooth fluoride concentration relates to fluoride ingestion (environmental). The objective of this study was to investigate the correlation between tooth fluoride concentration, DF severity, and tooth mechanical and materials properties. Three strains of mice (previously shown to have different susceptibility to DF) at weaning were treated with four different levels of F in their water (0, 25, 50, and 100 ppm) for 6 weeks. Mice teeth were tested for fluoride by instrumental neutron activation analysis (INAA), DF severity determined by quantitative lightinduced fluorescence [QLF], and tooth quality (enamel and dentin microhardness and dentin mineralization). Tooth fluoride concentration (environment factor) correlated positively with DF severity (QLF) (rs = 0.608), fluoride treatment group (rs = 0.952). However, tooth fluoride concentration correlated negatively with enamel microhardness (rs = )0.587), dentin microhardness (rs = )0.268) and dentin mineralization (rs = )0.245). Dental fluorosis (genetic factor) severity (QLF) correlated positively with fluoride treatment (rs = 0.608) and tooth fluoride concentration (rs = 0.583). DF severity correlated negatively with enamel microhardness (rs = )0.564) and dentin microhardness (rs = )0.356). Genetic factors (DF severity) and the environmental factor (fluoride concentration in tooth structure) have similar influence on tooth biomechanical properties, whereas

Correspondence to: M. D. Grynpas; E-mail: grynpas@ mshri.on.ca

only the environmental factor has an influence on tooth material property (mineralization). Key words: Genetic susceptibility — Dental fluorosis — Tooth properties

Dental fluorosis (DF) is a tooth malformation related to fluoride (F) ingestion during tooth development [1, 2]. Greater than optimal amounts of fluoride (taken systemically) contributes to a greater risk in developing DF. Fluoride interacts with mineralized tissues and at elevated concentrations disturbs the mineralization process [3]. The subsurface enamel all along the tooth becomes increasingly porous (hypomineralized) extending toward the inner enamel. Hypomineralization of the dentin results in an enhancement of the incremental lines. Following tooth eruption, the more severe forms of fluorosis are subject to extensive mechanical breakdown of the surface [4–6]. The continuum of fluorideinduced clinical changes has been well classified by the TF (Thylstrup Fejerskov) index. TFI reflects, on an ordinal scale, the histopathologic features that correlate with the clinical features seen in fluorotic teeth [7]. Clinically, enamel fluorosis is defined by the presence of characteristic opacities, which was referred to in the early literature as mottled enamel [8]. Very mild fluorosis manifests as few faint white flecks. With increasing severity the white flecking becomes more prominent (snowcapping or snowflaking) and involves more of the tooth surface. The most severe form of enamel fluorosis is characterized by dark brown staining and large posteruptive enamel defects (pitting). The risk of developing enamel fluorosis appears to be greatest when exposure occurs during both secretory and maturation stages of enamel formation [9–12]. A delay in the hydrolysis and removal of enamel proteins, particularly amelogenins, as the enamel matures may contribute to the subsurface porosity [13]. DF prevalence has increased throughout the world [14], ranging between 7.7% and 80.9% in areas with fluoridated water

18

and between 2.9% and 42% in areas without fluoridated water [15–17, 26]. This tooth malformation may, depending on its severity, influence tooth quality. Tooth quality relates to tooth structural and material properties, and those, in turn, relate to the tooth’s ability to fulfill its function and sustain mastigatory forces. Tooth microhardness and tooth mineralization levels are important parameters of tooth quality (mechanical and material properties). Previous works have shown little correlation between tooth fluoride concentration and DF severity [18–20]. Moreover, work from our laboratory has shown that, in unerupted third molars from cities with optimum and suboptimum levels of fluoride in the drinking water, there is a lack of correlation between enamel fluoride concentration and DF severity as well as weak correlation between dentin fluoride concentration and DF severity [21]. Even when a correlation was present, tooth fluoride concentration explained only 10% of DF severity. This fact calls into question the generally accepted hypothesis that the main factor responsible for DF severity is the fluoride concentration in tooth structure. Epidemiologic studies have shown that certain ethnic groups are more susceptible to DF than others [22]. In the Grand Rapids fluoridation study, DF was twice as prevalent in African American than in white children [23]. The odds ratio for African American children having DF, compared with Hispanic and non-Hispanic white children, was 2.3 in the 1980s Texas surveys [24]. This trend was also seen in a Georgia study, which showed higher prevalence of DF among African Americans [25]. A study on various strains of mice by Everett et al., where rigorous control of genotype, age, gender, food, housing, and drinking water fluoride levels were present, showed that animals from different strains, with the same level of fluoride concentration in their mineralized tissue, presented different degrees of DF severity [26]. This study also showed significant correlation between fluoride concentration in water and in bone (r = 0.869, P < 0.00001), which shows that the amount of fluoride found in subject mineralized tissue is related to the amount ingested (environmental factor). Despite all the information we have about fluoride ingestion and DF severity, it is still unknown if tooth quality is more affected by fluoride concentration in tooth structure or by DF severity. Because DF severity relates to individual susceptibility to fluoride (genetics) and tooth fluoride concentration relates to fluoride ingestion (environmental), the study of the relationship between DF severity, tooth fluoride concentration and tooth quality is important to separate the influence of genetic and environment factors in tooth quality. The aim of this study was to investigate the correlation between tooth fluoride concentration; tooth DF severity, and tooth quality in mice from different strains

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

subjected to different levels of fluoride concentration in the drinking water in order to distinguish between genetic and environmental factors.

Materials and Methods Mice and Fluoride Treatment Three strains of male mice were used in this study—A/J, 129P3/J and SWR/J and all were obtained from The Jackson Laboratory (Bar Harbor, ME). The selection of these mouse strains was based on results from a previous work from Everett et al. (2002), showing different levels of susceptibility to DF. Mice were fed regular laboratory rodent diet containing fluoride concentrations ranging between 5 lg/g to 8 lg/g. A total of 72 male mice, from three strains were treated using four different levels of fluoride treatment in the drinking water (0, 25, 50, and 100 ppm) for 42 days. Each treatment group had six mice. Mice were allowed water ad libitum and fluoride treatment was started when mice were 4 weeks old. Two mice from the 129P3/J strain (25 ppm fluoride treatment group) died before the end of the treatment; therefore, only 70 mice were evaluated in this study (only four in the 129P3/J 25 ppm fluoride treatment group). DF Severity DF severity in mice teeth was determined using quantitative light-induced fluorescence (QLF). After 42 days of treatment mandibular incisors were dissected and the coronal third was evaluated using QLF (Inspecktor Research System BV, Amsterdam, The Netherlands) as previously described by Everett et al. (2002). Fluoride Analyses Mice upper incisors were analyzed for fluoride concentration using instrumental neutron activation analysis (INAA). In INAA, each sample is bombarded with thermal neutrons that produce short-lived radioisotopes from the elements in the sample. These radioisotopes decay with specific half-lives, emitting gamma rays of discrete and characteristic energies. The relative amounts of gamma rays detected are proportional to the concentrations of the elements in the sample [27]. INAA is capable of measuring fluoride concentration >50 ppm (0.05%) to the 10 ppm level (0.01%). Its typical precision and accuracy vary from ± 2% to ± 10% depending on the element, the nature of the sample matrix, and absolute concentration of the element [28]. Tooth Quality—Mechanical and Material Properties After DF severity evaluation (using QLF), the coronal part of mice lower incisor was transferred separately to molds and embedded in epoxy resin (Epoxycure resin, Buehler, Markham, Canada), which was allowed to polymerize at room temperature (21
C) overnight. The blocks were ground using 400 grit sandpaper (Phoenix BETA Grinder/Polisher fitted with the VECTOR Power Head, Buehler) until half of the tooth (coronal-apical direction) was exposed. Ground surfaces were then polished using 6l and 1l polishing pastes. Half of the samples (n = 70) were tested for mechanical properties (microhardness), while the other half (n = 70) was tested for material properties (Mineralization level was determined using backscattered electron microscopy and image analysis.) Teeth microhardness was tested using a Vicker indentor (100 g per 10 seconds). Prior to indentation, teeth were rehy-

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

19

drated in distilled water for 10 minutes. A total of 12 indentations were performed per tooth. Four indentations were performed in enamel and eight in dentin at two sites, 100 lm (dentin 100 lm) and 300 lm (dentin 300 lm) from the dentin enamel junction (DET) (Fig. 1). The average value for all dentin indentations was also calculated (average dentin). The location of the indentations were based on a previous study that performed microhardness in mice teeth [29]. The calculation of the microhardness on each point was done by using the equation below: HV ¼

1:854P d21

where P is applied load (kg), d1 is diameter of the indentation (mm), and HV is Vicker’s microhardness (kg/mm2). Extensive attrition and enamel loss in the tooth samples from the 100 ppm A/J strain mice prevented microhardness testing; therefore, those enamel samples were not analyzed in this study. The degree of mineralization of the coronal third of the mice dentin was evaluated utilizing backscattered electron imaging (BSE). The image formed by the backscattered electrons exhibits a range of brightness that is related to the degree of mineralization. This image is divided into various gray levels (256 gray bins) of increasing mineral density. The area percentage of each image in each intensity level is calculated and a histogram based on those values is produced. In order to have one number representing the overall mineralization pattern of the tooth a Logit function was utilized (dentin mineralization logit). The value for X was set at the gray level 130 (approximately one half point of the 256 gray levels present in the distribution).   area  X Logit ¼ ln areaX The Logit value was used to quantify the mineralization patterns in the coronal dentin (one third of tooth—coronal section) Figures 2 and 3. A high logit value represents a high degree of dentin mineralization.

Statistics

Descriptive statistics, Spearman’s correlation, and Kruskall-Wallis test were used when appropriated. Significant values were set at P £ 0.05. Statistical analysis was done with the use of statistical analysis software (SPSS for Windows, SPSS Inc., Chicago, IL, USA)

Results

A total of 70 mice, from three different strains (A/J, 129P3/J, and SWR/J) were studied. Six in each group with the exception of the group 129P3/J 25 ppm fluoride treatment group, which only had 4 mice. Two mice died before the end of the experiment. The overall tooth F concentration varied between 76 ppm and 2385 ppm, whereas the QLF values (related to DF severity) varied between 0 and 2671. The enamel microhardness values varied between 27 kg/ mm2 and 256 kg/mm2 and the dentin microhardness between 31 kg/mm2 and 87 kg/mm2, whereas the dentin mineralization (total dentin mineralization logit)

Fig. 1. Schematic of sites where microhardness indentations were taken.

varied between 2.2 and 5.1. The data for each subgroup (strain and fluoride treatment) are presented in Table 1. When the strains, in the 0 ppm fluoride treatment group were compared (Kruskall-Wallis test), enamel microhardness and dentin mineralization differed significantly between the strains. The enamel microhardness in the 129P3/J strain was significantly harder than in the A/J and SWR/J strains (P < 0.05). The dentin of the SWR/J strain was significantly less mineralized than the A/J and 129P3/J strains (P < 0.05). In addition, there was a significant difference between the dentin microhardness values at 100 lm and 300 lm (from the DEJ) among the three strains in the 0 ppm fluoride treatment group, with the microhardness values being greater at 100 lm from the DEJ. In the other fluoride treatment groups there was no difference between microhardness performed 100 lm and 300 lm from the dentin enamel junction (DEJ). Due to lack of difference in most groups only dentin microhardness at 100 lm from the DEJ was used during correlation tests. Therefore, dentin microhardness and dentin microhardness at 100 lm from the DEJ are used interchangeably in the following text. When the strain A/J and SWR/J were analyzed (Kruskall-Wallis) for the four different fluoride treatment groups, significant differences were seen between the fluoride treatment groups in QLF and tooth fluoride concentration. It was shown that increasing values of fluoride regimen resulted in increasing values of QLF and tooth fluoride concentration. Interestingly, despite the similar values of tooth fluoride concentration in each fluoride treatment group (0, 25, 50, and 100 ppm), the QLF values (related to DF severity) for the A/J group were much higher than for the SWR/J group, demonstrating a higher susceptibility of the A/J group to fluoride (Table 1). The tooth fluoride concentrations of the 129P3/J strain were equivalent to the tooth fluoride concentration in the A/J and SWR/J strains.

20

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

Fig. 2. Schematic of mineralization distributions and logit calculations.

Fig. 3. Schematic of mineralization pattern in different mouse groups depicting increase in mineralization pattern. A shift on the curve, as seen in this figure, shows an increase in mineralization of one group over the other. The curve toward the more mineralized end of the curve (right side of the graph) represents a higher mineralized group. This shift can be detected by using the logit function.

When all the strain were evaluated together (Table 2), tooth fluoride concentration correlated positively with DF severity (QLF) (rs = 0.608, P < 0.001) and with fluoride treatment group (rs = 0.952, P < 0.001). However, tooth fluoride concentration correlated negatively to enamel microhardness (rs = )0.587, P < 0.001), dentin microhardness (rs = )0.268, P < 0.03) and dentin mineralization (rs = )0.245, P < 0.05).

Dental fluorosis severity (QLF) correlated positively with fluoride treatment (rs = 0.608, P < 0.01), DF severity correlated negatively to enamel microhardness (rs = )0.564, P < 0.001) and dentin microhardness (rs = )0.356, P < 0.03). This shows that when all strains are considered, tooth fluoride concentration is the most important factor influencing tooth quality, as it correlates with tooth mechanical and material properties. QLF, which is related to DF severity, correlates only with tooth mechanical properties. Correlation was also performed with each strain separately (Table 3). In the A/J strain, QLF correlated positively with fluoride treatment (rs = 0.803, P < 0.001), tooth fluoride concentration (rs = 0.717, P < 0.001) and negatively with enamel microhardness (rs = )0.663, P < 0.005) and dentin microhardness (rs = )0.485, P < 0.02). The tooth fluoride concentration in the A/J strain, on the other hand, correlated positively with QLF (rs = 0.717, P < 0.001) and fluoride treatment (rs = )0.969, P < 0.001), and negatively with enamel microhardness (rs = )0.853, P < 0.001). There was a trend between tooth fluoride concentration and dentin microhardness (rs = 0.397, P = 0.54) in the A/J strain. In the 129P3/J strain, DF severity (QLF) correlated with only fluoride treatment (rs = 0.554, P < 0.007) and tooth fluoride concentration (rs = 0.424, P < 0.05), whereas tooth fluoride concentration correlated positively with DF severity (rs = 0.424, P < 0.05) and fluoride treatment (rs = 0.914, P < 0.001), and negatively with enamel microhardness (rs = )0.598, P < 0.009) and dentin microhardness (rs = )0.478, P < 0.03). In the SWR/J strain, the DF severity (QLF) correlated positively with fluoride treatment (rs = 0.688, P < 0.001) and

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

21

Table 1. Descriptive statistics of mice strain groups Mouse strain

Water [F] (ppm)

A/J

0

25

50

100

129P3/J

0

25

129P3/J

50

100

SWR/J

0

25

50

100

Statistics

QLF (DF severity)

Tooth [F] (ppm)

Enamel microhardness

100 Pm Dentin microhardness (kg/mm2)

300 Pm Dentin microhardness (kg/mm2)

Dentin logit (dentin mineralization)

Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD

0 1 0.25 0.418 0 72 18.75 28.4 211 1517 595.7 479.6 56 2,671 886.0 1,138.2 0 2 0.3 0.6 0 1 0.3 0.5 0 40 6.7 16.3 0 425 85.2 167.3 0 13 2.1 5.2 0 0 0 0 0 30 5.2 12.2 0 177 66.0 74.8

97 196 147.8 43.9 341 560 444.2 79.0 894 1001 944.33 35.1 1827 2385 2,110.3 219.5 81 189 132.5 45.3 456 872 584.0 193.9 264 817 633.0 204.3 1377 1927 1632.7 213.9 76 150 113.7 31.0 439 591 513.7 64.1 718 1317 912.3 219.0 1472 2183 1800.8 243.2

139 205 175.2 26.7 79 210 129 50.1 32 84 51.6 20 — — — — 206 256 222.4 19.4 134 215 182.8 35.5 140 227 190 45 64 190 118.7 64.6 112 207 160.3 38.0 73 198 145.5 49.5 137 150 143 6.4 27 60 48.3 12.9

51 79 70.9 9.9 54 83 70.8 10.5 56 80 69.9 8.6 34 71 55.5 13.2 73 82 76.2 3.7 58 83 73.9 11.4 60 83 76.4 9 60 72 68.3 4.5 67 80 71 4.8 56 84 70.6 10 73 87 78 5.1 46 78 64 13.2

43 75 60.3 12.1 41 76 64.8 12.3 60 76 66.3 5.9 31 64 53.7 11.8 62 76 68.6 6.3 52 72 65.9 9.6 54 81 69.7 9.2 59 72 66.2 4.9 56 67 62.3 4 56 77 54.5 7.6 56 80 70 7.8 46 83 66 13.2

3.81 4.8 4.26 0.35 3.36 4.4 3.7 0.39 2.26 4.01 3.5 0.64 3.8 5.1 4.39 0.549 3.22 5.01 4.12 0.8 3.75 4.24 3.95 0.22 2.99 4.58 3.76 0.64 3.35 4.54 4.1 0.46 3.21 3.74 3.5 0.2 2.17 4.39 3.7 0.8 3.84 4.58 4.11 0.26 3.74 4.94 4.31 0.47

SD, standard deviation, [F], fluoride concentration

tooth fluoride concentration (rs = 0.637, P < 0.001), and negatively with enamel microhardness (rs = )0.727, P < 0.001). There was a negative trend between DF severity and dentin mineralization in the SWR/J strain (rs = )0.395, P < 0.06). In this same strain (SWR/J), tooth fluoride concentration correlated positively with DF severity (rs = 0.637, P < 0.001), fluoride treatment (rs = 0.969, P < 0.001), and ne-

gatively with enamel microhardness (rs = )0.573, P < 0.02) and dentin mineralization (rs = )0.678, P < 0.001). Those results showed how different strains differ in how they are influenced by different factors, demonstrating the importance of genetic factors in tooth quality. The coefficient of determination (r2), which expresses the proportion of variance in the dependent variable

22

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

Table 2. Correlation data (Spearman correlation) — all strains combined QLF (DF severity) QLF (DF severity) Fluoride treatment [F] in water Tooth [F] Enamel Microhardness Dentin microhardness (at 100Pm from DEJ) Dentin mineralization (total dentin logit)

Correlation coefficient Coefficient of determination n Correlation coefficient Coefficient of determination n Correlation coefficient Coefficient of determination n Correlation coefficient Coefficient of determination n Correlation coefficient Coefficient of determination n Correlation coefficient Coefficient of determination n

0.583* 34% 70 0.952* 91% 70

r2 - % r2 - % r2 - % r2 - % r2 - %

Tooth [F]

0.608* 37% 70 0.583* 34% 70 )0.564* 32% 52 )0.356** 13% 70

)0.587* 34% 52 )0.268** 7% 70 )0.245** 6% 69

r2 - %

*P < 0.001; **P < 0.05; %, percentage of variance in one variable explained by the other variable

Table 3. Correlation data (Spearman correlation)—strains separated A/J QLF QLF (DF severity)

Fluoride treatment ([F] in water) Tooth [F]

Enamel Microhardness Dentin microhardness (at 100Pm from DEJ) Dentin mineralization (total dentin logit)

*P < 0.001; **P < 0.05;

Correlation coefficient Coefficient of determination r2,% n Correlation coefficient Coefficient of determination r2, % n Correlation coefficient Coefficient of determination r2, % n Correlation coefficient Coefficient of determination r2, % n Correlation coefficient Coefficient of determination r2, % n Correlation coefficient Coefficient of determination r2, % n

129P3/J Tooth [F]

QLF

0.717* 51% 0.803* 64% 24 0.717* 51%

24 0.969* 94% 24

SWR/J Tooth [F]

QLF

0.424** 18% 0.554** 31% 22 0.424** 18% 22

22 0.914* 84% 22

24 )0.663** 44%

)0.853* 73%

16 )0.485** 24%

16 )0.397+ 16%

18 )0.478** 23%

24

24

22

)0.598** 36%

Tooth [F] 0.637* 41%

0.688* 48% 24 0.637* 41%

24 0.969* 94% 24

24 )0.727* 53%

)0.573** 33%

18

18

)0.395+ 16%

)0.678* 46%

24

24

+

Statistical trend (0.05 < P > 0.1)

explained by the independent variable [30], was calculated for each of the significant correlations (Tables 2 and 3). When enamel microhardness and dentin microhardness and mineralization (all strains collectively) were evaluated, enamel microhardness and dentin microhardness were more influenced by tooth fluoride

concentration, whereas dentin mineralization was influenced only by DF severity (QLF). When the correlation between QLF and dentin microhardness was analyzed, the r2 was 0.13, which means that 13% of the dentin 100 lm microhardness can be explained by the DF severity (QLF). The r2 in the correlation between

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

tooth fluoride concentration and dentin microhardness was 0.07 (7%). When enamel microhardness was being examined, QLF and tooth fluoride concentration showed significant correlation (r2 = 0.32 and 0.34, respectively). When dentin mineralization was being examined, only tooth fluoride concentration showed significant correlation, and even then, with very low r2 (r2 = 0.06).

Discussion

This is the first time that the relationship between tooth fluoride concentration, DF severity, and tooth property (mechanical and material) are studied in the same set of samples. Previous work have shown that the correlation between human tooth fluoride concentration and DF severity is weak, which means that teeth with the same level of fluoride have very different levels of DF [19–21]. Therefore, it is important to know if the tooth quality, which relates to the capacity of the tooth to fulfill its functions, is more affected by, and/or related to, DF severity or tooth fluoride concentration. Tooth quality can be analyzed by measuring tooth properties, such as material and mechanical. Material properties are those properties specific (intrinsic) to a material, whereas mechanical properties are those properties that reveal the reaction, either elastic or plastic, of a material to an applied stress. Tooth mechanical properties themselves are a function of both tooth material and architecture. In this article, material property was analyzed by measuring tooth mineralization whereas mechanical property was analyzed by using tooth microhardness. Previous studies in mice [31] and humans [21], as well as epidemiological studies [23–25] have demonstrated that severity of dental fluorosis cannot be explained simply by the amount of fluoride in the tooth structure, indicating that genetics (susceptiblity to fluoride) plays an important role in DF severity. Based on that, one can infer that in individuals ingesting the same amount of fluoride, the DF severity will be related to and/or based on individual susceptibility to fluoride (genetics). On the same note, the level of fluoride an individual ingests is dependent on the amount of fluoride he or she is exposed to in the environment. Previous studies have shown that the amount fluoride in an individuals calcified tissues is related to the amount of fluoride ingested [21, 31]. It is, therefore the environment of a person that ‘‘controls’’ the individual fluoride intake and consequently the amount of fluoride in its mineralized tissues (e.g., bone and teeth). Because DF severity relates to individual susceptibility to fluoride (genetics) and tooth fluoride concentration relates to fluoride ingestion (environmental), the study of the relationship among DF severity, tooth fluoride concentration, and tooth

23

quality are important to clarify, determine and separate the influence of genetic and environment factors in tooth quality. When all strains were analyzed together, an excellent correlation between tooth fluoride concentration and fluoride treatment (the amount of fluoride ingested) was shown, confirming the use of tooth fluoride concentration as an environmental factor. When DF severity was evaluated, a correlation was seen between QLF (DF severity) and tooth fluoride concentration (environmental factor), but only 34% of the variance of DF severity was explained by the concentration of fluoride in tooth structure. This observation shows that other factors apart from tooth fluoride concentration are important in DF severity. In addition, these findings confirmed previous studies in animal models and in humans [21, 31], which showed genetic susceptibility to be an important factor in DF severity. This study shows that both the genetic (DF severity) and environmental (tooth fluoride concentration) factors play a role in tooth quality. However, when evaluating all strains together, environmental (tooth fluoride concentration) was the only factor that correlated with both tooth mechanical and material properties. QLF correlated only with tooth mechanical property. One can also appreciate that the coefficient of determination values for enamel microhardness were much higher than for the other variables (dentin microhardness and dentin mineralization). This study showed that enamel was more influenced by environmental and genetic factors than dentin. Dentin mineralization is more regulated by proteins than enamel [32]. Some proteins, like phosphophyryn, sialoprotein, osteocalcin, osteonectin, osteopontin found in dentin and/or bone, are known to regulate mineralization [33]. Therefore, the influence of dental fluorosis and/or tooth fluoride concentration in dentin may be less noticed due to this stronger regulation of its mineralization. This factor would explain the small percentage of variance in this property explained by tooth fluoride concentration. The fact that a negative correlation was found between the environmental and genetic factor (tooth fluoride concentration and QLF) and tooth quality, shows that with increasing severity of dental fluorosis and/or tooth fluoride concentration, decreasing values of tooth quality were found. This demonstrates a negative influence of those factors (environmental and genetic) on tooth quality. In humans, histologic changes owing to ingestion of fluoride have been found more easily in the enamel, but in severe fluorosis the dentin has also shown histologic modifications [34, 35]. Mechanisms that have been proposed to explain the formation of fluorosed enamel include a systemic effect of fluoride on calcium homeostasis, altered matrix biosynthesis (protein secretion,

24

synthesis of mineral composition), a direct or indirect effect on matrix proteinases affecting protein removal, and specific effects on cell metabolism and function. Most available evidence indicates that fluoride has an effect on cell function, either directly through interactions with the developing ameloblasts or more indirectly by interacting with the extracellular matrix. Because of this defect in protein removal during tooth mineralization, a more porous enamel is formed, which is therefore less mineralized and more brittle. When evaluating the different strains separately, we noted that in the more DF susceptible strain (A/J strain), QLF correlated with more variables (enamel and dentin microhardness) than in the other strains. This shows that DF severity becomes a more important factor in tooth quality in susceptible strains. In addition, the percentage of variance of tooth mechanical property (enamel and dentin microhardness) related to DF severity (QLF) and tooth fluoride concentration is higher in the more susceptible strain (A/J). When strains were analyzed separately, only the SWR/J strain demonstrated correlation between tooth fluoride concentration and dentin mineralization (total dentin logit) and a trend between DF severity and dentin mineralization. The explanation for this is not clear; it may be due to a genetic predisposition, demonstrated by a less-regulated dentin mineralization in the SWR/J strain. This would result in a higher susceptibility of this strain to the influence of DF severity and tooth fluoride concentration. It is important to clarify that in the A/J strain the enamel of the 100 ppm fluoride treatment group, after processing, was not available for microhardness testing. This may be caused by the higher susceptibility of this strain (A/J) to F [31]. Fluoride action on the tooth enamel may have created a more porous (brittle, fragile) enamel, which did not retain its integrity during tooth dissection and preparation. Finally, the differences seen in enamel microhardness between the strains at 0 ppm of fluoride are interesting and may indicate that enamel hardness is genetically determined. It is also apparent by these data that DF susceptibility is somewhat independent of enamel hardness. In other words A/J and SWR/J have similar enamel hardness (at 0 ppm) yet different susceptibilities to DF. This is not seen in dentin hardness, which changes very little with fluoride exposure and does not vary much in the different strains.

Conclusions

This study showed that environmental (fluoride concentration in tooth structure) and genetic (susceptibility to DF/mice strain) factors both influence tooth quality. Genetic factor (DF severity, QLF) and environmental

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

factor (fluoride concentration in tooth structure) have similar influence on tooth biomechanical property, whereas only environmental factors have an influence on tooth material property (mineralization). However, the relative low coefficient of determination values found in some correlations, show that those results should be taken with caution and that further studies in this area are necessary.

Acknowledgments. We would like to thank Ms. Deidra Faust for her technical assistance during this study and Dr. Angeles Martienz-Mier for fluoride analyses on food and water samples. This work was funded by a grant from the Canadian Institute of Health Research (CIHR) and by the NIH/NIDCR (R01 - DE014853) to ETE. AV is the recipient of Harron and Connaught Scholarships.

References 1. DenBesten PK (1994) Dental fluorosis: its use as a biomarker. Adv Dent Res 8:105–110 2. Murray JJ, Rugg-Gunn AJ, Jenkins GN (1991) Epidemiology and measurement of dental fluorosis. Fluoride in caries prevention. Wright, Oxford, pp 222–261 3. Aoba T (1997) The effect of fluoride on apatite structure and growth. Crit Rev Oral Biol Med 8:136–153 4. Fejerskov O, Larsen MJ, Richards A, Baelum V (1994) Dental tissue effects of fluoride. Adv Dent Res 8:15–31 5. Fejerskov O, Manji F, Baelum V (1990) The nature and mechanisms of dental fluorosis in man. J Dent Res Spec 69:692–700 6. Yanagisawa T, Takuma S, Fejerskov O (1989) Ultrastructure and composition of enamel in human dental fluorosis. Adv Dent Res 3:203–210 7. Thylstrup A, Fejerskov O (1978) Clinical appearance of dental fluorosis in permanent teeth in relation to histologic changes. Commun Dent Oral Epidemiol 6:315–328 8. Dean HT (1934) Classification of mottled enamel diagnosis. J Am Dent Assoc 21:1421 9. DenBesten PK, Crenshaw MA (1987) Studies on the changes in developing enamel caused by ingestion of high levels of fluoride in the rat. Adv Dent Res 1:176–180 10. DenBesten PK, Crenshaw MA, Wilson MH (1985) Changes in the fluoride-induced modulation of maturation stage ameloblasts of rats. J Dent Res 64:1365–l370 11. DenBesten PK, Heffernan LM, Featherstone JD, Shields CP (1992) Fluoride binding by matrix proteins in rat mineralizing tissue. Arch Oral Biol 37:459–462 12. DenBesten PK, Thariani H (1992) Biological mechanisms of fluorosis and level and timing of systemic exposure to fluoride with respect to fluorosis. J Dent Res 71:1238–1243 13. DenBesten PK (1999) Mechanism and timing of fluoride effects on developing enamel. J Public Health Dent 59:247–251 14. Angmar-Mansson B, Jong E, Sundstrom F, Bosch J (1994) Strategies for improving the assessment of dental fluorosis: focus on optical techniques. Adv Dent Res 8:75–79 15. Clark DC (1994) Trends in prevalence of dental fluorosis in North America. Community Dent Oral Epidemiol 22:148–152 16. Mascarenhas AK (2000) Risk factors for dental fluorosis: a review of the recent literature. Pediatr Dent 22:269–277 17. Pendrys DG (2000) Risk of enamel fluorosis in nonfluoridated and optimally fluoridated populations: considerations for the dental professional. J Am Dent Assoc 131:746–755 18. Olsen T, Johansen E (1978) Fluoride content of human fluorosed enamel. J Dent Res 57:281

A. P. G. F. Vieira et al.: DF Genetic Susceptibility and Tooth Quality

19. Richards A, Fejerskov O, Baelum V (1989) Enamel fluoride in relation to severity of human dental fluorosis. Adv Dent Res 3:147–153 20. Richards A, Likimani S, Baelum V, Fejerskov O (1992) Fluoride concentrations in unerupted fluorotic human enamel. Caries Res 26:328–332 21. Vieira APGF, Hancock R, Maia R, Limeback H, Grynpas MD (2004) Is fluoride concentration in dentin and enamel a good indicator of dental fluorosis? J Den Res 83:76–80 22. National Research Council. Health effects of ingested fluoride. National Research Council. 1993. National Academy Press, USA 23. Russell AL (1962) Dental fluorosis in Grand Rapids during the seventeenth year of fluoridation. J Am Dent Assoc 65:608–612 24. Butler WJ, Segreto V, Collins E (1985) Prevalence of dental mottling in school-aged lifetime residents of 16 Texas communities. Am J Public Health 75:1408–1412 25. Williams JE, Zwemer JD (1990) Community water fluoride levels, preschool dietary patterns, and the occurrence of fluoride enamel opacities. J Public Health Dent 50:276– 281 26. Everett ET, McHenry MA, Reynolds N, Eggertsson H, Sullivan J, Kantmann C, Martinez-Mier EA, Warrick JM, Stookey GK (2002) Dental fluorosis: variability among different inbred mouse strains. J Dent Res 81:794–798 27. Mernagh JR, Harrison JE, Hancock R, McNeill KG (1977) Measurement of fluoride in bone. Int J Appl Radiat Isot 28:581–583

25

28. Gordon GE, Randel K, Goles G, Corliss J, Benson M, Oxley S (1968) Instrumental activation analysis of standard rocks with high resolution gamma-ray detectors. Geochmicia et Cosmochimica Acta 32:369–396 29. DenBesten PK, Machule D, Gallagher R, Marshall GW Jr., Mathews C, Filvaroff E (2001) The effect of TGF-beta 2 on dentin apposition and hardness in transgenic mice. Adv Dent Res 15:39–41 30. Streiner DL, Norman GR (2000) Simple regression and correlation. In: Norman GR, Streiner DL (eds) Biostatistics: the bare essentials. BC Decker, Hamilton, London, pp 118–126 31. Ten Cate AR (1994) Oral histology: development, structure, and function Mosby, St. Louis 32. Bartold PM, Narayanan AS (1998) Biology of the periodontal connective tissues. Quintessence Books, Carol Stream, Ill 33. Fejerskov O, Thylstrup A, Joost Larsen M (1977) Clinical and structural features and possible pathogenic mechanisms of dental fluorosis. Scand J Dent Res 85:510–534 34. Fejerskov O, Yeager JA, Thylstrup A (1979) Microradiography of the effect of acute and chronic administration of fluoride on human and rat dentine and enamel. Arch Oral Biol 24:123–130 35. Fejerskov O, Richards A, DenBesten PK (1996) The effect of fluoride on tooth mineralization. In: Fejerskov O, Ekstrand J, Burt BA (eds) Fluoride in dentistry Munksgaard Textbook, Munksgaard, pp 112–152