EFFECTS OF HEAT AND ORGANIC MATRIX ON ENAMEL DEMINERALIZATION AND DIFFUSION HUANG LI B.D.S A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE DEPARTMENT OF PREVENTIVE DENTISTRY NATIONAL UNIVERSITY OF SINGAPORE 2006 i ACKNOWLEDGEMENTS I wish to express my deepest gratitude to my supervisor Prof Hsu Chin-Ying Stephen, whose academic and professional motivation stimulated my continued pursuing of knowledge on this journey. I am greatly indebted to my family for their understanding, spirit and financial support, especially my mother, for her attention devoted to my son during my long and frequent absences from home. I would like to thank my colleagues in the cariology lab, Ms Liu Yuanyuan, Ms Deng Ying, Dr Gao Xiaoli, for their invaluable help during this working, and lab technician, Mr. Chan, for his patient technical support in the use of equipments. Special thanks go to Dr Nyi Lay Maung, who cheerfully endured my numerous lengthy discussions and conversations with patient reply. I would also like to thank the following individuals for their help and assistance with professional information: Dr Deng Xudong, Department of English, has not only helped me revise this thesis, but also teach me writing skills; Prof Thorsten Wohland, Department of Chemistry, has kindly guided me to acquire new knowledge and handle equipments; Dr Chan Yiong Huak, Biostatistics Unit, has dedicated to instruct me to do the statistical analysis. Finally, financial assistance provided by the Faculty of Dentistry, National University of Singapore in the form of research scholarship toward the completion of the research is thankfully acknowledged. ii TABLE OF CONTENTS ACKNOWLEDGEMENTS .......................................................................... i TABLE OF CONTENTS ............................................................................ iii SUMMARY ................................................................................................ viii LIST OF FIGURES ...................................................................................... x LIST OF TABLES ...................................................................................... xii LIST OF ABBREVIATION ..................................................................... xiii CHAPTER I INTRODUCTION ................................................................ 1 CHAPTER II LITERATURE REVIEW ................................................... 5 2.1 Human Dental Enamel.............................................................................................. 6 2.1.1 Structure of Enamel ........................................................................................... 6 2.1.1.1 Enamel Rod................................................................................................. 6 2.1.1.2 Enamel Crystals .......................................................................................... 7 2.1.1.3 Enamel Tufts, Lamellae, Spindles and Cracks ........................................... 9 2.1.1.4 Striae of Retzius........................................................................................ 10 2.1.1.5 Hunter-Schreger Bands............................................................................. 10 2.1.1.6 Dentino-Enamel Junction (DEJ) ............................................................... 10 2.1.1.7 Water Structure of Enamel........................................................................ 11 2.1.1.8 Porosity ..................................................................................................... 11 2.1.2 Chemical Properties ......................................................................................... 13 2.1.2.1 Inorganic components............................................................................... 13 iii 2.1.2.2 Organic components ................................................................................. 14 2.2 Heat-Induced Effect on Human Enamel ................................................................. 15 2.2.1Optical, Morphological, and Crystallographic Changes................................... 16 2.2.2 Chemical Reaction ........................................................................................... 18 2.2.3 Heat-Induced Reduction of Enamel Demineralization .................................... 19 2.3 Caries Formation and Progression.......................................................................... 22 2.3.1 Introduction...................................................................................................... 22 2.3.2 Protective Role of Sound Enamel Surface Chemistry ..................................... 22 2.3.2.1 Fluoride, Carbonate and Magnesium........................................................ 23 2.3.2.2 Organic Matrix.......................................................................................... 23 2.3.2.3 The pH value............................................................................................. 24 2.3.3 In vitro Demineralization ................................................................................. 25 2.3.4 Polarized Light Microscopy (PLM) and structure of carious enamel.............. 27 2.3.4.1 The Translucent zone................................................................................ 29 2.3.4.2 The Dark zone........................................................................................... 30 2.3.4.3 The Body of lesion.................................................................................... 30 2.3.4.4 The Surface zone....................................................................................... 30 2.4 Diffusion Phenomena of Enamel............................................................................ 31 2.4.1 Diffusion Phenomena and Diffusion Pathways ............................................... 31 2.4.2 The Published DC of Enamel .......................................................................... 32 2.4.3 Factors Affecting Enamel Diffusion ................................................................ 33 2.4.3.1 Pore Size ................................................................................................... 33 2.4.3.2 Charge of Surface Enamel ........................................................................ 33 iv 2.4.3.3 Organic Matrix in Enamel ........................................................................ 33 2.4.3.4 Chemical Reaction .................................................................................... 34 2.5 Confocal Laser Scanning Microscopy (CLSM) and Fluorescence Application on Dental Enamel............................................................................................................... 34 2.5.1 Introduction...................................................................................................... 34 2.5.2 Confocal Laser Scanning Microscopy ............................................................. 35 2.5.2.1 Principles and Theoretical Background .................................................... 35 2.5.2.2 Basic Structure .......................................................................................... 39 2.5.2.3 Application of CLSM in Dentistry ........................................................... 41 2.6 Fluorescence Recovery After Photobleaching (FRAP) .......................................... 42 2.6.1 Introduction...................................................................................................... 42 2.6.2 Basic Principles of Fluorescence ..................................................................... 42 2.6.3 Autofluorescence and Photobleaching in Dental Enamel................................ 45 2.6.4 Fluorescent Dyes Commonly Used for Dental Enamel ................................... 45 2.6.5 FRAP Principles and Applications .................................................................. 48 2.6.5.1 Two-Dimensional Model .......................................................................... 50 2.6.5.2 Three-Dimensional Model ........................................................................ 51 CHAPTER III MATERIALS & METHODS .......................................... 53 3.1 Overview................................................................................................................. 54 3.2 Sample Preparation ................................................................................................. 54 3.2.1 Tooth Selection and Preparation ...................................................................... 54 3.2.2 OM Extraction ................................................................................................. 54 3.2.3 Enamel Sections Preparation ........................................................................... 58 v 3.2.4 Heat Treatment................................................................................................. 58 3.2.5 Demineralization Process................................................................................. 58 3.3 Measurement of Lesion Depth (LD)....................................................................... 59 3.4 Measurement of DC................................................................................................ 60 3.4.1 Labeling Samples............................................................................................. 60 3.4.2 Theoretical Model For FRAP .......................................................................... 61 3.4.3 FRAP Procedures............................................................................................. 61 3.4.4 Data Collection and Analysis........................................................................... 64 3.5 Measurement of Birefringence ............................................................................... 65 3.6 Statistical Analysis.................................................................................................. 65 3.6.1 Evaluation of OM Extraction and Heat Treatment Effects on LD .................. 65 3.6.2 Evaluation of OM Extraction and Heat Treatment Effect on DC.................... 65 3.6.3 Evaluation of OM Extraction and Heat Treatment Effect on Birefringence ... 66 CHAPTER IV RESULTS .......................................................................... 67 4.1 Characterization of Lesion Depth (LD) using PLM ............................................... 68 4.1.1 Qualitative Results ........................................................................................... 68 4.1.2 Quantitative Results ......................................................................................... 68 4.2 Characterization of DC by FRAP/CLSM ............................................................... 72 4.2.1 Qualitative Characterization of Enamel........................................................... 72 4.2.2 Quantitative Characterization of Enamel......................................................... 75 4.3 Characterization of Birefringence under PLM ....................................................... 80 4.3.1 Qualitative Characterization of Enamel........................................................... 80 4.3.1.1 Stereomicroscopy (SM) ............................................................................ 80 vi 4.3.1.2 Polarized Light Microscopy (PLM).......................................................... 80 4.3.2 Quantification of Birefringence of Enamel...................................................... 84 CHAPTER V DISCUSSION ..................................................................... 87 5.1 Effect of Heat and OM on Enamel Demineralization............................................. 88 5.2 Effect of Heat and OM on Enamel DC................................................................... 90 5.2.1 Heat Effect ....................................................................................................... 91 5.2.2 OM Effect ........................................................................................................ 92 5.3 Effect of Heat and OM on Enamel Birefringence .................................................. 93 5.4 Summary................................................................................................................. 95 5.5 Confounding factors and Future Studies ................................................................ 98 CHAPTER VI CONCLUSSION ............................................................. 100 6.1 Conclusion ............................................................................................................ 101 REFERENCES.......................................................................................... 102 APPENDIX A ............................................................................................ 114 APPENDIX B1 .......................................................................................... 116 APPENDIX B2 .......................................................................................... 117 APPENDIX B3 .......................................................................................... 118 vii SUMMARY The reduced subsurface enamel demineralization caused by heat and the laser treatments have been studied in the past few decades. But the mechanism was not completely clear. The purpose of this experiment was to investigate the changes of the birefringence, ion diffusion, and lesion formation in enamel with & without organic matrix following the temperature increment during heating. Seven sound teeth were cut into halves and prepared into two groups: the normal and Organic Matrix (OM)-extracted groups. Six sound sections were chosen from each tooth half and heated in the temperature range of 100℃ to 500℃, with one section kept at room temperature as the control group. Their birefringences before and after heating were measured using Polarized Light Microscope (PLM). Thereafter, these sections were cut into two segments — the coronal and the cervical. The coronal segments were subjected to the measurement of Diffusion Coefficient (DC) quantitatively using Flurescence Recovery After Photobleaching (FRAP) coupled with Confocal Laser Scanning Microscpe (CLSM). The cervical segments were subjected to demineralization treatment and the Lesion Depth (LD) was evaluated by Polarized Light Microscope (PLM). The OM-extracted groups showed an increased LD, DC, and birefringence compared with those of the normal group. These results confirmed the previous assumption that the organic matrix has an inhibitory effect on the ion diffusion and demineralization in the enamel. LD reached its minimum at 300℃ in both normal and OM-extracted groups. Cracks appeared in surface enamel when temperature was above 300 ℃ . Therefore this viii temperature may be the limit of heating. There was an abrupt decrease of DC in the normal group in the temperature range 300℃ to 400℃. But not for the OM-extracted group. The DC continued to decrease in the temperature range 400℃ to 500℃ in both groups. This finding suggested that the DC decrease was not just caused by the decrease of permeability in the enamel after heating, especially above 300℃. It could also have been affected by the inorganic changes in the enamel. The increased birefringence in both groups indicated the increased space in both heated and OM-extracted enamel. The birefringence increase in the OM-extracted group at 200 ℃ was more than that of the normal group. The results suggested that the micropores in the enamel were related to the organic matrix. In conclusion, the reduced demineralization induced by heat was attributed to the mixed effect of both permeability reduction and compositional changes in enamel. The organic matrix has the retardation effect on the diffusion of ions in enamel. The heating in the temperature range 200 ℃ to 300 ℃ was considered beneficial to the prevention of incipient caries. The FRAP technique provided a novel approach to study ion diffusion in hard tissue. ix LIST OF FIGURES Figure 2-1. Crystal structure of hydroxyapatite.................................................................. 8 Figure 2-2. Schematic model of subsurface demineralization.......................................... 27 Figure 2-3. Diagram of the retardation formation ............................................................ 28 Figure 2-4. Schematic structure of a CLSM ..................................................................... 35 Figure 2-5. Diagram of the 3D-PSF.................................................................................. 37 Figure 2-6. Section through the 3D-PSF........................................................................... 38 Figure 2-7. The convolution and deconvolution used in the theoretical 3D-PSF............. 38 Figure 2-8. Diagram of the optical slicing in a specimen (www.zeiss.com) .................... 39 Figure 2-9. Basic structure of CLSM................................................................................ 39 Figure 2-10.Wavelength of laser spectrum....................................................................... 40 Figure 2-11. Jablonski diagram illustrating fluorescence ................................................. 43 Figure 2-12. Stokes shift ................................................................................................... 44 Figure 2-13. Optical spectra of fluorescein....................................................................... 47 Figure 2-14. Ionization equilibria of fluorescein .............................................................. 48 Figure 3-1. Flowchart of experimental procedures........................................................... 55 Figure 3- 2. OM extraction procedures............................................................................. 57 Figure 3-3. Coronal and cervical segments of one tooth section...................................... 59 Figure 3-4. Lesion Measurement ...................................................................................... 60 Figure 3-5. Picture of ROI under CLSM .......................................................................... 62 Figure 3-6. Diagram of FRAP setup in tooth section ....................................................... 63 Figure 3-7. The recovery curve from the image sequence................................................ 64 x Figure 4-1. Representative PLM photograph of lesions after demineralization............... 68 Figure 4-2. LD of two groups with heat treatment at different temperature..................... 69 Figure 4-3. Representative CLSM image of ROI in enamel ............................................ 73 Figure 4-4. Reprehensive T-series images of FRAP under CLSM .................................. 74 Figure 4-5. Fluorescent intensity curve ............................................................................ 75 Figure 4-6. DC changes of normal and OM-extracted groups after heat treatment ......... 77 Figure 4-7. Enamel sections after heat treatment under SM............................................. 81 Figure 4-8. Enamel sections after heat treatment under PLM (lower magnification 2X) 82 Figure 4-9. Respective pictures under PLM of enamel after heating (highrt magnification 10X) .......................................................................................................................... 83 Figure 4-10. Birefringence changes after OM-extracted and heat treatment.................... 85 xi LIST OF TABLES Table 2-1: Summary of the heating effect ........................................................................ 21 Table 4-1. Comparison between normal and OM-extracted groups................................. 69 Table 4-2. Percentages of LD reduction ........................................................................... 70 Table 4-3. Statistical significance of Tests Between-Subject Effects of LD.................... 71 Table 4-4. Pairwise comparisons in Post-hoc Tests of LD.............................................. 71 Table 4-5. Pairwise comparisons of Post-hoc Test in subgroups of LD........................... 72 Table 4-6. Means and standard deviations of K, τ and f in FRAP model....................... 76 Table 4-7. DC with heat treatment in normal and OM-extracted groups ......................... 76 Table 4-8. Percentages of DC reduction ........................................................................... 78 Table 4-9. Statistical results of Tests of Between-Subject Effects of DC ........................ 78 Table 4-10. Multiple Comparisons of Post hoc Test of DC ............................................. 79 Table 4-11. Pairwise Comparisons of Post-hoc Test in subgroups of DC........................ 79 Table 4-12. Raw Birefringence data in normal and OM-extracted groups...................... 84 Table 4-13. Statistical results of Tests Between-Subject Effects of birefringence........... 84 Table 4-14. Heating effect on birefringence .................................................................... 85 Table 4-15. Statistical results of heat-induced birefringence change ............................... 86 xii LIST OF ABBREVIATION AU Airy Unit BF Birefringence CLSM Confocal Laser Scanning Microscopy CaF2 Calcium Fluoride DC Diffusion Coefficient DCPD Dicalcium phosphate dihydrate DELF Dye-enhanced laser fluorescence DEJ Dentino-Enamel Junction f Fraction of recovery FITC Fluorescein Isothiocyanate FRAP Flurescence Recovery After Photobleaching FWHM Full Width at the Half Maximum Height GFP Green Fluorescent Protein HAP Hydroxyapatite HPO42- Acid phosphate K Leaching parameter LD Lesion Depth MRG Microradiography N.A. Numerical Aperture OM Organic Matrix PBS Phosphate Buffered Saline xiii PLM Polarized Light Microscope PMT Photomultiplier PSF Point Spread Function P2O74- Pyrophosphate R Retardation ROI Region of interest RI Refractive Indices SEM Scanning Electron Microscopy SM Stereomicroscopy t Thickness of section TEM Transmission Electron Microscopy β-TCP βeta-Tricalcium Phosphate 3-D Three-dimension 2-D Two-dimension τ Characteristic diffusion time xiv CHAPTER I INTRODUCTION 1 The caries prevalence and caries experience have been improved significantly in the past century. This is attributed to the development of oral health and the use of fluoride in water and oral health products. In recent years, early dental caries has attracted increased attention in the dental community. Diagnosis and treatment of the early caries would prevent the development of caries, and greatly improve people’s living quality. The primary caries preventive agents have been fluoride and fissure sealants. Recently, with the development of the laser technique, laser has been used in the research and treatment of caries. The first laser application in dentistry was reported by Stern and Sognnae (1964) and Goldman et al. (1964). Several early researchers (Stern, 1969; Stern and Sognnaes, 1972) reported that laser irradiation might increase caries resistance of enamel and has the potential capability in cares prevention. Several types of lasers including CO2 laser (Stern et al., 1972; Featherstone, et al., 1998), Nd-YAG laser (Yamamoto and Ooya, 1974), Argon laser (Oho and Morioka, 1990), and Ho:YLF laser (Bachmann et al., 2004) have been studied and were reported to have the effect of increased acid resistance and reduced subsurface demineralization in the enamel. However, the mechanism of laser-induced physical and /or chemical changes that cause the reduced demineralization is not clear. Stern et al. (1966) and Yamamoto and Sato (1980) attributed this reduction to the reduced permeability of enamel. But Borggreven et al. (1980) suggested that the reduced rate was caused by the chemical modifications. Fowler and Kuroda (1986) suggested that the formation of pyrophosphate might reduce the enamel solubility. Nelson et al. (1986) suggested that the inhibitory effect of laser was probably due to a combination of surface sealing, compositional changes, and effect on organic matrix. As the laser-irradiated enamel would normally have a temperature 2 gradient that decreases towards the DEJ, identification of the changes along this temperature gradient in tooth enamel will help us to understand the inhibitory mechanism on subsurface demineralization. As for the development of a subsurface cariogenic lesion, there exist two processes. First, enamel apatite crystals are dissolved by the acid ions. Second, the dissolved ions such as calcium, phosphate ions and hydrogen ions would diffuse out and into enamel, respectively, through an apparently intact surface (Moreno and Zahradnik, 1974). Several studies have revealed the compositional, structural, and phase changes of heated enamel (Holcomb and Young, 1980; Kuroda and Fowler, 1984; Palamara et al., 1987), and the changes of lesion depth (LD) (Sato, 1983). The minimum LD was found to occur at 300 ℃. There are relatively few studies to identify the diffusion phenomenon in enamel (Featherstone and Rosenberg, 1984), since it is a complicated process. Several methods have been used to study this process including diaphragm cell method (Moreno and Burke, 1974), penetration profile study (de Rooij et al., 1980), and conductometry measurement (Scholberg et al., 1984). These methods are time-consuming and difficult to reproduce. With the development of Confocal Laser Scannng Microscopy (CLSM) coupled with Fluorescence Recovery After Photobleaching (FRAP), it becomes straightforward to measure intracellular ions diffusion in the biomedical field. But there has been no application in the dental hard tissue until recently. In this experiment, the quantitative measurement of diffusion coefficient (DC) and birefringence in enamel would be carried out. Simultaneously, the Organic Matrix (OM) 3 and diffusion changes after heating and their roles in reducing the subsurface demineralization would be explored. The main objective of this study is to quantitatively evaluate the effect of temperature and /or organic matrix on the carious-like lesion formation, diffusion coefficient (DC) and birefringence of tooth enamel. 4 CHAPTER II LITERATURE REVIEW 5 2.1 Human Dental Enamel Enamel is the most highly mineralized tissue in human body. By weight, it consists of 96% of mineral, 4% of organic material and water (ten Cate, 1985). Embryologically, enamel is derived from cells of the oral epithelium. This cell, called ameloblast, is quite different from the internal dental epithelium. Anatomically, enamel is the outer layer of tooth structure, coving the anatomic crown of the tooth. Functionally and physiologically, enamel and its anatomical configuration provide a durable surface for tearing and chewing of food. They also help protect the underlying tissues, i.e. dentin and pulp. 2.1.1 Structure of Enamel 2.1.1.1 Enamel Rod The enamel prism or rod is the basic structural unit of enamel. In permanent teeth, its size ranges from 4 to 7 micrometer in diameter (Gwinnett, 1992). Initially enamel rods originate from the region that is quite close to the dentino-enamel junction (DEJ). Then they decussate into the two-thirds of the enamel and finally arrive at their parallel alignment in the outer third of the enamel. Osborn (1965) stated that this decussation produces an optical artifact that is known as Hunter-Shreger bands. The parallel rods are slightly oblique to a tangent of the natural surface of enamel. In the outmost side of permanent teeth, there are rodless enamels, usually in the pit, fissure and cervical regions (Gwinnett, 1967). In the cross sections, the prisms appear somewhat like “keyholes”. Usually the head of prism is oriented to the occlusal surface of the tooth and its tail toward the cervical region of the cross section (Meckel et al., 1965). 6 Within a prism, crystallites of hydroxyapatite are preferentially arranged (Schmidt and Keil, 1958). In the center of the prism, the crystals are tightly packed compared to those in the periphery and interprism enamel. Around the periphery of each prism there is a zone with relatively high organic contents where crystals are oriented in a different direction from the central axis of rod. The zone is known as rod sheath or interprismatic substance, and is believed to have the function of holding the rods together. 2.1.1.2 Enamel Crystals Crystallites of hydroxyapatite occupy 80-90% of the overall volume of the enamel. The remaining 10-20% consists of fluids and organic, usually proteinaceous materials (Robinson et al., 1971). Within a crystal, by weight, 37% is calcium, 52% is phosphate, and 3% is hydroxyl. The size of a crystal is approximately 0.03 by 0.04 by 0.2µm (Larsen and Bruun, 1986). They are the main compositions of apatite and are called major elements. Their relative amounts are quite stable in the enamel. In contrast to these major components, the minor components are scarcely distributed evenly through the enamel. Some of the components, e.g. fluoride, zinc, and lead, show a high concentration in the surface layers below which the concentration drops dramatically. Some components, e.g. sodium, carbonate, and magnesium, exhibit a reverse gradient. Still there are components, e.g. strontium and copper, are not affected by the depth (Larsen and Bruun, 1986). Each crystal has a long axis, called crystal c-axis or fiber axis in shape (Thewlis, 1940). It usually sits in parallel with the direction of the prisms, but has a tendency to deviate from the prism axes from the cusp to the cervical margin (Poole and Brooks, 1961). The carbonated-apatite crystals have a trend to extend from the dentine toward the enamel surface (Johansen, 1965). 7 The stoichiomatric formula for hydroxyapatite is Ca10 (PO4)6(OH) 2 (Kay et al., 1964). This is the unit cell that repeats in all directions to form the single enamel crystal that is approximately 50 nm wide by 25 nm thick, extending from the dentin toward the enamel surface (possibly up to 1mm) (Johansen, 1965). Robinson et al., (2000) described this stoichiomatric structure in arrangement of main ions of the crystals (Figure 2-1). The following statements are excerpted from his descriptions: The stoichiometric structure is most easily appreciated by a consideration of the arrangement of ions around the central hydroxyl column, which extends in the caxis direction through the long axes of the crystals. In the plane of the diagram, the hydroxyl ion is enclosed by a triangle of calcium ions (calcium II). This is in turn surrounded by a triangle of phosphate ions rotated out of phase by 60°. These triangles are in turn surrounded by a hexagon of calcium ions (calcium I). The entire crystal structure can be envisaged as a series of such hexagonal plates stacked one on top of another, each rotated 60° in relation to its immediate neighbors. Ca I Ca I P Ca II Ca II OH Ca I Ca I P P Ca II Ca I Ca I Figure 2-1. Crystal structure of hydroxyapatite Note: The overall planar hexagonal nature of the arrangement of calcium and phosphate ions around the central c-axis hydroxyl column can be seen (Robinson et al., 2000) However, there are a number of variations on this theme in enamel crystal structure. Such variations include missing ions, particularly calcium and hydroxyl, which dissolve from 8 the crystals and diffuse into surrounding fluid. At the same time, extraneous ions such as carbonate, fluoride, sodium, and magnesium are frequently found to precipitate within the crystal structure of surface enamel (Robinson et al., 2000). Such defects and substitutions do have a profound impact on the behavior of apatite, especially with regard to its solubility at low pH. In other words, it has a very close relationship with the enamel dissolution and initiation of the incipient caries. 2.1.1.3 Enamel Tufts, Lamellae, Spindles and Cracks In ground sections, both enamel tufts and lamellae are best demonstrated. Enamel tufts extend from the DEJ to the enamel in a short distance. These lower-mineralized tufts appear to be branched and contain greater concentrations of enamel protein than the rest of enamel. Lamellae project from the surface of enamel to the deeper enamel, consisting of linear, longitudinally oriented defects, which are filled with enamel protein or organic debris from the oral cavity. Origination of enamel tufts is the result of abrupt changes in the direction of packed rods during development. But the reason for the development of lamellae could be related to blocking of the relief of enamel internal strains produced by dimensional changes during enamel maturation. Another reason for the high concentration of organic contents in tufts and lamella is likely due to faulting of blocking of the exit for enamel protein after maturation. Enamel spindles are formed by some newly formed odontoblast processes that are trapped between adjoining ameloblast when enamel formation begins (ten Cate, 1985). 9 2.1.1.4 Striae of Retzius As the human enamel precipitates at a rate of approximately 4 µm per day, there appear periodic bands or cross striations, occurring 4 µm intervals across rods, on the ground enamel sections. Therefore, the striae of Retzius are actually incremental lines of this daily growth in enamel. They appear as concentric growth rings running from DEJ toward the occlusal surface and appear brownish under illuminating light in longitudinal sections. Accentuated incremental lines are produced by systemic disturbances, such as fever and nutritional changes. The lines’ surface manifestation is called Perikymata (Boyde, 1997). 2.1.1.5 Hunter-Schreger Bands The Hunter-Schreger bands are an optical phenomenon, not a real structure of the enamel. Under incident of polarized light, there appear dark and light alternating band in the inner four fifths of the enamel in the longitudinal ground sections. It was assumed that HunterSchreger bands are formed due to the changes of prism orientation. When groups of prism are cut transversely, they are known as diazones; those, cut more longitudinally, are known as parazones. They can also be reversed by an alteration of the direction of the incident illumination. 2.1.1.6 Dentino-Enamel Junction (DEJ) The DEJ is a junction area between enamel and dentine. It can be easily seen as a series of depressions or concaves towards surface of enamel. In scanning electron microscopy (SEM), the DEJ shows as a series of ridges that increase the surface area and probably enhance adhesion between enamel and dentine. 10 2.1.1.7 Water Structure of Enamel The water content in enamel has been reported around 12% by volume (Carlstrom et al., 1963) and 1 ~ 6% by weight (Brudevold et al., 1960). The water content appears to vary with the relative location of the enamel in the tooth and may also vary with age and tooth type. The amount of water in enamel could be reflected by the magnitude of the form birefringence (Carlstrom and Glas, 1963). Normally enamel has a negative birefringence under polarized light investigation. Angmar et al., (1963) found that after heating or drying, enamel shows a positive birefringence. It was believed that water was lost and air was substituted into the submicroscopic spaces. It increased the form birefringence and compromised the intrinsic birefringence and eventually changed the observed birefringence from negative to positive. Little et al., (1962) found that water was bound to enamel in two different ways - loosely and firmly. He also demonstrated that the firmly bound water occupied a greater part, associated with minerals of enamel. Carlstrom et al. (1963) stated that the ratio of loosely and firmly bound water should be around 1:4. The loosely bound water, at least part of it, was related to the organic matrix whereas the firmly bound water was related to the mineral phase. The evidence for the relationship of the water and organic matrix is that the enamel completely loses the ability of reimbibition in water at 200℃, while the organic matrix begins to carbonize in 200℃, and both water and organic matrix increase from the surface to DEJ (Carlstrom and Glas, 1963). 2.1.1.8 Porosity The crystal arrangement of enamel gives rise to two main categories of porosity, corresponding to the spaces between single crystals in the prism and those between 11 prisms. Moreno and Zahradnik (1973), using the isothermal water vapour sorption technique, have demonstrated that the intact enamel has a bimodal pore volume distribution. They also stated that the larger pores are related to the interprismatic regions and the smaller ones are probably associated with the intraprismatic spacing of dental enamel in this geometrical model. The bimodal distribution in the core body has peaks at 9 to 25 Å. A few other researchers have stated that the enamel possess a range of pores diameters from 30 to 140 Å (Ying et al., 2004) or 10 to 250 Å (Medema and Houtman, 1969). Orams et al. (1974) showed that the two dominant pores have sizes in a range from 1 to 10nm, observed by SEM. All these range coincided with each other. The pore size of mature enamel is very small. The function of the pores was described in different ways such as “a molecular sieve” (Darling et al., 1961 and Poole et al., 1961), a semipermeable and osmotic membrane (Atkinson, 1947). So only air and water (Angmar et al., 1963), methanol (Darling et al., 1961), and molecules and ions of the same order of magnitude (Fosdick and Hutchinson, 1965) could be diffused into these microscopic spaces of mature enamel. However, Moreno and Zahradnik (1973), with the help of vapour uptake technique, suggested that dental enamel did not contain micropores and the sieve behavior was probably due to the presence of pore constrictions that in turn are related with the organic matter. Brudevold et al. (1960) found that the enamel sorption capacity increased from the enamel surface to near the DEJ, which probably reflected the fact that the organic matrix increased from surface to DEJ as well. So Brudevold used a simple experiment to prove this assumption: the low-temperature (100℃, carbonization was observed as the blackening of tissue and smoke escape from the tissue. And finally, at >300℃, melting occurred in the hard tissue. 16 These observed changes were related to the interaction between laser-induced heating effect and hard tissue. In addition, there are some techniques used to observed different enamel changes with different temperatures. Firstly, Sato (1983) and Palamara et al. (1987) used polarized light microscopy (PLM) in their studies. They observed the heat-induced birefringent changes with different temperatures. In their findings, between the temperature ranges of 100℃ to 300℃, the first thermal change was seen in the inner region of the enamel. At 300℃, the positively birefringent region extends to the middle enamel and outer enamel shows a bluish color. At 400℃, the entire enamel showed positive birefringence with slight opacity. At 600℃, all enamel became completely opaque. Secondly, Sato (1983) used scanning electron microscope (SEM) to observe the changes of heated enamel. Below 300℃, it showed there were no noticeable changes in the samples. At 400℃, enamel crystals are sharply outlined and there were dotty microspores. At 500 ℃ , a great number of pores appeared and some of them were enlarged. Furthermore, fusion between neighboring enamel crystals was clearly observed. Thirdly, Sakae (1988) used x-ray diffraction to study the changes of crystallites of enamel. From 200℃ to 400℃, crystallites in enamel initially became smaller along the hexagonal a-axis direction. Beyond 400℃, these crystallites grew bigger. That was quite consistent with the observations made by of LeGeros et al. (1978) and Young and Holcomb (1984). In these latter studies, they also showed crystallites gradually became bigger along the caxis (beyond 240℃). 17 Lastly, transmission electron microscope (TEM) was used by Palamara et al. (1987). His study showed inter- and intra-crystalline voids formation in heated enamel at temperatures as low as 200℃. Between 200℃ and 350℃, the negative birefringent regions of the near surface enamel showed minimal or nearly no changes in void volume. However at 350℃, intra-crystalline voids appeared mainly. From 350℃ to 600℃, the increase in the number and size of voids was more significant in the positive birefringent area. In summary, all these techniques have contributed to the research of the heating effect on enamel and the results are summarized in Table 2-1. 2.2.2 Chemical Reaction The chemical changes of heated enamel have been studied by several researchers (Corcia and Moody, 1974; Sakae, 1988; Hsu et al, 1994; Fowler and Kuroda, 1986). Their findings can be summarized as follows: 1. The water content decreases with increasing temperature and an abrupt decrease happens at 250 ℃ to 300 ℃ losing about one-third of the amount initially incorporated water in enamel. The decrease of water content coincides with a sharp contraction of lattice parameter (a-axis) at the same temperature range. Therefore, Holcomb and Young (1980) stated that the enlargement of the a-axis length is related to the structurally incorporated water. 2. With the increasing temperature, there is a consistent loss and rearrangement of CO32- ions: the substitution of CO32- for PO43- decreases and the substitution of CO32- for OH-increases (Holcomb and young, 1980). 18 3. The content of OH- increases progressively to a maximum in the range 300℃ to 500℃(Fowler and Kuroda, 1986). 4. The endothermic water maximum is in a range from 100℃ to 140℃, up to a peak at 140℃. The pyrolysis and volatilization of organic constituents change from 250℃ to 400℃, up to a peak at 350℃ (protein decompose). The oxidation of carbon content attends the peak at 500℃. A glass phase is formed between 400℃ and 460℃ (Corcia and Moody, 1974). 5. The acid phosphate ions condense to form pyrophosphate (P2O74-) ions and its content progressively increases in the temperature range 200℃ to 400℃. 6. The formation of more soluble Beta-TriCalcium Phosphate (β-TCP) in the enamel after 400℃ (Palamara et al., 1987). 7. The decrease of solubility from 150℃ to 400℃ is the main reason for the increased acid resistance (Hsu et al, 1994). From above chemical changes, enamel gradually lost its translation materials (water and protein) and becomes more stable and more resistant to the acid solution. 2.2.3 Heat-Induced Reduction of Enamel Demineralization After heat treatment, surface enamel showed an increased birefringence with increasing temperatures. Oho and Morioka (1990) attributed it to the loss of the protein and the formation of microspaces in enamel acting as a trap for the deposition of ions released by acid attack. This trap could compromise acid effect and reduce the demineralization rate. Sakae (1988) used X-ray diffraction to study the size change of crystallite after heat 19 treatment of the enamel samples. Contraction of the a-axis length was observed and attributed to the loss of structurally incorporated water from the apatite lattice. Holcomb and Young (1980) found that in the range 300℃ to 500℃ OH- content increases and at 500℃ pyrophosphate (P2O74-) appears in the hydroxyapatite. Sato (1983) found the minimal calcium dissolution at 350℃ and attributed to the products of pyrolysis of organic matrix, which closed up the porosity newly formed in heated enamel. All these changes could affect the ions diffusion and dissolution of enamel during acid attack. The heating effects on the constituents of the tooth enamel, birefringence, and artificial lesion depth are summarized in Table 2-1. The arrows indicate the trend of the changes, whereas the numbers specify the articles in which the specific effect was studied. 20 Table 2-1: Summary of the heating effect Factors R.N. OM Porosity H2O a-axis Soluble Carbonate A B OH- Total A+B Pyrophosphat e Enamel melting / fusing - - X1 X2 X2 X1 X2 X2 +- +- - - + + T(℃) 24-100 100-200 200-300 300-400 - ++ - - - + + -- + + -- - + ++ - + - 11 11 10,11 1,5,10,11 5,10,11 400-500 11 12 11 12 8,12 2,8,12 2,8 12 1,9,11,12 1,5,9, 10,11,12 1,5,9,12 1,9,12 - 1,9,12 7 7 1,7,9,11 7,11 + 7 - 9 9,12 1,9 9,12 1,9 1,9 +1, 9 9,12 1,9,12 9, 12 1,9,12 9, 12 1,9,12 + + 1,9, - 1, 9,12 9,12 1,9, 12 1,9, 12 1,9, 12 - 1, 6, 9,12 + 1 1 1,9 ΒTCP X2 X2 X2 X2 - 5,10,11 + 8 12 - 1,9,12 + 7 - 9 1,9,12 - 1,9,12 - 1, 6, 9, 12 LD * - X2 + + + 1,3,5,9 + 2 2 2 2 + + + 8 8 + 2 * 8 12 500-600 BF 2 + + + 8 8 2 Note 1: “+” indicates increment with temperature Note 2: “-” indicates decrement with temperature Note 3: “x” indicate no change Note 4: “x” indicates no change Note 5: “*” indicates no data Note 6: Number of relevant articles is as follows: 1 ─ Fowler and Kuroda, 1986 2 ─ Sato, 1983 3 ─ Herman H, Dallemagne MJ, 1961 “ 4 ─ Hsu et al., 1994 5 ─ lin et al., 2000 6 ─ Arends j, Davidson CL ,1975 7 ─ Sakae, 1988 8 ─ Palamara et al., 1987 9 ─ Holcomb and Young, 1980 10 ─ Corcia and Moody, 1974 11 ─ Oho and Morioka, 1990 12 ─ Ying et al, 2004 21 2.3 Caries Formation and Progression 2.3.1 Introduction In dental caries, the disease occurs within the tooth hard tissues where the mineral substance of the tooth is dissolved by acid, and subsequently the organic substance is destroyed by proteolysis. The acid is created by oral bacteria that metabolize and convert carbohydrates, especially sugars, into acid. Caries-susceptible individuals have many such kind of acidogenic (acid-producing) bacteria in their saliva and dental plaques. Because the acids in the plaque are in contact with the tooth surface, enamel beneath the plaque is slightly dissolved. This process is the beginning of a caries lesion. The presence of an apparently intact surface layer overlying a subsurface demineralization is the feature of the early lesion. A slight increase in enamel porosity changes the optical properties of the enamel in such a way that light is scattered. Because of this, enamel gradually becomes less and less translucent with increasing tissue porosity in the early demineralization. Clinically, this can be observed as the appearance of whitish (opaque) changes of the enamel, or “white spot”. Because carious dissolution follows the direction of the rods, the lesion appears triangular in sections cut through the central lesion part. As the mineral loss increases, the surface enamel loses the support of the beneath structure and finally collapses. The cavities consequently appear on the enamel surface (Fejerskov O and Kidd EAM, 2003) . 2.3.2 Protective Role of Sound Enamel Surface Chemistry The physico-chemical integrity of dental enamel in the oral environment is heavily hinged upon the composition and chemical behavior of the surrounding fluids. 22 Normally, saliva and plaque fluid are supersaturated with respect to enamel apatite. Under this chemical equilibrium, it not only prevents enamel from dissolving, but also tends to precipitate apatite, partly as calculus in the form of crystal growth on surface enamel. But this equilibrium could be broken if the exterior ions added or the concentration of ions changed (Moreno and Zahradnik, 1974). Factors affected the stability on enamel apatite are the free active concentrations of calcium, phosphate, fluoride, organic matrix and pH in solution. 2.3.2.1 Fluoride, Carbonate and Magnesium The surface enamel contains, for example, high concentrations of fluoride. Fluoride is of particular importance in stabilizing the surface enamel by the reaction of fluoride with the dental apatite. The major product is called calcium fluoride (CaF2) (Moreno and Zahradnik, 1974): Ca10 ( PO4 )6 (OH ) 2 + 20 F − + 11H + → 10CaF2 + 3H 2 PO4 − + 3HPO4 2− + 2 H 2O The surface enamel also contains low concentrations of carbonate and magnesium, which have a destabilizing effect. When moving inward away from the surface, gradients of fluoride decrease, while gradients of both carbonate and magnesium increase together with increasing porosity (Hallsworth et al., 1972). Therefore, as caries process progresses inward toward the deeper enamel, the chemistry of dissolution will change, and the solubility will increase. 2.3.2.2 Organic Matrix The presence of organic matrix on or in the enamel surface is a contributor to surface zone formation by reducing mineral loss or acting as a barrier. When protein material has 23 been removed, the natural lesion is able to take up more calcium from the external environment (Robinson et al., 1990). This supports the view that the protein layer can slow the transmission of mineral ions through the enamel surface. 2.3.2.3 The pH value When pH falls, the solubility of the enamel apatite will increase dramatically. The rationale is that: the hydroxyl concentration is inversely proportional to the hydrogen concentration and the concentration of the phosphate ionic species depends on the pH of the solution. When the pH decreases, more PO43- ions are transformed to HPO42- that in turn reacts to H2PO4-. The conversion of phosphate and the effect on the solubility can be illustrated by the reaction (Fejerskov O and Kidd EAM, 2003): Ca ( PO4 )6 (OH ) 2 = 10Ca 2+ + 6 PO4 3− + 2OH − H + ↑↓ H + ↑↓ HPO4 2− H 2O H + ↑↓ H 2 PO4 − The pH at which the fluids are exactly saturated with respect to enamel apatite is defined as the “critical pH”. It is usually around 5.5. However it is not a fixed number. It is determined by the concentration of calcium and phosphate presented in the oral fluids (Dawes, 2003). When the pH goes below the critical level the aqueous phase is unsaturated with respect to hydroxyapatite due to the decreased activity of PO43- and of hydroxyl. 24 If the pH increases again, the aqueous environment of the enamel surface will gradually return to a state of supersaturation with respect to hydroxyapatite and/or fluorapatite. Caries progression from ultrastructural changes to visible decay should therefore be regarded as the cumulating effect of a long series of alternating dissolution at a low pH and a partial reprecipitation when pH rises. 2.3.3 In vitro Demineralization To study the formation of lesion, many laboratory models (Darling,1956; Gray and Francis, 1963; Holly and Gray, 1968) have been built, trying to understand the mechanism behind this phenomenon. It has been related with the driving forces and the kinetics in the carious process and the pertinent equations based on diffusion theory have been developed. The initial enamel dissolution rate is largely a function of the total buffer concentration, buffer acid strength, and pH (Gray, 1962). Several theories have been proposed to explain the phenomenon of preferential subsurface dissolution occurred when the dental enamel is subjected to acid attack. These theories are usually focused on discussion of a specific factor. However it is increasingly clear that many factors are simultaneously contributing to the process of demineralization. Firstly, the anatomical variations in the structure and composition of enamel are the cause of the subsurface demineralization. The outer enamel surface, which is intact in early caries, is protected by inhibitors, for example, F+, derived from the oral environment and adsorbed on the outer surface. It makes the surface less soluble (Gray, 1965). Conversely fluoride ions are less in subsurface enamel. But the concentration of carbonate and magnesium, which are related to the solubility of enamel, increases from the surface of 25 enamel to DEJ. The gradients favoring dissolution of mineral content stimulated the subsurface demineralization (Theuns et al., 1986a). Secondly, the initial stage of lesion formation is described as the dissolution of mineral with outward diffusion of calcium and phosphate ions. As this diffusion ions accumulated at the surface, the surface enamel was in a dynamic process of demineralization/remineralization. Margolis and Moreno (1985) suggested that the precipitation of CaHPO4·2H2O (DCPD) from dissolved enamel in the subsurface region was the cause of intact surface layer. Moreno and Zahradnik (1974) stated that the quasiequilibrium was maintained by a kinetic balance between the rate of transfer of dissolved ions across the enamel-solution interface and the rate of precipitation of DCPD, HA and FA. Theuns et al. (1986a) attributed the formation of surface layer to the transformation of the apatite to a more stable calcium phosphate in the presence of acid, as more F+ or less carbonate contained in an apatite. Thirdly, following the formation of the incipient caries, a model acting as a pumping mechanism can be formulated as schematically illustrated in Figure 2-2 (Moreno and Zahradnik, 1974). The model works as follows: Firstly, the plaque bacteria, particularly Streptococcus mutans and lactobacilli, produced organic acids (HB). Second, the acids diffuse through the pellicle into the surface enamel. Third, the plaque pH drops to maybe 5.5, the critical pH. Some phase transformation occurs in the surface enamel, including dissolution of the enamel surface, followed by precipitation of the solid phases CaHPO4·2H2O (DCPD) and Ca5F(PO4)3 (FA). Therefore there are three solid phases, DCPD, FA and the bulk of the tooth enamel, in the surface enamel. Last, basic constituents will diffuse from the inner region into the surface zone, and from the surface 26 zone into saliva. This diffusion rate of transportation from surface zone to saliva at least matches the rate of precipitation in the surface zone. Otherwise, an incipient caries would appear (Moreno and Zahradnik, 1974). SALIVA PELLICLE PLAQUE SURFACE ENAMEL INNER ENAMEL BASE B-H+ HA FA DCPA DEMIN ENAMEL Ca ACID Ca,P P Figure 2-2. Schematic model of subsurface demineralization Last but not least, it was related to the differing diffusion coefficient and charges of ions. The permselectivity of the enamel was suggested as an important factor governing the relative rates of ion diffusion, hence attributing to the process of demineralization (Brown and Chow, 1986). 2.3.4 Polarized Light Microscopy (PLM) and structure of carious enamel The PLM is quite similar to a conventional light microscopy with the addition of a polarizer, a rotating stage and an analyzer. It has been used to study the optical properties of tooth enamel for more than a century. Under PLM, an anisotropic, uniaxial crystal is oriented at 45º to the plane of the polarized light. The crystal splits the light into two beams with a refractive index known as n0 for 27 the ordinary ray and ne for the extraordinary ray (Figure 2-3) if ne is greater than n0, the sign is positive, and if ne is less than n0, the sign is negative (www.olympusmicro.com). Retadation no no Crystal ne ne d Figure 2-3. Diagram of the retardation formation The basic unit structure of enamel is the tightly packed hydroxyapatite (HAP) crystal. It is a uniaxial crystallite and has a negative intrinsic birefringence in relation to the prism length. The organic matrix of the enamel has a positive birefringence. But it is too small and can be ignored (Carlstrom and Glas, 1963). The form birefringence is caused by fluids or gases with refractive indices (RI) different from that of enamel (RI=1.62). It is positive in relation to the prism. The amount produced depends on the relative volume of spaces presented and on the difference between the RI of the enamel and that of the medium within the spaces. Hence the form birefringence can be avoided by examining specimens in a medium with the same RI as enamel. The observed birefringence is the sum of these two kinds of birefringence (Carlstrom and Glas, 1963). It is closed to (-) 0.003 (Cape and Kitchin, 1930) and (-) 0.00292 (Keil, 1936) for the fresh human enamel. 28 During the carious dissolution, the increased spaces in the enamel give rise to the form birefringence. Therefore, the greater the spaces in enamel, the more positive form birefringence will increase, which can reduce, compensate or reverse the negative intrinsic birefringence of the normal enamel. In the end, the observed birefringence can change to positive. The observed colors of a thin section of enamel viewed with PLM are produced by inserting a ¼ lambda color tint into the light path. Every time the stage is rotated by 90º, the specimen will change color. The two quadrants in which sound enamel is blue-green are said to be negative while the rest quadrants are positive. Therefore the negative birefringence of sound enamel becomes positive one due to the increased form birefringence after demineralization. Darling (1956) and Silverstone (1968) have characterized four zones in the early lesion of enamel caries without cavitations- the translucent zone, the dark zone, the body of lesion, and the surface zone. 2.3.4.1 The Translucent zone The translucent zone of enamel is not seen in all lesions. It lies beneath the dark zone and between it and the normal enamel at the advancing front of the lesion. Its main feature is the production of a considerable amount of large spaces than the healthy enamel. During the attack of caries, these small spaces may persist or may be lost. Mummery (1926) indicated that it was hypercalcified and may be a protective reaction against caries. 29 2.3.4.2 The Dark zone The dark zone acts as a molecular sieve (Barrer, 1949). It contains a series of spaces of various sizes larger than translucent zone, having a pore volume of 2-4%. It is dominated by an almost equally large amount of minute spaces that are inaccessible to quinoline. The spaces are accessible only to small molecules such as those of air, water and methanol. It has positive birefringence and appears dark in water. Keil (1936) suggested that this may be caused by gases, e.g. CO2 retained in the decalcified enamel. 2.3.4.3 The Body of lesion The body of lesion has only large spaces that are accessible to quinoline. It appears positive birefringence and has a pore volume of 25% (Silverstone, 1968). It appears as small wedge-shaped areas with the base at the enamel surface and occurs in relation to the striae of Retzius. The body of the lesion shows decalcification with cross-striations of the prisms and enhanced striae of Retzius. But it is not known whether there is any change in the organic matrix at this stage. The continuous enlargement of the pores of this zone ultimately leads to cavitation. 2.3.4.4 The Surface zone The surface zone is a well-calcified zone lying over the lesion. It retains negative birefringence even when examined in air, thereby indicating a pore volume of less than 1%. This relatively unaffected surface zone is approximately 20 to 40 microns in depth (Silverstone, 1968). 30 2.4 Diffusion Phenomena of Enamel 2.4.1 Diffusion Phenomena and Diffusion Pathways The physical and chemical compositions of enamel indicate that it is definitely not a solid impermeable mass of hydroxyapatite. Enamel is a complex porous solid, composed of long thin crystals surrounded by a matrix of water and organic materials. Enamel of an erupted tooth has an intimate relationship with fluid around it. The interior of the enamel is in contact with the enamel fluid and the exterior with the saliva or the gingival fluid. Bergman (1963) has found that there was a fluid transportation between enamel and dentine. As the size of pore in enamel is so small compared to that in dentine, some large molecules, which exist in dentinal fluid, could not diffuse into enamel. Fosdick (1962) illustrated a relationship between the relative osmolality of saliva and blood, which indicated a constant flow of water transporting from the tooth surface to the pulp. But in some specific situations, for example, higher hydrostatic pressure of the blood or ingestion of concentrated sugar, the direction of transportation would be reversed. Calcium and sodium would then diffuse from the blood to the saliva, and the phosphate and potassium to the blood. As early as 1929, Bodecker (1929) has assumed that the “enamel fluid” or “dental lymph” served as a transport medium for molecules and ions, both during the developing period of enamel and in the fully mature tooth. Deakins (1942) agreed with this statement. Through this diffusion medium, diffusion of acid into and minerals out of enamel are feasible during the process of enamel caries. Therefore, understanding the diffusion behavior of enamel is a crucial part to fully comprehend the mechanism of dental caries at the molecular level. 31 The pathways for molecules and ions through the enamel are those parts where the crystallites are less densely packed and not so well oriented. These parts are largely represented by enamel sheaths and inter-rod substance. These places that have high concentration of organic matrix (Melfi and Alley, 2000) noted that the loosely bound fraction of water, or free water, is related to the organic matrix (Carlstrom et al., 1963). Therefore it is reasonable to conclude that the ions diffusion in enamel was related to loosely bound water. Brudevold et al. (1960) showed that there are relatively less amount of loosely bound water near surface enamel, and progressively more approaching the DEJ. This leads to the assumption that the rate of ions diffusion in the enamel would be the least in the surface and the greatest near DEJ. In the above statement, the terms, such as “enamel fluid” and “dental lymph” (Bodecker, 1929), are actually referring to the free water phase of the enamel with its dissolved molecules and ions. Zahradnik and Moreno (1975) further pointed out that the high activity energy of the water diffusion through enamel was caused by the particular structure of water in the pores of enamel. 2.4.2 The Published DC of Enamel Moreno and Burke (1974) used the plano-parallel enamel membrances to checked the self diffusion of water in human enamel and obtained the diffusion coefficient (DC) was 1.16 ± 0.08X10-8 cm2/sec. Van Dijk et al (1979) estimated the the DC of acetic acid in artificial caries was near 10-8~10-7 cm2/sec. Featherstone (1984) made a summary of the DC for the diffusion of ions and molecules through enamel. The data of DC compiled in his paper was in a range of 10-7 to 10-12 cm2/sec. 32 2.4.3 Factors Affecting Enamel Diffusion 2.4.3.1 Pore Size Zahradnik and Moreno (1975) showed that dental enamel has a bimodal pore distribution. The size of pores in mature enamel is so small that it can only admit air and water, or molecules or ions of the same order of magnitude (Fosdick, 1963). Burke and Moreno (1975) used tritiated water to show that the small pore size contributed to the pore constriction of the enamel, which worked as energy barriers to diffusion. 2.4.3.2 Charge of Surface Enamel Enamel acts like a charged membrane (Waters, 1975) and has an ion selection behavior where the cations were more mobile than the anions. The molecular sieve effect of enamel has been studied by Borggreven et al., (1977). de Rooij (1980) used radioactive phosphate to measure the diffusion profiles and compared it with calcium and fluoride to show that the phosphate profiles differed considerably from the others. Therefore different ions had different diffusion properties. 2.4.3.3 Organic Matrix in Enamel The diffusion of Na22+ in untreated and boiled enamel tissue showed that boiling reduced diffusibility. It might be caused by the coagulating effect of the organic matrix of the enamel (Arwill et al., 1965). The lipid component of enamel acts as an inhibitor to the ions diffusion during demineralization. As a result, the lesion progress in lipid extracted enamel is more than doubled that in normal enamel (Featherstone and Rosenberg, 1984). 33 2.4.3.4 Chemical Reaction Some molecules and ions can pass freely without changing the composition of the enamel, while others can react with the enamel and eventually alter its physical and chemical properties. Featherstone (1984) summarized the diffusion properties for different ions and molecules in different diffusion matrix. Some of his statements are excerpted here: The net apparent diffusion coefficient for overall caries process was about 10-9 cm2/s or less in enamel, while these values are 100- to 10,000- fold smaller than those for simple diffusion in water. The diffusion process in plaque is many times faster than those in enamel where there is no chemical reaction with enamel mineral. The calcium, phosphate and fluoride, which are part of, and also react with, enamel mineral, have much smaller measured DCs. A DC in the order of 108 cm2/s is appropriate for the diffusion of non reacting species through enamel. It is clearly from the above statements that the chemical reaction between enamel and molecules or ions affects the ions diffusion in enamel. 2.5 Confocal Laser Scanning Microscopy (CLSM) and Fluorescence Application on Dental Enamel 2.5.1 Introduction CLSM is now a well recognized technique used in the fields of biology and medicine. Compared with the conventional epi-illumination light microscopy that simultaneously illuminates the entire view of a sample (full-field illumination), CLSM just illuminates a single point of the specimen (point scanning illumination). This gives CLSM a significant high axial resolution, improves plane resolution, and removes out -of-focus blur. With these advantages, it could be used for a series of non-invasive optical sectioning. In addition, CLSM is compatible with the computer image storage technique, which allows 3-D measurements within semitransparent samples available. Hence CLSM is a convenient technique to acquire 3D image data without damaging specimens to be 34 observed. When CLSM is equipped with the feature to photobleach a user-defined region, it becomes a handy tool to perform fluorescence recovery after photobleaching (FRAP) measurements. 2.5.2 Confocal Laser Scanning Microscopy 2.5.2.1 Principles and Theoretical Background The idea of optical scanning microscope was firstly introduced by Minski as early as in 1957. But it nearly took yet another thirty years to develop CLSM. The confocal principle is illustrated schematically in Figure 2-4. Figure 2-4. Schematic structure of a CLSM (www.staff.kvl.dk/~als/confocal.htm) A laser beam with specific wavelength is chosen to pass through an excitation aperture (excitation pinhole), which adjust laser beam to a thinner one, and then is reflected by a beamsplitter and is adjusted by a scanning unit. The scanning unit inside microscope can deflect the laser source in the x- and y- directions, and facilitate the point-by-point 35 scanning in a specimen. The adjusted thin laser beam is focused on the focused plane within the specimen by an objective. The emitted longer-wavelength fluorescent light is collected by the objective and passes through the dichroic mirror and is focused on an emission pinhole. The emission pinhole is in a confocal (conjugate) position with the excitation pinhole, which is able to eliminate all the out-of-focus light, i.e., all light coming from the region above or below the focus plane is eliminated. Therefore, CLSM provides an excellent resolution both in the plane of section (x- and y-directions), and between section planes (z-direction). All information of specimen points is collected by a photomultiplier (PMT), positioned behind the emission pinhole. All signals are digitized and transmitted to a computer. The three-D image of the specimen can be generated in a computer. Compared with the conventional microscope, the CLSM has two important advantages that endow CLSM special function to calculate the image intensity of a point, which are (a) the point spread function and (b) optical sectioning: (a) The point spread function With an ideally small confocal aperture, the image of a point can be described in quantitative terms by the point spread function (PSF), which represents the intensity distribution in the image space. For the purpose of determining PSF, both illumination and observation (detection) are limited to a point, also called PSF. To investigate the three dimensional imaging properties of a confocal LSM, it is necessary to consider the 3D image or 3D PSF. In the ideal diffraction-limited case (no optical aberrations and homogeneous illumination of the pupil, the 3D-PSF is likely to be of cornet-like, rotationally symmetrical shape. It 36 has a close relationship with the working depth and Numerical Aperture (NA) of objective., as shown in Figure 2-5. Figure 2-5. Diagram of the 3D-PSF Note: High magnification, high N.A. objectives are accompanied by very shallow depth of field and short working distance from the specimen. (www.zeiss.com) The central maximum of the 3D-PSF, in which 86.5% of the total energy available in the pupil is concentrated, can be described as a cylinder of rotation (Figure 2-6). For consideration of optical slice thickness it is useful to define the half-maximum area of the cylinder, i.e. the well-defined area in which the intensity of the 3D point image in axial and lateral direction has dropped to half of the central maximum (i.e. the full width at the half-maximum height, FWHM). So any reference of PSF exclusively refers to the halfmaximum area (www.zeiss.com). 37 Figure 2-6. Section through the 3D-PSF Note: The left image is in Z direction, and the right image is in XY-direction. (www.zeiss.com) The total PSF (PSFtot) behind the emission pinhole is composed of the PSF of the illumination and detection ray path that are called PSFill and PSFdet, respectively. ( PSFtot ( x, y, z ) = PSFill ( x, y, z ) ⊗ PSFdet ( x, y, z ) ). The imaging properties of a CLSM are determined by the interaction between PSFill and PSFdet. This interaction is termed convolution ( ⊗ ). Later in the computer the signal of the image can be deconvolved and the original shape of the object is restored (Figure 2-7). (a) (b) Figure 2-7. The convolution and deconvolution used in the theoretical 3D-PSF Note: (a) An object is convolved by the optical system’s PSF. (b) The object image can be deconvolved and the original shape of the object restored. (www.zeiss.com) (2) Optical Sectioning Instead of using a microtome to slice a thin section out of a thick sample, thin section inside the sample can be imaged. With scanning, a CLSM can image a whole plane inside a thick sample and then focus deeply inside a specimen to image a different layer, and 38 those two images do not interfere with each other. With proper controls, a CLSM can image a whole stack of optical sections, which can later be assembled into a threedimensional display Figure 2-8. Figure 2-8. Diagram of the optical slicing in a specimen (www.zeiss.com) 2.5.2.2 Basic Structure Figure 2-9 shows the basic structure of the CLSM. It is consisted of six main parts, which is stated in the following paragraphs: Figure 2-9. Basic structure of CLSM (www.zeiss.com) (1) Laser: To excite fluorescent dyes and fluorescent proteins, the CLSM is equipped with different lasers emitting a number of wavelengths of light in a range from the UV to visible spectral (Figure 2-10). Currently, the most common laser line used in a CLSM is 39 Argon Ion laser (488nm and 514 nm) and Helium-Neon Laser (543nm, 595nm, and 633nm). Figure 2-10.Wavelength of laser spectrum (2) Scanner: It contains of motor-driven collimators, scanning mirrors, adjustable pinholes, and highly sensitive detectors. All these components are arranged to ensure optimal specimen illumination and efficient collection of reflected or emitted light. (3) Objective lens: Users can select the lens to get the best combination of resolving power, aperture (N.A.), working distance and correction for their specific applications. The higher magnification of objective is, the higher the numerical aperture, and the shorter working distance (Figure 2-5). (4) Beam Splitter: It is a combination of diachronic mirrors and emission filters. Users can choose different filters to reduce the interference of excitation and emission wavelength of the fluorescent dyes. (5) Pinhole: It is used to prevent the out-of-focus light to be detected by microscope; it also determines the thickness of optical slice. Increasing the size of pinhole will increase the signal intensity, it also compromise the effect of confocal. Usually 1 Airy Unit (AU) is the optimal choice and it could be calculated from equation: 1AU = 1.22 × λ NA 40 λ ------- Wavelength of the illuminating laser light NA----- Numerical aperture of the objective (6) Photomultiplier Tube (PMT): It is used to detect the photons emitted/reflected by the specimen. 2.5.2.3 Application of CLSM in Dentistry CLSM is a powerful technology providing new optical information different from the conventional microscope. It can provide high contrast images, individual continuous optical sectioning without slicing specimen, and stereographs that are observed by three dimensional reconstructions with little blur and high spatial contrast. CLSM is frequently employed in the measurement of intracellular ion concentration, such as Ca2+, H+ ions (Lipp and Niggli, 1993). One major advantage of CLSM is that it requires less specimen preparation, which contributes to its wide use in the study of dental hard tissues. As early as 1987, Watson and Boyde (1987) used CLSM to study the behavior of adhesives on dentine. The most common applications of CLSM in dentistry include visualization of bonding structures (Shimada anf Tagami, 2003), caries research (Okuda, et al., 2003), biofilm research (Hope et al., 2002) and 3-D studies of dental hard tissues (Nakamura and Kawata, 1990). Along with the development of new dyes, CLSM may provide a valuable new technique for the visualization in tooth enamel. 41 2.6 Fluorescence Recovery After Photobleaching (FRAP) 2.6.1 Introduction FRAP is a well-known method to measure the mobilities driven by diffusion. Peter et al. (1974) has developed FRAP as a tool to study the molecular mobility in several media, including aqueous solutions, gels, and living cells. Being a relatively old technology, FRAP was originally used to study membrane diffusion of lipids and proteins coupled to fluorophores in the cell (Axelrod et al., 1976; Edidin et al., (1976). But with the development of fluorescent proteins, FRAP makes it possible to understand the physical and physiological properties of the cell and its nucleoplasm. More recently, noninvasive fluorescent tagging with the green fluorescent protein (GFP) or its variants has stimulated the use of FRAP to measure protein dynamics inside the living cell (Reits and Neefjes, 2001) and the surrounding environment (Carrero et al., 2003). Thereby the fluorephores are important part of FRAP. 2.6.2 Basic Principles of Fluorescence Fluorescence is a well-known optical phenomenon that occurs when the molecule absorbs a photon and triggers the emission of another photon with a longer wavelength. The diagram shown in Figure 2-11 illustrates the process. 42 Figure 2-11. Jablonski diagram illustrating fluorescence (www.invitrogen.com) The process is characterized by three stages - excitation, excited-state lifetime and fluorescence emission. In the excitation stage, the fluorophore absorbs a photon energy hνEX (where “h” is the Planck’s constant, “νEX” is the frequency of light) coming from external exciting source and jumps to the excited state (S1'). The lifetime of this state, however, is very short, and then the fluorophore will transit to the less excited state (S1). At the end, the fluorophore experiences the emission stage, which transits from the less excited stated (S1) to the ground state (S0). Meanwhile, a photon of energy hνEM is emitted. Because of the energy dissipation during the excited-state, the energy of photon (hνEM) is lower than the absorbed (hνEX). As a result, the emission photon has a longer wavelength than the excitation photon. The difference between the two wavelengths or frequencies (νEX -νEM) is called Stokes shift (Figure 2-12). Stokes shift is fundamental to the sensitivity of fluorescence technique because it allows emission photons to be detected separately from excitation photons. 43 Relative Fluorescence 1.0 0.8 0.6 Excitation Emission 0.4 0.2 0.0 300 350 400 450 500 550 600 650 700 Wavelength Figure 2-12. Stokes shift (http://www.fluorescence.com) The entire fluorescence process is cyclical and the same fluorophore can be excited and detected repeatedly until the fluorephore is irreversibly destroyed. A single fluorophore can generate many thousands of detectable photons. If a fluorophore is designed to localize within a specific region of a biological specimen or to respond to a specific stimulus, the fluorescent change of the specific fluorescent probes will reflect the psychological changes in the targeted items. In practical experiments with polyatomic molecules in solution, the discrete electronic transitions hνEX and hνEM in Figure 2-11 will replaced by much more broad energy spectra - fluorescence excitation and emission spectra. The bandwidth of these spectra is of particular importance for applications in which two or more different fluorophores are simultaneously detected. 44 2.6.3 Autofluorescence and Photobleaching in Dental Enamel Fluorescence detection sensitivity is severely degraded by background signals, which may originate from endogenous sample constituents (referred to as autofluorescence) or from unbound or nonspecifically bound probes (referred to as reagent background). When the tooth was excited by argon laser light of wavelength 488nm, intact enamel was found to be fluorescent with a yellowish light around wavelength 550nm (AngmarMansson and ten Bosch, 1987; Alfano and Yao, 1981). Gonzalez-Cabezas et al. (1998) stated that early enamel mineral changes can be monitored by measuring the changes in autofluorescence of the enamel with CLSM. Noisy autofluorescence can be minimized either by selecting filters that reduce the transmission of emission energy, relative to excitation energy or by selecting probes that emit longer wavelength laser beams. Under high-intensity illumination conditions, the irreversible destruction or photobleaching of the excited fluorophore becomes the factor limiting fluorescence detectability. The most effective remedy for photobleaching is to maximize detection sensitivity by using lower laser power and the widest emission bandpass filters to reduce the photobleaching effect. However everything has two edges. The disadvantage of the photobleaching has made it specific function to measure the movement of the molecules over time used in FRAP. 2.6.4 Fluorescent Dyes Commonly Used for Dental Enamel Several dyes have been widely used in the dental researches including Rhodamine B and fluorescein. Rhodamine B 45 Rhodamine B isothiocyanate is comprised of electro-neutral, undissociated molecules, which do not dissolve in water and, therefore, lead to a contrast under CLSM. This particular feature has made it the most commonly used dyes for dental hard tissue with CLSM (Pioch et al., 1997). Watson and Boyde (1987) were the first to study the behavior of adhesive resins on the dentin surfacing using the CLSM, with the help of rhodamine B. Watson (1989, 1991) developed a new technique with rhodamine B to visualize the bonding structure such as the hybrid layer in dentin with labeled primer component. It had no destruction to the specimen and could clearly distinguish components of bonding agents. Benn and Watson (1989) also stated that the use of this fluorescent marker (rhodamine B) can greatly increases the sensitivity of this method. That imbibition of a fluorescent dye into porosities of demineralized or remineralized enamel could be used to measure the thickness of samples quantitatively under CLSM (Gonzalez-Cabezas, et al., 1998). Fluorescein Fluorescein isothiocyanate (FITC) is a well-established fluorescence probe in biological research. The dye-enhanced laser fluorescence (DELF) with the use of Sodium fluorescein dye has become a more sensitive tool to detect early demineralization (Eggertsson et al., 1999). Some features of fluorescein make it possible to be used in dental tissue under CLSM 1. Fluorescein is an amine reactive dye. It has a molecular formula C20H12O5, and molecular weight 332.31. Its diameter is around 0.69±0.02nm (Cvetkovic et al., 2005). The small diameter makes its possible to diffuse inside the porous enamel. 46 2. Fluorescein has a maximum excitation of 496nm and a maximum emission of 512nm (Figure 2-13). The excitation maximum (494 nm) that closely matches the 488 nm spectral line of the argon-ion laser. This makes it an important fluorophore for confocal laser-scanning microscopy (CLSM) applications. Figure 2-13. Optical spectra of fluorescein Note: (a) absorption and (b) emission wavelength (www.invitrogen.com) 3. Fluorescein exhibits multiple, pH-dependent ionic equilibria (Figure 2-14). Because only the monoanion and dianion of fluorescein are fluorescent, with quantum yields of 0.37 and 0.93, respectively. The wavelength and shape of the emission spectra resulting from excitation are close to the dianion absorption peak of 490 nm that is relatively independent of pH. But the fluorescence intensity is dramatically reduced at acidic pH. Therefore, the buffer solution – 1% PBS water, prepared to dissolve fluorescein, has a pH 7.2. 47 Figure 2-14. Ionization equilibria of fluorescein (www.invitrogen.com) 4. The main disadvantage of using fluorescein in dental hard tissue is that its excitation and emission spectra may overlap with those of dental tissue. It has a relatively high rate of photobleaching. However, the lower laser power (2% of 25mW), high-numerical aperture objective (“Plan-Neofluar” 10X/0.3), and sensitive filters (BP 500-550) can be chosen to maximize the emission signal. 2.6.5 FRAP Principles and Applications The underlying principle in a FRAP experiment is that extrinsic fluorophores tagged biomolecules or the natural fluorescent molecules can be checked kinetically by examining the movement after photobleaching of the fluorescent molecule in a defined area during the redistributing time. In a typical FRAP experiment, measuring the movement consists of two phases: Firstly, a high-intensity laser beam with a wavelength near the excitation peak of fluorophores is transmitted into a region of interest (ROI) in specimens within a short period. As a 48 consequence, most of the fluorophores inside ROI are irreversibly destroyed and their fluorescence are lost permanently. This phenomenon is known as photobleaching. Secondly, as molecules attached by fluorophores move in and out of the ROI, fluorescence inside the ROI increases, and eventually an equilibrium is reached. The subsequent recovery of the fluorescence intensity in ROI is monitored by a low-intensity laser beam at time intervals after photobleaching. Analysis of the rate of recovery of the fluorescent signal after photobleaching is a method to determine the diffusion rate of fluorophores or their tagged molecules. Both two-dimensional and three-dimensional models of FRAP have been developed (Wedekind et al., 1994 & 1996; Peter and Kubitscheck, 1999). The FRAP models describing the two-dimensional (2-D) lateral mobility for fluorophore and observation of fluorescence recovery by the fluorescence microscopy were studied by several researchers (Axelrod et al., 1976; Gordon et al., 1995). Blonk et al., (1993) has successfully developed in new FRAP equations to measure three-dimensional (3-D) diffusion coefficient of fluorescent species with objective lens of low numerical aperture (NA) under CLSM. More features have been added into CLSM, including the capability to bleach an arbitrary region in the specimens selected through the computer and the ability to modulate the intensity according to the designed pattern. CLSM now becomes an excellent standard tool to perform FRAP experiment. But the 3-d situation is much more complicated, Braeckmans et al., (2003) designed a three-dimensional model to quantify FRAP experiments in a disk-shaped geometry by a CLSM. 49 2.6.5.1 Two-Dimensional Model To examine the theoretical two-dimensional lateral mobility model of fluorescence recovery curves, several criteria should be fulfilled: 1. The fluorescent molecules in specimens must be initially uniformly distributed. 2. Pure two-dimensional diffusion is monitored by a laser beam of Gaussian intensity profile that has a uniform circular disc profile 3. The fluorescence recovery should be the result of pure diffusion without flow. 4. The diffusion is assumed to occur in an infinite medium to avoid boundary effect 5. Ideally, the fluorophore should not be bleached during the monitoring period 6. The high illumination has no phototoxicity to specimens during photobleaching for FRAP experiment The purpose of early methods applying FRAP was to measure the rates of lateral transport of protein and lipids in cell membrane. Axelrod et al. (1976) developed a standard model to estimate the effective diffusion coefficient. In his model, Axelrod assumes that the concentration of fluorescent species u (x,t) after photobleaching at position x = (x, y) and time t can be represented by the simple diffusion equation: ∂ u( x, t ) = Deff Vu( x, t ), t > 0 (1) ∂t u ( x, 0) = f ( x) Vu = ∂ 2u ∂ 2u + ∂x 2 ∂y 2 (2) (3) (Axelrod et al., 1976) Where Deff is the effective diffusion coefficient, and f(x) is the initial condition of fluorescent species right after photobleaching. This initial condition depends on the intensity profile I(x) of the laser beam that is used to photobleach a ROI. Axelrod et 50 al.(1976) solved Eq. (1) using a Gaussian intensity profile to obtain following explicit theoretical fluorescence recovery curve, which can be used to fit the normalized data: (− K ) N 2t −1 F (t ) = ∑ [1 + N (1 + )] τ Deff N! N =0 ∞ (4) Where τ Deff =W2 / (4Deff) is the characteristic diffusion time, W is the half-width of the intensity at e-2 height, and K is a parameter describing the amount of bleaching. Note that W and K are parameters characterizing the bleaching profile of the laser beam whose values can be obtained directly from the experimental data. 2.6.5.2 Three-Dimensional Model Most of the two-dimensional and three-dimensional models developed are based on accurate numerical approach (Wedekind et al., 1994 & 1996; Peter and Kubitscheck, 1999). An approximate three-dimensional (3-D) model has been developed for use with objective lenses of low numerical aperture (NA) on a CLSM (Blonk et al., 1993). Although the model is only an approximation, it is practically very useful. Derivation of the FRAP approximation model was based on two assumptions: (1) The bleaching phase was very short so that the amount of fluorescence recovery that would take place during bleaching was negligible. (2) The bleaching reaction can be described by an irreversible first-order equation. ∂C ( x, y, z , t ) = −α I b ( x, y, z )C ( x, y, z, t ) (5) ∂t where C(x,y,z,t) is the spatial concentration of fluorophores at time t, α is the bleach rate specific for a certain type of fluorophore in a particular medium, and I(x,y,z) is the 3-D intensity distribution of the bleaching beam. 51 In CLSM, 3-D intensity distribution is designated as point-spread function (PSF). Therefore I(x,y,z) is referred to as the bleaching PSF. Equation 5 is an application of the bleaching for an arbitrary geometry on a disk with radius W and a constant bleaching intensity. For objective with low numerical aperture (NA) and if the resolution of the detecting PSF is much smaller than the radius of the bleaching disk, the following equation can be used: ⎧⎪ +∞ ⎡ ( − K 0 ) n 1 ⎤ Ftot (t ) τt τ t ⎫⎪ −2(τ t / t ) = 1 + ⎨∑ ⎢ × − + (1 e ( I (2 ) I (2 ))) ⎬ (6) 0 1 ⎥ F0 n ! t t + 1 n ⎦ ⎩⎪ n =1 ⎣ ⎭⎪ F0 is the total fluorescence inside the disk before bleaching, τ t = (W2 / (4D) is the radial characteristic diffusion time, K0 is bleaching parameter: Finally DC can be calculated from W2/ τ , where w is radius of bleaching disk and τ is the characteristic diffusion time (Braeckmans et al., 2003). 52 CHAPTER III MATERIALS & METHODS 53 3.1 Overview All steps performed in this experiment are shown in Figure 3-1. It includes: teeth selection, tooth cutting and sectioning, OM extraction, heat treatment, quantification of birefringence and LD using PLM, DC characterization using FRAP/CLSM, and statistical analysis. 3.2 Sample Preparation 3.2.1 Tooth Selection and Preparation Seven non-carious human upper and lower molars (stored in 0.1% thymol solution), brushed with detergent solution, and rinsed in de-ionized water, were selected, using a stereomicroscope (Zoom Stereo Microscope, SZ4045TR, Olympus, Tokyo, Japan). Each tooth was cut into two halves longitudinally (disto-mesia) by using micromotor (Vmax, Nakanishi Inc., Japan). 3.2.2 OM Extraction One of the two halves of each tooth was randomly chosen for OM extraction. To extract the organic matrix, each tooth half was first treated with lipid extraction, followed by protein extraction, and then another lipid extraction. The extraction procedures were illustrated in Figure 3-2. 54 7 teeth selected Each tooth cut longitudinally to get buccal and lingual tooth halves OM extraction (7 tooth halves) Control group (7 tooth halves) 6 sections per tooth-half randomly selected Quantification of birefringence (PLM-I) Heat treatment for 5 groups at different temperatures plus one control group 24 ℃ 100 ℃ 200 ℃ 300 ℃ 400 ℃ 500 ℃ Quantification of birefringence (PLM-II) Each tooth section cut into coronal 1/3 and cervical 2/3 segments Demineralization (cervical segments) FRAP (coronal segments) Lesion Depth (LD) Measurement DC Measurement Statistical analysis for LD, DC, and BF with OM & heating as independent variables Figure 3-1. Flowchart of experimental procedures 55 In the first lipid extraction, the tooth halves were first washed with methanol for 10 seconds, put into a 20-ml glass vial, which was already filled with twelve milliliters of chloroform/methanol (2:1) solvent, and then incubated for 48 hours. During this period, the vials were stirred at a speed of 120rpm on a shaker at temperature 43±1℃. A 5minute sonication was subsequently carried out when this processing is completed, and it was repeated for one more time. The samples were then washed with 100% methanol for 2 minutes, and this completed the first lipid extraction. At the beginning of protein extraction, the samples were transferred into a 10ml sodium hypochlorite solution with available chlorine from 10 to 13%. Then they were stirred at a speed of 120rpm on a shaker at temperature 24±1 ℃ for 2 hours. This process was followed by a 5-minute sonication, and was repeated once again with a fresh solvent. Another 5-minute sonication was subsequently carried out. Finally the samples were washed with distilled and de-ionized water for 10 minutes, and this completed protein extraction. Following protein extraction was the second lipid extraction. It consisted of 5 steps: (1) Incubating the samples in 12-ml chloroform/methanol (2:1) solvent, stirred at a speed of 120 rpm overnight at 24±1℃; (2) Repeating step (1), but with a fresh solvent for 2 hours only; (3) Repeating step (1), but with 12-ml chloroform/methanol (1:1) solvent for 2 hours; (4) Repeating step (2); (5) Washing the samples with methanol and then storing in distilled water. When OM extraction was completed, all tooth halves were kept in de-ionized and distilled water till the next procedure. 56 W ash the samples with M ethanol 10 seconds Chloroform :m ethanol (2:1 v/v) 43 ±1℃,120 rpm stirring and shaking for 48 hours Sonica tion for 5 m inute s Lipid extraction I Chloroform : M ethanol (2:1 v/v) 43 ±1℃, , 120 rpm stirring and shaking for 48 hours W ashed the samples w ith 100% methano l 2 minutes 10 ml N aO C l ( 10-13% C l2) 24 ±1℃,120 rpm stirring and shaking for 2hrs Sonication for 5 minutes 10 ml N aO C l ( 10-13% C l2) 2 4 ±1℃,120 rpm stirring and shaking for 2hours Protein e xtraction Sonication for 5 minutes W ashed the samp les w ith distilled and d e-ionized w ater (10 minutes) Chloroform : M ethanol (2:1 v/v) 24 ±1℃, 120 rpm stirring and shaking , overnight C hloroform : M etha nol (2:1 v/v) at room T , 12 0 rpm stirring and shaking for 2 hours C hloroform : M ethanol (1 :1 v/v) at room T , 12 0 rpm stirring and shak ing for 2 hours Lipid extraction II C hloroform : M ethanol (2:1 v/v) at room T , 120 rpm stirring and shaking for 2 hours W ashed the samples with methanol and stored in d istilled water Figure 3- 2. OM extraction procedures 57 3.2.3 Enamel Sections Preparation Tooth halves were cut perpendicularly to the occlusal surface into enamel sections of thickness140±10µm using microtome (Series 1000 Deluxe, Sci, Fab, Littleton, CO, USA) equipped with a Buehler® Diamond Wafering Blade (Series 15LC, Buehler Ltd. Lake Bluff, IL, USA) as illustrated in Figure 3-3. The morphology and integrity of the tooth sections were then characterized with Polarized Light Microscopy (PLM) (Olympus B.X51, Tokyo, Japan). From each tooth half, six enamel sections without any defect were selected to undergo heat treatment at different temperatures. 3.2.4 Heat Treatment The heat treatment was carried out in air for five of the six sections, using a hightemperature furnace (HT 08/17, Nabertherm, German) at temperatures 100℃, 200℃, 300 ℃, 400℃, and 500℃, respectively. At each temperature the treatment lasted for 2 hours. The remaining one section was left at room temperature (24℃) as a control. The heat was increased at a rate of 2℃/ minute. After heating, the samples were allowed to cool down to room temperature in the furnace. 3.2.5 Demineralization Process After heating, the sections were cut with a disposable micrtome blade (LEICA® model 818, LEICA Instruments GmbH, Nussloch, Germany) into two segments -- the coronal and cervical segments, in the direction perpendicular to the cut tooth surface. Since there were 14 tooth halves and each contributed 6 sections, there would be 84 cervical segments in total. The cut surfaces of cervical segments were covered with the transparent varnish to facilitate lesion measurement under PLM after pH cycling. The 58 natural surfaces were covered with transparent varnish, leaving a window of 0.14x0.2mm2 near the incision line (Figure 3-3). These segments were immersed in demineralization solution (pH 4.5) containing 0.05M acetic acid, 2.2 mM calcium, and 2.2 mM phosphate ions. A two-day (48 hours) demineralization was performed with a stirring speed of 130 rpm at 37℃. The demineralization solution was refreshed daily. In the end, tooth segments were washed in the de-ionized and distilled water for 5 min and the caries-like lesions were measured using PLM. Whole Tooth View Half-Tooth View One-Third Coronal Segment Tooth Section View Spot for FRAP measurement on cut surface Incision Line Two-Third Cervical Segment Window for pH-cycling on natural surface Figure 3-3. Coronal and cervical segments of one tooth section 3.3 Measurement of Lesion Depth (LD) Using the image analysis software (Micro ImageTM, Olympus, Japan) with PLM, one box (100µm in width) was first drawn covering the lesion area. Then the contour of the lesion in the box was traced and the area was measured. This is illustrated in Figure 3-4. The 59 lesion depth was then calculated by dividing the lesion area with the width, which was 100µm. Figure 3-4. Lesion Measurement 3.4 Measurement of DC 3.4.1 Labeling Samples The coronal segments were used to check the diffusion coefficient using FRAP, while the cervical segments were subjected to demineralization treatment. A confocal laser scanning microscope (LSM 510 Meta, Carl Zeiss, Germany) coupled with FRAP was used to check the DC in coronal segments near incision line (Figure 3-3). The 84 coronal segments, which would be subjected to FRAP, were incubated in 0.8mL Fluorescein 20µM (Aldrich, Milwaukee, Wisc., USA) overnight in the chamber on a #1.0 German Borosilicate Coverglass (Lab-Tek® Chambered Coverglass System, Nalge Nunc International Corp., IL) for 12 hours at the room temperature. The DC were quantified at 10µm beneath the natural surface with FRAP measurement performed near the incision line three times for each spot. 60 3.4.2 Theoretical Model For FRAP Assuming that the resolution of the detecting Point Spread Function is much smaller than the radius of the bleaching disk, the following model can be obtained, ⎧⎪ +∞ ⎡ ( − K 0 ) n 1 ⎤ Ftot (t ) τt τ t ⎫⎪ −2(τ t / t ) = 1 + ⎨∑ ⎢ × − + (1 e ( I (2 ) I (2 ))) ⎬ 0 1 ⎥ F0 t t ⎪⎭ (7) 1+ n ⎦ ⎪⎩ n =1 ⎣ n ! where F0 is the total fluorescence inside the disk before bleaching, τ t = W2 / (4D) is the radial characteristic diffusion time, K0 is bleaching parameter and: K0 = α I ob Z 2 v0Vy0 (8) where α is the bleach rate, I ob is the intensity of bleaching beam, Z stands for zoom factor, v is the scanning speed, and Vy is the interline distance. Finally DC can be calculated from W2/ τ , where w is the radius of bleaching disk and τ is the characteristic diffusion time (Braeckmans et al., 2003). 3.4.3 FRAP Procedures A series of pilot studies were done to set the appropriate parameters for the FRAP procedure under LSM 510 Meta (Carl Zeiss, Germany). The general procedure can be summarized as follows: (1) The control samples were mounted in the parafilm-sealed chamber coverglass, imbibed in 1X Phosphete Buffered Saline (PBS), which contains 137mM NaCL, 2.7mM KCL, 8mM Na2HPO3, and 2mM KH2PO4, and checked under CLSM. The parameters were adjusted to render the level of autofluorescence negligibly low. 61 The intensity level was recorded as background. After that, dye-labeled sections were mounted, and the same procedures as that of the controlled samples were followed. The pilot study showed that there was little photobleaching in the samples. (2) The area selected for FRAP should be clear of any obvious defect especially cracks. The prerequisite of the FRAP is that the fluorescein intensity should be distributed homogeneously in the enamel. All optical settings for FRAP were optimized using a low-magnification/low-NA objective with a cylindrical illumination profile (Fig 3-6). (3) The enamel sections were characterized under CLSM, using the 488-nm line of a 25mW argon laser as the excitation laser beam with an objective of PlanNeoFluar (Carl Zeiss, Jena, Germany) 10X/0.3 N.A, an output laser power of 100%, and BP 500-550 IR filter. (4) A predefined region of interest (ROI) (a circle with diameter of 25-µm) was photobleached at the full laser power (100%) as illustrated in Figure 3-5. Recovery of fluorescence was monitored by scanning the ROI at the low laser power (0.2%). A photomultiplier tube is used to measure the intensity of the fluorescent signal. Figure 3-5. Picture of ROI under CLSM 62 (5) The theoretical model has assumed that the radius w of the bleached disk should be sufficiently large compared to the radial resolution r0 of the bleached PSF. Therefore the radial resolution r0 is considered to be independent of the axial coordinate z (Kubitscheck et al., 1998). The measured DC values are independent of the radius of ROI if larger values of radius w are used. Deviations from the 2-D approximation are likely to occur because of an increasing contribution to the fluorescence recovery by diffusion in the z-direction. In this experiment, the lateral resolution r0 of the bleached PSF is 0.497µm. Hence the area to perform FRAP was selected as a cylindrical area (25 µm in diameter and ~50 µm in height) inside the middle of the sample, as shown in Figure 3-6. Enamel Cut Surface Enamel Natural Surface 45μm 140±5μm 50μm 45μm DEJ 25μm 10μm Cylindrical shaped FRAP volume (25μmX50μm) Figure 3-6. Diagram of FRAP setup in tooth section (6) During FRAP, images were acquired as 12-bit TIFF files (256X256 pixel frame). The photobleachig time for ROI was 133 milliseconds by using a 63 bidirectional scan model and bleaching iteration was set to 2. The scan zoom of 20 was selected. During each measurement, the ROI was scanned 5 times before bleaching, after bleaching (~0.26seconds) and monitoring for 40seconds to record the recovery of fluorescence intensity in the bleached area as illustrated in Figure 3-7. 0.9 0.8 0.7 0.6 0 10 20 30 40 50 Figure 3-7. The recovery curve from the image sequence (7) For each selected spot, FRAP was performed 3 times and the mean values were calculated for DC of each section. 3.4.4 Data Collection and Analysis The fluorescence intensity flux in the ROI over time during the photobleaching process was first retrieved from the intensity-frequency table and profile of LSM 510 Meta. The mean fluorescence intensity is then normalized to the mean fluorescence intensity using the background reading. A nonlinear least-squares fit of theoretical 3-D model (Braeckmans et al., 2003) was subsequently employed to determine the experimental parameters, such as characteristic diffusion time ( τ ), leaching parameter (K) and fraction of recovery (f). The curve fitting was performed using self-written software in Mathematica 5.1 © (1988-2003, Wolfram research Inc). 64 3.5 Measurement of Birefringence A spot is marked on the cut surface by pencil. It is 50µm away from the natural surface toward the inner enamel. The light was monochromatized by a tint plate (U-TP530, Olympus, Japan). The objective 10X/0.3 (UplanFI, Olympus, Tokyo, Japan) was used to get a clear picture for the points within the coronal segments. The retardation (R) was measured with a Thick Berek Compensator (U-CTB, Olympus, Japan). Three measurements were performed on the same spot both before and after heat treatment. The birefringence was subsequently calculated according to the following equation: Birefringence (B) = Retardation (R) / Thickness of Section (t) (http://micro.magnet.fsu.edu) 3.6 Statistical Analysis 3.6.1 Evaluation of OM Extraction and Heat Treatment Effects on LD After demineralization processing, LD of the normal group and the OM-extracted group that was with and without heat treatment were evaluated. LD was chosen as the dependent variable. The “OM treatment” and the “heat treatment” were chosen as the independent variables. The “tooth” was the random factor. A two-way ANOVA with Bonferroni correction were performed with SPSS 13 software (SPSS Inc., Chicago, IL) to evaluate the significance level of factors and the potential interactions between factors. 3.6.2 Evaluation of OM Extraction and Heat Treatment Effect on DC The effects of OM and heat on DC was evaluated. DC was chosen as the dependent variable. Both “ OM extraction ” and the “ heat treatment ” were chosen as the 65 independent variables. The “ tooth” was a random factor. A two-way ANOVA with Bonferroni correction were performed with SPSS 13 software to evaluate the significance level of factors and the potential interactions between factors. 3.6.3 Evaluation of OM Extraction and Heat Treatment Effect on Birefringence The mean birefringence of the normal group with and without heat treatment and that of the OM-extracted group with and without heat treatment were evaluated. The mean birefringence was chosen as the dependent variable. The “OM extraction”, the“heat treatment ” and the “ temperature ” were chosen as the independent variables. The “tooth” was a random factor. The three-way ANOVA were constructed to evaluate the statistical significance of factors and their interaction. For the purpose to evaluate the heat effect among the different temperatures, the birefringence changes after heating was calculated. The “birefringence difference” was dependant variable. The “OM” and the “temperature” were independent variables. The “tooth” was a random factor. A two-way ANOVA with Bonferroni correction was adopted with SPSS 13 software (SPSS Inc., Chicago, IL) to evaluate the significance level of factors and their interactions. 66 CHAPTER IV RESULTS 67 4.1 Characterization of Lesion Depth (LD) using PLM 4.1.1 Qualitative Results The lesion areas in the sub-surface enamel show as a region of positive birefringence (Figure 4-1). Usually it is brown. The inner enamel heated with increasing temperatures changes from negative birefringence to positive birefringence. It appears yellowish at 100 ℃ , and gradually changes to a bluish color. At 400 ℃ it becomes opaque. The representative images showing LD at normal and OM-extracted groups in the temperature range 24℃ to 400℃ are shown in the Appendix-A. (a) (b) Figure 4-1. Representative PLM photograph of lesions after demineralization Note: (a) – Normal tooth enamels; (b) – OM-extracted tooth enamels (24℃) 4.1.2 Quantitative Results Initially there are 84 enamel sections. After heat treatment, however, 14 sections, which are heated to 500℃, are too dark to measure their lesion depths. The raw LD data for the rest 70 sections are shown in Appendix B. The means and standard deviations of the two groups (normal and OM-extracted) at different temperatures are listed in Table 4-1. The corresponding LD trends are shown in Figure 4-2. 68 Table 4-1. Comparison between normal and OM-extracted groups Lesion Depth (µm) Temperature( ℃) 24 100 200 300 400 Normal (n = 7) Mean 137.75 110.10 101.41 79.25 86.17 SD 15.12 10.90 10.70 10.13 11.04 OM (n = 7) Mean SD 141.15 11.19 126.84 12.70 108.14 10.09 99.00 16.34 102.39 12.96 160 140 120 100 80 60 0 100 200 Temperature (℃) 300 400 Normal 500 OM Figure 4-2. LD of two groups with heat treatment at different temperature Note: the vertical bars are standard deviation (SD) Descriptive Statistical Analysis: The curves in Figure 4-2 show that LDs decrease gradually with increasing temperatures from 100 ℃ to 300℃, both in the normal and OM-extracted groups. However, when temperature increases to 400℃, LD increases slightly with temperature increase in both groups. At 500℃, the tooth sections are too dark to measure LDs. Generally, LDs in normal groups are smaller than those in OM-extracted groups. 69 Table 4-2 shows LD reduction percentages of normal and OM-extracted groups at different temperatures. The percentage is computed based on the control group at temperature 24℃. Table 4-2. Percentages of LD reduction Percentage Reduction Temperature ( ℃ ) *24 VS 100 *24 VS 200 *24 VS 300 *24 VS 400 *100 VS 200 *200 VS 300 *300 VS 400 Normal -20.07% -26.38% -42.47% -37.44% -6.31% -16.09% 5.03% OM -10.14% -23.38% -29.86% -27.46% -13.24% -6.48% 2.40% Note: “-“sign means percentage decrease, no sign means percentage increase Reference group is indicated with * Table 4-2 shows percentage of LD reduction in heated groups in the temperature range 100℃ to 400℃. The percentage is computed relative to the control group at temperature 24℃. The percentage reductions at temperatures 100℃, 200℃, 300℃, and 400℃ are 20.07%, -26.38%, -42.47% and -37.44% for the normal groups, and -10.14%, -23.38%, 29.86% and -27.46% for the OM-extracted groups. Differences in percentages from 100 ℃ to 200℃, 200℃ to 300℃, and 300℃ to 400℃ are -6.31%, -16.09% and 5.03% for the normal groups, and -13.24%, -6.48% and 2.4% for the OM-extracted groups. Apparently the normal groups show the largest LD drop from temperature 200℃ to 300℃ (-16.09%) and a LD increase from temperature 300℃ to 400℃ (+5.03%). As a comparison, the largest LD drop for OM-extracted groups occurs when temperature increases from 100℃ to 200℃ (-13.24%), and LD also slightly increases when temperature increases from 300 ℃ to 400℃ (+2.4%). 70 Inferential Statistical Analysis: The results of statistical analysis using a two-way ANOVA to evaluate LD are shown in Tables 4-3, 4-4 and 4-5. The Levene’s test shows that variances are homogenous in these groups (p=0.997). The factorial analysis indicates that both “OM” and “temperature” have significant impact on the LD (both p[...]... intact enamel samples only in the integrity of organic matter The ashed samples did 12 not exhibit the phenomenon of activated diffusion So it can be concluded that the pore constriction has a close relationship to organic matter of the enamel 2.1.2 Chemical Properties 2.1.2.1 Inorganic components The main mineral component of enamel is hydroxapatite, nearly 80-90% by weight Since the surface of tooth... process of mineralization of enamel involves the displacement of water by minerals Therefore, the water concentration and the degree of mineralization have an inverse relationship 2.2 Heat- Induced Effect on Human Enamel Previous studies have shown that enamel after heat treatment and enamel with laser irradiation had reduced subsurface demineralization when enamel was exposed to acid solution (Yamamoto... attributed to the products of pyrolysis of organic matrix, which closed up the porosity newly formed in heated enamel All these changes could affect the ions diffusion and dissolution of enamel during acid attack The heating effects on the constituents of the tooth enamel, birefringence, and artificial lesion depth are summarized in Table 2-1 The arrows indicate the trend of the changes, whereas the... the Organic Matrix (OM) 3 and diffusion changes after heating and their roles in reducing the subsurface demineralization would be explored The main objective of this study is to quantitatively evaluate the effect of temperature and /or organic matrix on the carious-like lesion formation, diffusion coefficient (DC) and birefringence of tooth enamel 4 CHAPTER II LITERATURE REVIEW 5 2.1 Human Dental Enamel. .. environment of saliva with supersaturated calcium and phosphate relative to hydroxapatite, analysis of successive layers of enamel has demonstrated that the chemistry of the surface enamel differs from that of the interior enamel in several respects: Fluorine, zinc, lead and iron accumulate in the surface enamel and the concentrations of these elements virtually depend on those of the external environment... the changes of prism orientation When groups of prism are cut transversely, they are known as diazones; those, cut more longitudinally, are known as parazones They can also be reversed by an alteration of the direction of the incident illumination 2.1.1.6 Dentino -Enamel Junction (DEJ) The DEJ is a junction area between enamel and dentine It can be easily seen as a series of depressions or concaves towards... 2.1.1.3 Enamel Tufts, Lamellae, Spindles and Cracks In ground sections, both enamel tufts and lamellae are best demonstrated Enamel tufts extend from the DEJ to the enamel in a short distance These lower-mineralized tufts appear to be branched and contain greater concentrations of enamel protein than the rest of enamel Lamellae project from the surface of enamel to the deeper enamel, consisting of linear,... solubility of apatite mineral (LeGeros, 1984) In addition, Bachra et al (1963) has stated that the dental enamel is a quite stable crystal as it contains less amount of impurities, e.g Mg, Na, CO3, HPO4, citrate, etc 2.1.2.2 Organic components The organic contents of mature dental enamel are at very low concentrations, approximately 1.2% by weight or 2% by volume, consisting of 58% of protein, 40% of lipids,... Another reason for the high concentration of organic contents in tufts and lamella is likely due to faulting of blocking of the exit for enamel protein after maturation Enamel spindles are formed by some newly formed odontoblast processes that are trapped between adjoining ameloblast when enamel formation begins (ten Cate, 1985) 9 2.1.1.4 Striae of Retzius As the human enamel precipitates at a rate of approximately... linear, longitudinally oriented defects, which are filled with enamel protein or organic debris from the oral cavity Origination of enamel tufts is the result of abrupt changes in the direction of packed rods during development But the reason for the development of lamellae could be related to blocking of the relief of enamel internal strains produced by dimensional changes during enamel maturation Another ... to the mixed effect of both permeability reduction and compositional changes in enamel The organic matrix has the retardation effect on the diffusion of ions in enamel The heating in the temperature... effect of the organic matrix of the enamel (Arwill et al., 1965) The lipid component of enamel acts as an inhibitor to the ions diffusion during demineralization As a result, the lesion progress... stage of lesion formation is described as the dissolution of mineral with outward diffusion of calcium and phosphate ions As this diffusion ions accumulated at the surface, the surface enamel