Saturation of human saliva with respect to calcium salts

Archives of Oral Biology (2003) 48, 317—322

Saturation of human saliva with respect to calcium salts M.J. Larsena,*, E.I.F. Pearceb a

Royal Dental College, Vennelyst Blvd., DK-8000 Aarhus C, Denmark Dental Research Group, Wellington School of Medicine and Health Sciences, PO Box 7343, Wellington South, New Zealand

b

Accepted 21 December 2002

KEYWORDS Saliva; Calcium phosphates; Solubility

Summary It may be assumed that free ionic concentrations of calcium and phosphate in resting saliva tend to equilibrate with those in plaque fluid, and that salivary data can therefore be used to illustrate chemical conditions in both saliva and plaque. In the present study, salivary data collected from the literature or obtained in our laboratory were used to calculate degrees of super- and undersaturation with respect to apatites, brushite, b-tricalcium phosphate, octacalcium phosphate, calcium carbonate and calcium fluoride in the pH range from 3 to 9. Concentrations of calcium, phosphate, fluoride, carbonate, and background ion strength of resting parotid saliva, resting submandibular saliva, and resting and stimulated whole saliva were entered into a computer program, and curves illustrating saturation in the pH range 3—9 constructed. It was found that oral fluids are supersaturated with respect to apatites above pH 5.3 and with respect to octacalcium phosphate and b-tricalcium phosphate above pH 6. Parotid saliva was undersaturated with respect to brushite whilst submandibular saliva was supersaturated with respect to that salt in the pH range 6—8. Stimulated whole saliva with 25 mmol/l carbonate became supersaturated with respect to calcium carbonate only above pH 7.3, which may explain the absence of this salt in the human oral cavity. To maintain the saturation of oral fluids with respect to calcium fluoride, i.e. to ensure its survival in the mouth required 6 ppm fluoride in the aqueous phase. Therefore, this salt, the outcome of topical fluoride therapy, will inevitably dissolve in the oral fluids. ß 2003 Elsevier Science Ltd. All rights reserved.

Introduction Critical pH in saliva is defined as the pH at which saliva is saturated with respect to enamel apatite. For the experimental determination of this value, Ericsson1 suspended powdered enamel in whole saliva samples whose pH had been adjusted from pH 4.5 to 7.5, and observed in which samples enamel dissolved, in which apatite was formed by crystal growth, and in which neither occurred


Corresponding author. Tel.: þ45-89424093; fax: þ45-86202202. E-mail address: [email protected] (M.J. Larsen).

macroscopically. He arrived at a critical pH close to 5.2. For practical reasons and to operate with a safe value, pH 5.5 has been generally adopted as the critical value below which enamel may dissolve and above which enamel does not dissolve. In the oral cavity a variety of calcium phosphate salts other than low fluoride hydroxyapatite may be found: brushite, octacalcium phosphate, b-tricalcium phosphate, fluorapatite and calcium fluoride may all occur either as transient phases or more or less permanently. Fluorapatite forms in the surface layer of enamel caries lesions, brushite, octacalcium phosphate and b-tricalcium phosphate appear in salivary gland stones2 and in supra- and subgingival

0003–9969/03/$ — see front matter ß 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0003-9969(03)00007-4

318

Table 1

M.J. Larsen, E.I.F. Pearce

Ion concentrations on which the graphs were based.

Saliva

N

pH

Calcium (mmol/l)

Phosphate (mmol/l)

Fluoride (ppma)

0.85 0.23 2.1 0.86 0.46

7.08 1.56 6.0 7.01 4.12

0.05 0.05 0.05

1.11 0.21

3.72 0.73

0.05

Resting Parotidb Submandibularc Whole salivad

22

5.94 0.40 6.4 7.07 0.46

Stimulated Whole salivae

31

7.28 0.21

6

Carbonate (mmol/l)

Ion strength (mmol/l)

3.5 4.0 4.4 2.3

28.8 4.0 35 44.6 11.6

25

45.1 11.4

a

Fluoride concentrations were estimated. . c6 (N and S.D. were not given). d Pearce (unpublished). e 16 . b 16

calculus,3,4 while calcium fluoride occurs as a shortlived product of topical fluoride treatment.5 McCann6 gathered data from the ample literature on saliva and calculated degrees of saturation with respect to the important salts in parotid saliva at its physiological pH. He demonstrated that saliva was supersaturated with respect to hydroxyapatite and fluorapatite, close to saturation with respect to octacalcium phosphate and brushite depending on secretion rate, and invariably unsaturated with respect to calcium fluoride. Under normal conditions the salivary concentrations of many ions remain within narrow limits despite a circadian variation,7 an individual variation8 and a variation due to stimulation.9 The concentrations of four ions, however, vary considerably, namely those of fluoride, sodium, carbonate and hydrogen. The fluoride concentration varies according to previous and recent intake of fluoride.10 In resting saliva the concentration of carbonate is low, around 1—3 mmol/l and pH varies around 5.5—7. At high flow rates, the carbonate concentration may rise to 40 mmol/l and pH may increase to 8.9,11 Further, in plaque, the pH may drop to almost 4,12 giving a pH range in the mouth from 4 to 8. Also, the concentration of sodium increases with increasing flow rate, adding somewhat to the ionic strength of saliva. The free ionic concentrations in plaque fluid tend to equilibrate with those of saliva,13—15 with the exception of that of Hþ produced in plaque by fermentation of carbohydrates. Therefore, a description of the saturation of saliva in the whole pH range observable in saliva and plaque may to some extent indirectly describe that of plaque fluid. The aim of the present study was to describe the saturation of saliva, resting and stimulated, with respect to apatites, brushite, octacalcium phosphate, b-tricalcium phosphate, calcium carbonate and calcium fluoride as a function of pH.

Materials and methods Analytical data on ion concentrations in saliva were obtained from the literature on resting parotid saliva,16 resting submandibular saliva6 and stimulated whole saliva (Table 1).16 A requirement was that the concentrations of calcium, phosphate, pH and ion strength were given for individual samples. The carbonate concentration in the data of Grøn was calculated from the concentration of CaHCO3þ, from the concentration of calcium and the association constant as used by Grøn.16 Data for unstimulated whole saliva (Pearce, unpublished) were obtained from 22 adult subjects who drooled into an ice-cooled collecting tube. Analytical methods were as previously described17 except for Ca (specific ion electrode model MI-600, Microelectrodes Inc., USA) and CO2 (electrode model 95-02, Orion, USA). The mean fluoride concentration in resting saliva in subjects living in a low fluoride area has been found to range from 0.02 to 0.05 ppm.18 Overall, the concentration of fluoride in saliva varies from 0.01 to 0.1 ppm depending on past and recent intake of fluoride.10 Hence, for the calculations below, a fluoride concentration of 0.05 ppm was assumed. The data of Table 1 were entered into a computer program47 (redeveloped from Larsen45) that uses the following constants (37 8C): solubility products (pK) for Ca5(PO4)3OH 58.6;19 Ca5(PO4)3F 60.1;6 CaHPO42H2O 6.6;20 Ca4H(PO4)32.5H2O 48.8;21 bCa3(PO4)2nH2O 29.5;22 CaF2 10.56 and CaCO3 8.2.23 The constants used were: pKw ¼ 13:6 and dissociation constants for: H3PO4/H2PO4 2.2;24 H2PO4/HPO42 7.19;25 HPO42/PO43 12.16;20 H2CO3/HCO3 6.3;26 HCO3/CO32 10.2;26 HF/F 3.25.27 Association constants were for CaH2PO4þ 4.52;20 CaHPO4 4.01;20 CaPO4 35  105 ;28 CaHCO3þ 18.2 and CaCO3 30,300;29 CaOHþ 25.30 The constants for the Debye—Hu ¨ckel equation (log f ¼ AI0:5 = 0:5 ð1 þ BaI Þ) were A ¼ 0:5115, B ¼ 0:329  108 and

Saturation of saliva

the a value for calcium was 6  108 , for fluoride 3:5  108 , for phosphates and carbonates 4:3  108 and for Hþ 8  108 .31 The concentration of the calcium-protein complex in saliva was calculated according to Hay et al.32 and the concentration of free ionic calcium was corrected accordingly. For every 0.2 pH unit in the pH range from 3 to 9, the negative logarithms of the ion activity product (pI) for each salt was calculated, subtracted from its pK and divided by the number of ions (n) in the salt according to the expression: ðpK  pIÞ=n, and for visualization, plotted in a diagram. The expression ðpK  pIÞ=n conveniently allows a comparison of saturation with respect to salts with a different number of ions in the ion product. Positive values indicate supersaturation and negative values undersaturation. The data of Table 1 are average data and the central curves in the figures are based on these average data. The grey zones around the central curves indicate the range of saturation values calculated from the individual ion concentrations for each type of saliva so that all values in the tables from where the data were drawn are covered within the grey zones.

Results For resting parotid saliva the critical pH for hydroxyapatite was 5.2. Below the critical pH the saliva became increasingly undersaturated with respect to hydroxyapatite and above the supersaturation increased steadily (Fig. 1). The critical pH for fluorapatite was 4.7. For b-tricalcium phosphate, octacalcium phosphate and brushite the critical pH values were close together: 6.1, 6.2 and 6.3, respectively. Around and above pH 7 parotid saliva was considerably supersaturated with respect to b-tricalcium phosphate and octacalcium phosphate. At pH 8.6 resting parotid saliva became saturated with respect to CaCO3. It remained unsaturated with respect to CaF2 in the whole pH 3—9 range. In resting submandibular saliva critical pH values were invariably lower than in parotid saliva: For hydroxyapatite it was 5.0 and for fluorapatite 4.2. For octacalcium phosphate, b-tricalcium phosphate and brushite it was 5.6, 5.7 and 5.9, respectively, whilst for CaCO3 it was 7.8 (Fig. 2). Further, submandibular saliva was consistently more supersaturated than was parotid saliva (Fig. 3). For resting and stimulated whole saliva, the diagram is not very different from that for parotid saliva, except for CaCO3 and brushite (Figs 4 and 5). With stimulation, the carbonate concentration increased from a few mmol/l to 25 mmol/l, and

319

Figure 1 Degree of saturation of resting human parotid saliva with respect to various calcium salts as a function of pH. For calculation see text. Positive values indicate supersaturation, negative values undersaturation (FAP: fluorapatite; HAP: hydroxyapatite; b-TCP: b-tricalcium phosphate; OCP: octacalcium phosphate; BSH: brushite).

Figure 2 Degree of saturation of resting human submandibular saliva with respect to various calcium salts as a function of pH. For calculation see text. Positive values indicate supersaturation, negative values undersaturation (FAP: fluorapatite; HAP: hydroxyapatite; b-TCP: b-tricalcium phosphate; OCP: octacalcium phosphate; BSH: brushite).

320

Figure 3 Degree of saturation of resting parotid (P) and of resting submandibular (S) saliva with respect to various calcium phosphates as a function of pH. For calculation see text. Positive values indicate supersaturation, negative values undersaturation (FAP: fluorapatite; HAP: hydroxyapatite; OCP: octacalcium phosphate; BSH: brushite).

M.J. Larsen, E.I.F. Pearce

Figure 5 Degree of saturation of stimulated human whole saliva with respect to various calcium salts as a function of pH. For calculation see text. Positive values indicate supersaturation, negative values undersaturation (FAP: fluorapatite; HAP: hydroxyapatite; b-TCP: b-tricalcium phosphate; OCP: octacalcium phosphate; BSH: brushite).

the critical pH for CaCO3 shifted from 8.6 to 7.3, the shift being due to the increase of the carbonate concentration. Supersaturation with respect to brushite was substantially reduced in stimulated whole saliva.

Discussion

Figure 4 Degree of saturation of resting whole saliva with respect to various calcium salts as a function of pH. For calculation see text. Positive values indicate supersaturation, negative values undersaturation (FAP: fluorapatite; HAP: hydroxyapatite; b-TCP: b-tricalcium phosphate; OCP: octacalcium phosphate; BSH: brushite).

Solubility diagrams such as those presented in this paper show what minerals may form and survive in the mouth and what will most likely dissolve in saliva and disappear. If saliva is considerably supersaturated with respect to a salt it may form and survive, while if it is only slightly but constantly supersaturated the salt may survive for at least some time. If saliva is unsaturated with respect to a salt it will eventually dissolve. Most likely, resting saliva has the greatest importance for the integrity of the teeth as stimulated saliva bathes the mouth for a relatively short time in relation to the many hours with resting saliva. During the long periods of resting saliva the ion concentrations of saliva may equalise with concentrations of fluids elsewhere in the mouth, e.g. in plaque and over calculus. The concentrations of free calcium, fluoride and phosphate in plaque fluid are not very

Saturation of saliva different from those in resting saliva14,15 in contrast to that of Hþ, which shifts up and down as a result of bacterial fermentation. Therefore, we considered it of interest to examine saturation with respect to calcium salts over an extended pH range. It should, however, be realised that when pH in plaque drops, calcium and/or phosphate may be released from complexes and/or solid calculus in the plaque, add to the ion concentrations and thereby move the plaque curves of Figs. 1—5 to the left. It was observed that saliva is supersaturated with respect to calcium carbonate above pH 8.6 in resting saliva and in stimulated above pH 7.3. The considerable undersaturation below these values indicates that if present, calcium carbonate will not survive. To the best of our knowledge, solid calcium carbonate has not been found in the human mouth, unlike that of the dog.33 All variants of saliva were considerably undersaturated with respect to calcium fluoride over the pH range 3—9, which indicates its dissolution in the mouth and explains its disappearance from the teeth in the hours and days after a topical treatment.5,34 To lift the CaF2 curve up to the saturation line (cf. Figs. 1, 2 and 4) a recalculation showed that a concentration of approx. 6 ppm fluoride in resting and stimulated saliva is required. This also indicates that solid CaF2 tends to maintain such a fluoride concentration in its vicinity as long as it survives. The differences of parotid and submandibular saliva should be seen in the light of the recent findings of Sas and Dawes35 who demonstrated that outside the dental arch parotid saliva prevails with a minor contribution from labial glands whilst submandibular saliva dominates inside the arch. Resting parotid saliva was supersaturated with respect to brushite in the pH range from 6.3 to almost 9. Resting submandibular saliva became saturated already at pH 5.9 and was more supersaturated above pH 6. It indicates a greater tendency to formation and survival of brushite where submandibular saliva prevails, i.e. within the lower dental arch.35 However, pH may quite often range below 5.9 in saliva and plaque in most people and therefore, brushite is not likely to survive in the mouth unless special conditions are given, e.g. special food conditions. Brushite has occasionally been found in calculus as a transient component.3,4 Another aspect of the observation that saliva is supersaturated with respect to brushite above pH 6 is that incorporation of solid brushite as an abrasive in dentifrice is unlikely to provide additional ionic calcium and phosphate for remineralisation. The critical pH of octacalcium phosphate and btricalcium phosphate ranged around pH 5.5—6, i.e. somewhat lower than that of brushite and the

321

supersaturation over the critical value increased considerably. Therefore, the formation and persistence of these two salts is more likely. b-Tricalcium phosphate has been found as a frequent component of subgingival calculus and both salts occur as occasional components of supragingival calculus.3,4 These salts appear in the mouth although they are more sensitive to periods with low pH than are apatites. However, in the presence of fluoride, octacalcium phosphate is likely to transform into fluorhydroxyapatite.36 Of most significance is the saturation of saliva with respect to hydroxyapatite, as this is the mineral of the teeth and that found most often in calculus. Critical pH in stimulated whole saliva was 5.2 — 5.3, very close to that found experimentally by Ericsson.1 Due to a higher calcium content in resting submandibular saliva the critical pH there was as low as 5.0. Most supragingival calculus consists of fluorhydroxyapatite with a fluoride content in the western world around 200 ppm in the core and up to 3000 ppm in the outer layers.37,38 The higher supersaturation with respect to fluor- and hydroxyapatite of submandibular saliva (Fig. 3) may play a role in formation of calculus close to the orifices of the submandibular salivary glands. The caries lesion in enamel before cavitation is characterised by a subsurface demineralised area, the lesion body, covered by a 20—50 mm thin fluoride-rich layer.39,46 The formation of lesion body is due to undersaturation with respect to hydroxyapatite of the surrounding aqueous phase. The formation of the surface layer has experimentally been demonstrated to be due to a supersaturation and consequently a concurrent formation of fluorapatite in the outer enamel during dissolution of hydroxyapatite below the critical pH of hydroxyapatite.40—43 The more supersaturated with respect to fluorapatite the aqueous phase is during enamel dissolution the more fluoride-rich mineral is deposited in the surface layer.41—44 Finally, erosion of the teeth occurs when enamel is exposed to an aqueous phase undersaturated with respect to both fluorapatite and hydroxyapatite, both apatites dissolve and no surface layer is formed.40,41 Our study shows that this occurs below pH 4.3—4.5 which is the fluorapatite critical pH.

References 1. Ericsson Y. Enamel apatite solubility. Acta Odont Scand 1949;8(Suppl 3):1—139. 2. Burstein LS, Boskey AL, Tannenbaum PJ, Posner AS, Mandel ID. The crystal chemistry of submandibular and parotid salivary gland stones. J Oral Pathol 1979;8:284—91.

322

3. Jensen AT, Danø M. Crystallography of dental calculus and the precipitation of certain calcium phosphates. J Dent Res 1954;33:741—50. 4. Schroeder HE, Bambauer HU. Stages of calcium phosphate crystallization during calculus formation. Arch Oral Biol 1966;11:1—14. 5. Brudevold F. Fluoride therapy. In: Bernier JL, Muhler JC, editors. Improving dental practice through preventive measures. St. Louis: Mosby; 1974. p. 77—103. 6. McCann HG. Inorganic components of salivary secretions. In: Harris RS, editor. Art and science of dental caries research. New York: Academic Press; 1968. p. 55—73. 7. Ferguson DB, Botchway CA. A comparison of circadian variation in the flow rate and composition of stimulated human parotid, submandibular and whole salivas from the same individual. Arch Oral Biol 1980;25:559—68. 8. Larsen MJ, Jensen AF, Madsen DM, Pearce EIF. Individual variation of pH. Arch Oral Biol 1999;44:111—7. 9. Dawes C. The effect of flow rate and duration of stimulation on the concentrations of protein and the main electrolytes in human parotid saliva. Arch Oral Biol 1969;14:277— 94. 10. Ekstrand J, Spak C-J, Vogel G. Pharmacokinetics of fluoride in man and its clinical relevance. J Dent Res 1990;69: 550—5. 11. Bardow A, Madsen J, Nauntofte B. The bicarbonate concentration in human saliva does not exceed the plasma level under normal physiological conditions. Clin Oral Invest 2000;4:245—53. 12. Nyvad B, Fejerskov O. Development, structure and pH of dental plaque. In: Thylstrup A, Fejerskov O, editors. Textbook of clinical cariology. Copenhagen: Munksgaard; 1996. p. 99—110. 13. Carey CM, Vogel GL. Measurement of calcium activity in oral fluids by ion selective electrode: method evaluation and simplified calculation of ion activity products. J Res Natl Inst Stand Technol 2000;105:267—73. 14. Matsuo S, Lagerlo ¨f F. Relationship between total and ionized calcium concentrations in human whole saliva and dental plaque fluid. Arch Oral Biol 1991;36:525—7. 15. Vogel GL, Zhang Z, Carey CM, Ly A, Chow LC, Proskin HM. Composition of plaque and saliva following a sucrose challenge and use of an a-tricalcium-phosphate-containing chewing gum. J Dent Res 1998;77:518—24. 16. Grøn P. The state of calcium and inorganic orthophosphate in human saliva. Arch Oral Biol 1973;18:1365—78. 17. Poff AM, Pearce EIF, Larsen MJ, Cutress TW. Human supragingival in vivo calculus formation in relation to saturation of saliva with respect to calcium phosphates. Arch Oral Biol 1997;42:93—9. 18. Bruun C, Givskov H. Fluoride concentrations in saliva in relation to chewing various supplementory fluoride preparations. Scand J Dent Res 1979;87:1—6. 19. McDowell H, Gregory TM, Brown WE. Solubility of Ca5(PO4)3OH in the system Ca(OH)2—H3PO4—H2O at 5, 15, 25 and 37.5 8C. J Res Natl Bur Stand 1977;81A:773—82. 20. Gregory TM, Moreno EC, Brown WE. Solubility of CaHPO4, 2H2O in the system Ca(OH)2—H3PO4—H2O at 5, 15, 25 and 37.5 8C. J Res Natl Bur Stand 1970;74A:461—75. 21. Tung MS, Edelman N, Sieck B, Brown WE. Octacalcium phosphate solubility product from 4 to 378C. J Res Natl Bur Stand 1988;93:613—23. 22. Gregory TM, Moreno EC, Patel JM, Brown WE. Solubility of b-Ca3(PO4)2 in the system Ca(OH)2—H3PO4—H2O at 5, 15, 25 and 37.5 8C. J Res Natl Bur Stand 1974;78A:667—74.

M.J. Larsen, E.I.F. Pearce

23. Garrels RM, Tompson ME. A chemical model for sea water at 25 8C and one atmosphere total pressure. Am J Sci 1962; 260:57—66. 24. Bates RG. First dissociation constant of phosphoric acid from 0 to 608C. J Res Natl Bur Stand 1951;47:127—34. 25. Bates RG, Acree SF. H values of certain phosphate chloride mixtures and the second dissociation constant of phosphoric acid from 0 to 608C. J Res Natl Bur Stand 1943;30:129—55. 26. Ellis AJ. The system Na2CO3—NaHCO3—CO2—H2O at temperatures up to 200 8C. Am J Sci 1959;257:287—96. 27. Broene HH, DeVries T. Thermodynamics of aqueous HF solutions. J Am Chem Soc 1947;69:1644—6. 28. Chughtai A, Marshall R, Nancollas GH. Complexes in calcium phosphate solutions. J Phys Chem 1968;72:208—11. 29. Nakayama FS. Calcium activity, complex and ion-pair in saturated CaCO3 solutions. Soil Sci 1968;106:429—34. 30. Davies CW, Hoyle BE. The dissociation constant of calcium hydroxide. J Chem Soc 1951;233—234. 31. Kielland J. Individual activity coefficients of ions in aqueous solutions. J Am Chem Soc 1937;59:1675—8. 32. Hay DI, Schluckebier SK, Moreno EC. Equilibrium dialysis and ultrafiltration studies of calcium and phosphate binding by human salivary proteins. Implications for salivary supersaturation with respect to calcium phosphate salts. Calcif Tissue Int 1982;34:531—8. 33. LeGeros RZ, Shannon IL. The crystalline components of dental calculi: human vs. dog. J Dent Res 1979;58:2371—7. 34. Larsen MJ, Ravnholt G. Dissolution of various calcium fluoride preparations in inorganic solutions and in stimulated human saliva. Caries Res 1994;28:447—54. 35. Sas R, Dawes C. The intra-oral distribution of unstimulated and chewing-gum-stimulated parotid saliva. Arch Oral Biol 1997;42:469—74. 36. Brown WE, Smith JP, Lehr JR, Frazier AW. Octacalcium phosphate and hydroxyapatite. Nature 1962;196:1048—54. 37. Okumura H, Nakagaki H, Kato K, Ito F, Weatherell JA, Robinson C. Distribution of fluoride in human dental calculus. Caries Res 1993;27:271—6. 38. Huang S, Nakagaki H, Okumura H, et al. Fluoride distribution in human dental calculus obtained from different sites on the tooth surface. J Periodont Res 1996;31:149—56. 39. Pearce EIF, Coote GE, Larsen MJ. The distribution of fluoride in carious human enamel. J Dent Res 1995;74:1775—82. 40. Larsen MJ. Dissolution of enamel. Scand J Dent Res 1973; 81:518—22. 41. Larsen MJ. Chemically induced in vitro lesions in dental enamel. Scand J Dent Res 1974;82:496—509. 42. Pearce EIF. A microradiographic and chemical comparison of in vitro systems for the simulation of incipient caries in abraded bovine enamel. J Dent Res 1983;62:974—89. 43. ten Cate JM, Duijsters PPE. Influence of fluoride in solution on tooth demineralization. II. Microradiographoic data. Caries Res 1983;17:513—9. 44. Margolis HC, Moreno EC, Murphy BJ. Effect of low levels of fluoride in solution on enamel demineralization in vitro. J Dent Res 1986;67:23—9. 45. Larsen MJ. An investigation of the theoretical background for the stability of the calcium phosphate salts and their mutual conversion in aqueous solutions. Arch Oral Biol 1986;31:757—61. 46. Pearce EIF, Larsen MJ, Coote GE. Fluoride in enamel lining pits and fissures of the occlusal groove-fossa system in human molar teeth. Caries Res 1999;33:196—205. 47. Larsen MJ. Ion products and solubility of calcium phosphates. Aarhus: Royal Dental College; 2001.