The bio-availability of uoride from black tea

Transcription

1 Journal of Dentistry 29 (2001) 15±21 Journal of Dentistry The bio-availability of uoride from black tea A. Simpson, L. Shaw, A.J. Smith* School of Dentistry, University of Birmingham, Birmingham, UK Received 2 March 2000; revised 27 July 2000; accepted 17 October 2000 Abstract Objectives: To investigate the oral retention of uoride from tea and its association with the tooth surface and acquired pellicle. Methods: Oral retention of uoride after rinsing in vivo was assessed from expectorated samples with an ion speci c electrode methodology. Interaction of uoride with the tooth surface and acquired pellicle was examined in situ with enamel blocks mounted on partial removable appliances. In vitro models were used to examine uoride binding to enamel particles. Results: Thirty four percent of the uoride was retained in the oral cavity after rinsing with tea. Differences in retention at the tooth surface in the presence and absence of an acquired pellicle were not statistically signi cant at incisor or molar sites. Fluoride from tea showed strong binding to enamel particles, which was only partially dissociated by solutions of ionic strength considerably greater than that of saliva. Binding studies demonstrated strong avidity of enamel for tea and salivary pellicle components. Conclusions: This study has demonstrated that tea can provide an effective vehicle for uoride delivery to the oral cavity where it may interact with the oral tissues and their surface integuments. This may lead to local topical effects of the ingested uoride as well as systemic effects following oral and gastro-intestinal absorption. q 2001 Published by Elsevier Science Ltd. All rights reserved. Keywords: Fluoride; Tea; Enamel; Acquired pellicle; Bio-availability; Diet 1. Introduction Intake of uoride is variable and is derived from a number of sources in the diet and the environment. Whilst a number of uoride compounds may contribute to air pollution, atmospheric exposure to uoride makes only a small contribution to the total intake of this element for people who do not live in heavily polluted areas [1,2]. The major uoride intake for most of the population is derived from dietary sources. There are very wide variations in the levels of uoride naturally occurring in fresh water ranging from 0.01 to 100 ppm [3], although the majority will contain 1 ppm of uoride or less. Where arti cial uoridation of water supplies is undertaken, a level of 1 ppm in temperate climates is generally regarded as being optimal for prevention of dental caries and minimising clinical symptoms of uorosis [4]. Sea water contains approximately 1.5 ppm uoride and reports of uoride levels in sh have been contradictory. An early study suggested levels of 6± 27 ppm uoride in sh [5], although a later study carefully avoiding contamination with bones reported much lower * Corresponding author. Tel.: ; fax: address: a.j.smith@bham.ac.uk (A.J. Smith). levels of 0.06±1.69 ppm [3]. Fresh or unprepared foods generally have uoride concentrations in the range 0.01± 1.0 ppm [3], although preparation of that food often using uoride-containing water can lead to slight increases in concentrations. A range of concentrations from 0.02 to 2.74 ppm have been reported for food prepared with local water containing 1 ppm uoride for hospital patients [6]. A very rich dietary source of uoride is tea where the dry leaves may contain 4±400 ppm uoride and the brewed tea 0.34±6 ppm [7±9]. Fluoride ingestion from swallowing toothpaste and other uoride-containing products may also be signi cant [10±12]. It is clear that dietary uoride intake may be quite variable and average daily intakes have been estimated to be in the range 1.2±1.8 mg/day for adults [6,13,14] and 0.32± 0.61 mg/day for children in the age group of 6 months±2 years [15,16]. Tea consumption may provide an appreciable proportion of the daily dietary intake of uoride [17,18] for some persons. The bioavailability of dietary uoride from this and other dietary sources has focussed on its gastrointestinal absorption [19]. However, little information exists on the bio-availability of uoride from tea for oral absorption or its topical interaction with the oral tissues. The latter interactions may be important in respect of modifying the carious process at the tooth surface and this study /01/$ - see front matter q 2001 Published by Elsevier Science Ltd. All rights reserved. PII: S (00)

2 16 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 investigated these interactions. The aims were to investigate: (a) the uoride retained in the oral cavity after rinsing with tea, and (b) the association of ingested uoride from tea with the surface of dental enamel and its acquired salivary pellicle using in vitro and in vivo approaches. 2. Materials and methods 2.1. Standardised preparation of tea solutions Tea infusion solutions (1% w/v) were prepared from black tea (World BlendÐTTHRA) using a commercial Bunn-O-Matic tea brewer (Bunn-O-Matic Corp, USA) with 40 gm tea leaves and 4 l double distilled deionised water. Solutions were prepared fresh daily and allowed to cool to room temperature prior to use. The uoride content of each infusion was assessed after preparation with an ion speci c electrode in the presence of TISAB (Total Ionic Strength Adjustment Buffer). The mean (^s.d.) uoride content of these tea infusions was 3.47 (^0.33) ppm Oral retention of uoride after mouth-rinsing with tea in vivo This part of the study aimed to investigate how much uoride was retained in the oral cavity after mouth-rinsing, which would be potentially available for either absorption through the oral mucosa or interaction with the surface of the dentition and its acquired salivary integuments. Eleven healthy volunteers aged 22±23 years with good standards of oral health, acted as experimental subjects. All experiments were performed in the middle of the day to avoid in uence of diurnal variations in salivary ow rate. Subjects avoided using any dental hygiene products, drinking tea or consuming uoride rich foods for 4 h prior to the experiment. Each subject initially rinsed their mouth with 10 ml distilled water for 30 s before discarding the expectorate. A second rinse volume of 8 ml distilled water was measured into a pre-weighed specimen pot and the weight of the pot plus water was recorded. Subjects rinsed for 1 min and the expectorate was collected in the original specimen pot. A third rinse of 8 ml tea (analysed for uoride) was carried out as for the water rinse. After a 15 min break the tea rinse was repeated. Each of the three samples of expectorated solutions were re-weighed and the uoride content determined with an ion speci c electrode in the presence of TISAB. Absolute amounts (mg) of uoride recovered in the postrinse solutions were determined to avoid errors arising from dilution of solutions. Any contribution to recovered uoride from other oral sources was controlled by subtraction of the uoride content of the control pre-tea expectorated solutions from the tea-rinse expectorated solutions In situ study of interaction of uoride from tea mouthrinsing with the tooth surface The studies in Section 1 above provide data on the overall bio-availability of uoride within the oral cavity and this part of the study aimed to examine the proportion of that uoride which can directly interact with the tooth surface both in the presence and absence of an acquired salivary pellicle. Blocks of enamel (approximate surface area 11 mm 2 each), prepared from sterilised extracted human teeth, were attached to partial removable appliances (vacuum formed splints of the dentitions of subjects from Section 1 above). Each enamel block was covered with nail varnish leaving a 10.5 mm 2 enamel window uncovered. Two types of appliances were usedðone covered the maxillary canine and incisor teeth, whilst the other covered the right maxillary pre-molar and molar teeth. The enamel blocks were attached to the palatal surface of the middle third of the maxillary left central incisor or the maxillary right rst permanent molar. Each subject wore the appliances for a period of 3 h, during which dietary contact was avoided, to allow the development of an acquired salivary pellicle on the enamel specimen surface. After pellicle development, an identical enamel block prepared from the adjacent area of the same tooth was attached to the appliance alongside the pellicle covered enamel block. The subject then mouth-rinsed three times with 10 ml samples of tea using the following protocol: 3 min tea rinse, 5 min break, 3 min tea rinse, 5 min break, 3 min tea rinse, 5 min break. The appliance was then removed from the mouth prior to analysis of the specimens. Fluoride loosely associated with the acquired salivary pellicle on one block from each site of each subject and from the specimen surface of the relevant control block was solubilised by wiping the specimen surface twice for 1 min with cotton pellets saturated with approximately 100 ul 2% sodium dodecyl sulphate. Each pellet was then extracted into 0.5 m perchloric acid prior to determination of uoride levels with a uoride ion speci c electrode in the presence of TISAB In vitro binding studies of uoride from black tea with dental enamel Interaction of uoride with dental enamel and its substitution into the hydroxyapatite crystal lattice represents one of the ways in which uoride may be able to reduce the incidence of dental caries. Studies in Section 2 above examined the interaction of uoride from black tea with the acquired salivary pellicle on the surface of the tooth where the uoride would be potentially available both to in uence plaque bacterial metabolism of dietary sugar and to substitute into the hydroxyapatite crystal lattice of enamel. In this part of the study, the direct binding of uoride from tea to dental enamel was examined. An in vitro approach of binding to powdered enamel particles was

3 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 17 adopted to provide the sensitivity for discrimination of bound from endogenous uoride. Human dental enamel was dissected under a stereomicroscope from thick sections (~200 mmðprepared with a cooled rotary diamond-edged saw-blade) of healthy, extracted human teeth after routine prophylaxis with a non- uoride paste. The enamel fragments were powdered in a liquid nitrogen cooled percussion mill and sieved to mesh size 40±60. A chromatography column (1 12 cm) was packed under pressure with a suspension of the powdered enamel particles in double-distilled deionised water and equilibrated with the same eluent. A 10 ml sample of tea solution was applied to the column and eluted with a linear gradient of equal portions of double distilled deionised water and a solution comprising of all of the inorganic components of an arti cial saliva [20] based on the composition of human saliva, followed by isocratic elution with 1 M- and 2 M-NaCl solutions. 1 ml fractions were collected from the column elution and each fraction monitored for absorbance at 370 nm (absorbance maximum for tea solution) and uoride content using a uoride ion speci c electrode in the presence of TISAB. Firmly bound uoride not eluted from the column was investigated by cutting the column into 12 equal slices and assaying the uoride with an ion speci c electrode in the presence of TISAB in sodium acetate buffered (ph 5.2) 1 M perchloric acid solubilised samples of the enamel column slices, which had been ashed overnight at 6508C In vitro competitive binding studies of black tea moities with dental enamel The uoride±enamel binding studies in Section 3 above indicated that there was strong discolouration of enamel after tea binding, which could not be easily dissociated. This implied that a number of constituents of tea may be strongly binding to enamel and in this part of the study, an in vitro competitive binding assay was used to investigate avidity of binding of these constituents to available sites on the enamel. A modi cation of the alizarin red competitive binding assay of Myers [21] was used in which powdered enamel prepared as in Section 3 above was substituted for hydroxyapatite powder. has a high avidity for hydroxyapatite and in these studies, the binding of this dye to the powdered enamel was studied either in the absence or presence of an arti cial acquired salivary pellicle and in the absence or presence of tea solution. The acquired pellicle on the enamel particles was formed by incubation of the enamel particles with freshly collected stimulated whole human saliva at 378C with agitation for 1 h. The binding assays were performed at 378C for 5 min with agitation and the concentration of unbound alizarin red was determined by measurement of the absorbance of the post-binding supernatant solutions at 520 nm. Ten replicate samples were examined for each test condition. Table 1 In situ uoride (mg) retention at the tooth surface in the presence and absence of an acquired pellicle at incisor and molar sites after tea mouthrinsing 3. Results 1 Pellicle (mg F) 2 Pellicle (mg F) Incisor site Mean s.t.d Molar site Mean s.t.d Oral retention of uoride after mouth-rinsing with tea in vivo Analysis of the mean uoride recovered in the tea solutions after mouth-rinsing for all of the subjects indicated that there was 65.93% (s.d ) recovery of uoride after the rst rinse with tea and 66.08% recovery after the second rinse. This represented a mean retention of 9.7 ug uoride after the rst rinse and 9.4 mg after the second rinse for the group of subjects. The mean volumes of rinse recovered for the water and tea rinses were 8.3 and 8.4 ml respectively indicating a small contribution from the oral uids during rinsing. All uoride determinations were performed in absolute amounts (mg) to avoid errors arising from dilution of solutions In situ study of interaction of uoride from tea mouthrinsing with the tooth surface There appeared to be a small amount of uoride associated with the pellicle at both incisor and molar sites after tea mouth-rinsing (Table 1), although there was considerable individual variation between subjects at both sites. Students t-test analysis of the results indicated that the differences between uoride retention at the tooth surface in the presence and absence of a pellicle were not statistically signi cant (p values and for the incisor and molar sites respectively) and neither was the uoride retention in the presence of a pellicle between incisor and molar sites (p ˆ ) In vitro binding studies of uoride from black tea with dental enamel Application of the tea solution resulted in a strong brown staining of the enamel particles in the upper areas of the chromatography column. Pre-gradient elution of the column with water led to elution of either loosely or un-bound material on the basis of release of material absorbing at 370 nm (Fig. 1). Approximately 12.8% of the uoride applied in the tea was eluted in this pre-gradient material. Elution with a gradient of arti cial saliva released some 370 nm absorbing

4 18 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 Fig. 1. Elution pro le of uoride from a column packed with enamel particles after application of a sample of tea solution. material and a gradual leaching of uoride from the column, although only approximately 15% of the uoride applied to the column was eluted by the arti cial saliva. Eluants of greater ionic strength (1 M- and 2 M-NaCl) released little 370 nm absorbing material and only approximately 9.6% of the uoride applied to the column. Thus, only approximately 37% of the uoride applied to the column could be released by elution of solutions of increasing ionic strength. This retention of tea constituents was paralleled by retention of the brown stain in the upper areas of the column after elution. Retention of uoride from tea in the upper areas of the column was con rmed by analysis of the uoride content of the enamel particles derived from the successive slices taken from the column, where the upper areas showed an increased uoride content (Table 2) In vitro competitive binding studies of black tea moieties with dental enamel Table 3 shows the change in alizarin red binding to enamel in the absence or presence of an acquired pellicle and tea. The decreased absorbance of alizarin red solutions (or alternatively, the percentage change in absorbance) is indicative of binding of alizarin red to charged sites on the enamel particles and the lesser decrease in absorbance in the presence of pellicle and/or tea indicates reduced binding of alizarin red to the particles. The rank order of avidity of the competing binding moieties was enamel 1 pellicle 1 tea was greater than enamel 1 tea, which was greater than enamel 1 pellicle; all of these were signi cantly different (p, 0.005) when examined by analysis of variance tests. 4. Discussion Whilst tea has long been recognised as a rich dietary source of uoride, information is lacking on the bioavailability of uoride from tea as a topical agent in the oral cavity. The results of the present study have demonstrated that uoride is retained in the oral cavity

5 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 19 Table 2 Fluoride concentration of ashed enamel particles derived from slices taken through the column Area of column from which enamel derived (cm from top of column) 0± ± ± ± ± ± ± ± ± ± ± ± Control enamel Enamel F concentration (mf/mg ashed enamel) after mouth-rinsing with tea and some of this uoride interacts with the tooth surface and its acquired pellicle integument. This approach was chosen for the study, rather than simply examining salivary uoride levels with time, since it provides direct information on its presence at the tooth surface where any topical action might be exerted. Thus, ingested uoride from tea will potentially have both local topical effects in the oral cavity, and more generalised systemic effects arising from gastrointestinal absorption. Approximately 65% of the uoride was recovered after mouth-rinsing indicating that the remaining 35% was retained in the body. Since all subjects reported minimal swallowing of the tea, it is probable that the majority of this retained uoride was initially bound to the soft- and hard-tissue surfaces in the oral cavity. However, some of this bound uoride may have become subsequently displaced and become available for systemic absorption. Oral absorption of uoride is probably of lesser importance than gastrointestinal absorption, although it does occur [22,23]. Although data are only available from studies on experimental animals, 7% of a dose of 18 F was reported to be absorbed orally during a 2.5 h period [23]. Solution ph and uoride concentration appear to be important determinants of oral absorption in the hamster cheek pouch [24]; however, above ph 5.0 (a similar ph to that of tea) the ph dependence was not apparent. The nature of the uoride binding to the oral tissues is likely to be complex and involve a variety of interactions. The present in situ studies of uoride binding suggest that uoride may interact with components of the salivary acquired pellicle. It seems probable that uoride also interacted with the salivary mucous covering on the oral mucosal tissues. Only limited information exists on the interaction of uoride with salivary components. Birkeland [25] reported that it binds and precipitates with inorganic substances in saliva, probably calcium phosphate. The acquired pellicle and saliva derived mucous coating on the oral mucosa is Table 3 In vitro competitive binding of solutions to enamel Components Absorbance (520 nm) (mean ^ SD) ^ enamel ^ enamel 1 pellicle ^ enamel 1 tea ^ enamel 1 pellicle 1 tea ^ % reduction in absorbance likely to include both salivary (glyco-)proteins and mineral ions and thus, might be expected to provide a suitable substratum for uoride interaction. However, the dynamic nature of the development of these integuments suggests that such interactions will not be static. The present study has not allowed us to discriminate which components of the salivary acquired pellicle that the uoride was associated with. The consensus view is that ionic uoride does not bind to plasma proteins [3] and there appears to be little evidence in the literature for binding to salivary proteins. Thus, it seems probable that the uoride detected was associated with mineral ions. We purposely studied the loosely bound uoride in the present study since this pool of any retained uoride would be readily available for participation in any of its topical effects. There have been many reports of uoride accumulation in dental plaque [26±28] and clearly, uoride can become associated with oral deposits after ingestion. The inverse relationship between plaque uoride levels and caries experience [29,30] demonstrates the potential importance of such association of uoride with oral deposits. It is unclear, however, as to what degree the more insoluble pools of uoride associated with plaque are able to directly in uence the carious process as opposed to the more soluble readily available pools of this ion. The alizarin red binding experiments in the present study have shown that a complex solution like tea can bind strongly with enamel and that with a salivary pellicle covering the enamel, binding still occurs. Whilst some of this may re ect binding of uoride from the tea, it seems probable that other tea constituents are also binding. The latter may contribute to the anecdotal reports of staining of the tooth surface in heavy tea drinkers with poor oral hygiene. Constituents of tea also bind avidly to the surface of enamel in the absence of a pellicle. Some of this represents binding of uoride ions as evidenced by the in vitro column binding experiment, although this latter experiment also demonstrated strong staining of the enamel by other constituents of tea. Speirs [31] observed complex interactions between tea infusions and powdered hydroxyapatite and suggested that the presence of aluminium and manganese in part reduced the uptake of ionized uoride from tea by hydroxyapatite. However, it is probable that a number of

6 20 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 other constituents of tea also participate in these complex interactions. The binding of uoride to enamel was probably due predominantly to its interaction with the mineral phase of the tissue. Ionic substitution of uoride into the hydroxyapatite crystal lattice in skeletal tissues has long been recognised and in the present experiments, probably involved predominantly surface substitution since the system was not undergoing active mineralisation. Our in vitro study of elution of uoride from enamel particles after application of a tea solution demonstrated the high avidity of enamel for uoride after topical application. Only limited dissolution of the uoride was observed after elution with solutions up to the ionic strength of saliva and even solutions of much greater ionic strength were unable to release all of the bound uoride. Thus, under in vivo conditions it is probable that uoride incorporated in this way will only be released during demineralisation of the enamel. It is unclear, however, as to whether reduced enamel solubility after uoride incorporation represents the predominant mechanism in the cariostatic action of uoride [32,33]. Pellicle or plaque-associated uoride may depress plaque bacterial metabolism of dietary carbohydrates to form acids and/or promote enamel remineralisation [34±37]. Thus, teaderived uoride might be expected to exert in uences on the carious process at the tooth surface in several ways. Other components of tea may also contribute to inhibition of caries. It has been reported that the tannins in tea can inhibit salivary amylase thereby reducing the cariogenic potential of starch-containing foods [38]. The results of this study have demonstrated that tea in the diet can provide an effective vehicle for uoride delivery to the oral cavity where it may then become associated with the oral tissues and their surface integuments. Subsequently, the uoride may exert local topical effects although systemic effects arising from oral and gastrointestinal absorption will be expected to contribute to its actions. Acknowledgements We are grateful to the International Steering Committee of the Tea Trade Health Research Association for their support of this study and gratefully acknowledge help with statistical advice from Dr P Murray. References [1] Hodge HC, Smith FA. Occupational uoride exposure. Journal of Occupational Medicine 1977;19:12±39. [2] Smith FA, Hodge HC. Airborne uorides in man. Critical Reviews in Environmental Control 1979;9:1±25. [3] Whitford GM. The metabolism and toxicity of uoride. Basel: Karger, [4] Murray JJ, Rugg-Gunn AJ, Jenkins GN. Fluorides in caries prevention. Oxford: Butterworth-Heinemann, [5] Muhler JC. Ingestion from foods, Fluoride and human health. Geneva: WHO, p. 32±40. [6] Taves DR. Dietary intake of uoride ashed (total uoride) v. unashed (inorganic uoride) analysis of individual foods. British Journal of Nutrition 1983;49:295±301. [7] Cremer HD, Buttner W. Absorption of uorides, Fluoride and human health. Geneva: WHO, p. 75±91. [8] Wei SHY, Hattab FN, Melberg JR. Concentration of uoride and selected other elements in tea. Nutrition 1989;5:237±40. [9] Chan JT, Koh SH. Fluoride content in caffeinated, decaffeinated and herbal teas. Caries Research 1996;30:88±92. [10] Whitford GM. Fluoride in dental products: safety considerations. Journal of Dental Research 1987;66:1056±60. [11] Burt BA. The changing patterns of systemic uoride intake. Journal of Dental Research 1992;71:1228±37. [12] Rock WP, Sabieha AM. The relationship between reported toothpaste usage in infancy and uorosis of permanent incisors. British Dental Journal 1997;183:165±70. [13] SanFilippo FA, Battistone GC. The uoride content of a representative diet of the young male adult. Clinica Chimica Acta 1971;31:453±7. [14] Singer L, Ophaug RH, Harland BF. Fluoride intake of young male adults in the United States. American Journal of Clinical Nutrition 1980;33:328±32. [15] Ophaug RH, Singer L, Harland BF. Estimated uoride intake of 6- month-old infants in four dietary regions of the United States. American Journal of Clinical Nutrition 1980;33:324±7. [16] Ophaug RH, Singer L, Harland BF. Estimated uoride intake of average two-year-old children in four dietary regions of the United States. Journal of Dental Research 1980;59:777±81. [17] Duckworth CS, Duckworth R. The ingestion of uoride in tea. British Dental Journal 1978;145:368±70. [18] Walters CB, Sherlock JC, Evans WH, Read I. Dietary intake of uoride in the United Kingdom and uoride content of some foodstuffs. Journal of the Science of Food and Agriculture 1983;34:523±8. [19] Trautner K, Siebert G. An experimental study of bio-availability of uoride from dietary sources in man. Archives of Oral Biology 1986;31:223±8. [20] Shellis RP. A synthetic saliva for cultural studies of dental plaque. Archives of Oral Biology 1978;23:485±8. [21] Myers HM. Structure-activity relationships (SAR) of hydroxyapatitebinding molecules. Calci ed Tissue International 1987;40:344±8. [22] Gabler WL. Absorption of uoride through the oral mucosa of rats. Archives of Oral Biology 1968;13:619±23. [23] Patten JR, Whitford GM, Stringer GL, Pashley DH. Oral absorption of radioactive uoride and iodide in rats. Archives of Oral Biology 1978;23:215±7. [24] Whitford GM, Callan RS, Wang HS. Fluoride absorption through the hamster cheek pouch: a ph-dependent event. Journal of Applied Toxicology 1982;2:303±6. [25] Birkeland JM. The effect of ph on the interaction of uoride and salivary ions. Caries Research 1973;7:11±18. [26] Jenkins GN, Edgar WM. Distribution and forms of F in saliva and plaque. Caries Research 1977;1:226±42 11 suppl.. [27] Carey C, Gregory T, Rupp W, Tatevossian A, Vogel GL. In: Leach S, editor. Factors relating to de- and re-mineralisation of the teeth, Oxford: IRL Press, p. 163±74. [28] Duckworth RM, Morgan SN, Murray AM. Fluoride in saliva and plaque following use of uoride-containing mouthwashes. Journal of Dental Research 1987;66:1730±4. [29] Agus HM, Schamschula RG, Barmes DG, Bunzel M. Associations between the total uoride content of dental plaque and individual caries experience in Australian children. Community Dentistry and Oral Epidemiology 1976;4:210±6. [30] Agus HM, Un PSH, Cooper MH, Schamschula RG. Ionized and bound uoride in resting and fermenting dental plaque and individual human caries experience. Archives of Oral Biology 1980;25:517±22.

7 A. Simpson et al. / Journal of Dentistry 29 (2001) 15±21 21 [31] Speirs RL. Correlations between the concentrations of uoride and some other constituents in tea infusions and their possible dental caries-preventive effect. Archives of Oral Biology 1983;28:471±5. [32] Chow LC. Tooth-bound uoride and dental caries. Journal of Dental Research 1990;69:595±600. [33] Arends J, Christoffersen J. Nature and role of loosely bound uoride in dental caries. Journal of Dental Research 1990;69:601±5. [34] Cate ten JM, Duijsters PPE. In uence of uoride in solution on tooth demineralization. I. Chemical data. Caries Research 1983;17:193±9. [35] Cate JM, Duijsters ten PPE. In uence of uoride in solution on tooth demineralization. II. Microradiographic data. Caries Research 1983;17:513±9. [36] Silverstone LM. Remineralization and enamel caries: signi cance of uoride and effect on crystal diameter. In: Leach SA, Edgar WM, editors. Demineralization and remineralization of teeth. Oxford: IRL Press, p. 185±205. [37] Tatevossian A. Fluoride in dental plaque and its effects. Journal of Dental Research 1990;69:645±52. [38] Zhang J, Kashket S. Inhibition of salivary amylase by black and green teas and their effects on the intraoral hydrolysis of starch. Caries Research 1998;32:233±8.