Journal of Dental Research http://jdr.sagepub.com/
Demineralization and Remineralization of Dental Enamel _____ E.C. Moreno and R.T. Zahradnik J DENT RES 1979 58: 896 DOI: 10.1177/00220345790580024301 The online version of this article can be found at: http://jdr.sagepub.com/content/58/2_suppl/896
Published by: http://www.sagepublications.com
On behalf of: International and American Associations for Dental Research
Additional services and information for Journal of Dental Research can be found at: Email Alerts: http://jdr.sagepub.com/cgi/alerts Subscriptions: http://jdr.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jdr.sagepub.com/content/58/2_suppl/896.refs.html
>> Version of Record - Feb 1, 1979 What is This?
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
Demineralization and Remineralization of Dental Enamel E. C. MORENO and R. T. ZAHRADNIK Forsyth Dental Center, 140 The Fenway, Boston, Massachusetts 02115, U.S.A. J Dent Res 58(B):896-902, March 1979 Experimentation in vitro using organic acid buffers calcium phosphates. This inhibition has been demonstrated for some macromoleas demineralizing media shows that caries-like lesions can be obtained which are very similar in cules at concentrations much lower than morphology and developmental stages to early those present in saliva. Consideration of oral lesions formed naturally under oral conditions. conditions in experimental design could The use of these chemical systems and of mechanis- enhance the possibility of developing retic models advanced to explain the unique histo- mineralizing solutions useful at the clinical logical features of incipient caries have yielded a good understanding of the processes involved in level. caries formation. The study of natural and induced factors influencing the demineralization process has been greatly facilitated by the use of bacterio- Demineralization. logical systems in which demineralization is proEnamel demineralization under oral conduced by direct colonization of cariogenic microis generally accepted to result from ditions organisms on the surfaces of extracted teeth. Comparison of results obtained with these latter acid conditions produced by the metabolism of cariogenic microorganisms in dental systems and with chemical systems has allowed us, .for example, to elucidate the mechanism by which plaque. However, the formation of a dental acquired salivary pellicles and fluoride topical caries is more complex than a simple dissolutions decrease the rate of enamel demineraliza- solution process of the hard tissue. The tion. The pellicle retards transport of matter across striking histological feature of the incipient the enamel surface, whereas the fluoride topical lesion is the presence of a relatively sound solutions decrease the cariogenicity of the colonizsurface layer of enamel overlying the deing bacteria.
Remineralization of incipient carious lesions occurs naturally in the oral environment when the cariogenic challenge is minimized. In vitro experimentation shows that remineralization can be attained by exposure of demineralized enamel to solutions containing calcium and phosphate ions. The remineralization rate appears to be proportional to the degree of supersaturation of the solutions with respect to hydroxyapatite and it is enhanced by fluoride at low concentrations. These observations are significant because they are very similar to those reported in relation to crystal growth rates in supersaturated solutions
seeded with hydroxyapatite crystals. In spite of the considerable volume of literature dealing with remineralization in vitro, there are not enough clinical studies to assess the possible benefits of synthetic remineralizing solutions. Furthermore, it seems that many laboratory studies have been conducted, paying little attention to specific oral factors which may affect remineralization, e.g., the presence of enamel pellicles and salivary macromolecules that actually inhibit the formation of basic
mineralized zone, i.e., the bulk of the mineral loss in the early stages of demineralization occurs at some distance from the enamel surface. For this reason the term "subsurface lesion" is used to describe this stage of the process. Another term widely used in this connection is "white spot", which describes the appearance of the affected enamel surface. Detailed histological descriptions of incipient caries lesions have been given by numerous investigators. However, comprehensive models explaining the controlling factors in the development of these features are relatively few. Some of these modelst were developed on the basis of diffusion theory under steady-state conditions and were used to relate the rate of enamel demineralization to factors such as pH, and acid concentrations in in vitro experimentation, using organic acid buffers as demineralizing media. The presence of the unaltered surface enamel is explained then by ascribing special properties to this enamel region, e.g., resistance to acid dissolution by adsorption of organic matter onto the enamel crystallites. Other models2 have been developed on the basis of the present
896 Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
Vol. 58(B)
897
knowledge of the chemistry of calcium phosphates. In these latter cases, the importance of kinetics is recognized, but the mechanism advanced depends on the driving forces operating on the system, i.e., degrees of saturation with respect to various calcium phosphates. The relatively sound enamel layer at the surface is then explained by a reprecipitation process occurring at the expense of the underlying enamel. The subsurface demineralization patterns produced in vitro by acid buffers are comparable to those found in natural lesions, although the morphologies of the demineralized zones are often different. In the artificial lesions the demineralization front is generally smooth and parallel to the enamel surface, whereas in the natural lesions this front displays a typical serrated morphology. However, if the demineralization with the acid buffer is controlled so as to cause a very slow process, the resulting morphology is quite comparable to that of a natural
lesion.3 According to microscopic studies, it appears that, in vivo, the first mineral lost in enamel demineralization is that present in the interprismatic regions;4 the dissolution proceeds then through crossstriations into the enamel prisms, and continues along the prism cores. A similar conclusion has been reached for the case of demineralization produced by acid buffers in vitro; the conclusion in the case based on porosity measurements and pore volume distributions obtained at various demineralization stages using water vapor adsorption techniques.5 With these techniques, however, increases in porosity have been observed that can not be attributed simply to mineral dissolution; they coincide with increases in both the specific surface area and the pore volume occupied by pores with the smaller radii in the bimodal distribution of dental enamel.5 This point will be discussed further in relation to enamel remineralization. The laboratory models using acid buffers have been very useful in studying the processes responsible for subsurface demineralization. Under oral conditions, however, enamel demineralization is affected by a number of factors, either of natural occurrence or specifically induced to check demineralization. In the latter case, it is not easy to determine whether demineralization is affected by changes in the physicochemical
properties of the enamel mineral or by modification in the microbial activity. For this reason, models have been developed that have a closer resemblance to the oral environment, i.e., artificial mouths. In these models demineralization is induced by acid-produced metabolically by microorganisms. A modification of this approach is a bacteriological model which involves the use of pure cultures of microorganisms.6 In this case colonization by the microoganisms is confined to a window of defined area on smooth surfaces of extracted teeth, the rest of the tooth surface being protected by dental wax. After a period of several days, depending on the bacterial strain used, subsurface enamel demineralization is observed and the caries-like features are quite similar to those obtained with acid buffers. We have reported6 results obtained with two strains of Streptococcus mutans. The combination of the chemical (acid buffers) and bacteriological models is a useful tool to investigate the mechanism by which induced modifications of the tooth surface are effective in reducing caries. One example concerns the effectiveness of fluoride topical solutions. It has been found7 that extracted teeth treated with acidulated phosphate-fluoride solutions (APF) or neutral sodium fluoride undergo demineralization by acid buffers to the same extent as untreated teeth. However, the same treatments result in a substantial decrease in demineralization, in comparison with untreated teeth, whenthe tooth surfaces are colonized by cariogenic microorganisms. Representative data illustrating this point are shown in table 1. The observed protection is lost when the treated tooth surfaces are exposed to a concentrated solution of KOH prior to microbial colonization. The alkaline treatment removes CaF2 formed during the topical treatment. Thus, the results indicate that the protection against demineralization by the microorganisms is related to the presence of CaF2 at the surface of the enamel. The presence of this salt likely affects both the rate of cell attachment7 and the glycolytic metabolism8 of the cariogenic bacteria. In the past, little consideration has been given to the presence of the acquired tooth pellicle in relation to enamel demineralization. Nevertheless, it is to this integument that the microbial cells attach and it is through this pellicle that acidity and dis-
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
898
MORENO & ZAHRADNIK
J Dent Res Special Issue B, 1 9 79
TABLE 1 DEMINERALIZATION OF APF-TREATED TEETH IN TWO IN VITRO SYSTEMS
Control
Depth of lesion (pm) Maximum mineral loss (%) Intact layer thickness (pum)
Lactic Acid* APF-1 hr
106 43 36
102 43.5 37
IB1600* Control APF-1 hr
130 45 40
102 8 38
*Exposure to lactate buffer for 72 hr. Duration of bacteriological test, 5 days.
solved enamel must diffuse. Thus, there are at least two ways in which this integument may affect caries formation: modifying cell attachment and imposing limitations on the kinetics of the demineralization process. The present consensus is that this pellicle is formed, at least in its initial stages, by selective adsorption of salivary peptides and glycoproteins. Acquired pellicles can be developed in vitro by exposing extracted teeth to saliva. We have reported9 on some of the electrochemical properties of these pellicles, i.e., a significant ionic permselectivity. This property has been related to the marked protection that acquired pellicles (developed in vitro) confer to enamel against demineralization by acid buffers.9 Furthermore, both the ionic permselectivity and the degree of enamel protection increase with the pellicle development time; in fact, these properties are observed only with pellicles developed for more than several days. These latter fmdings are consistent with the behavior of pellicles developed in situ in relation to enamel erosion by mineral acids.10 Although the changes in the pellicle properties with developing time are not fully understood at present, most probably they are related to changes in its composition and ultrastructure. In terms of the model previously proposed2 to explain the formation of a subsurface enamel lesion, the acquired pellicle affects the kinetics of the process without altering the driving forces involved to a significant extent. It is conceivable that an understanding of the pellicle formation and its properties at the molecular level may lead to the development of agents more effective than the natural pellicle to check enamel demineralization. Until recently, there was no information concerning the possible protective action of the acquired pellicle under oral conditions. Elucidation of this point using in vivo
conditions is practically impossible. However, the use of the bacteriological model has made possible some advances in this direction. We recently have found that a significant reduction in enamel demineralization by S. mutans (IB 1600) is attained when the organism is allowed to colonize the tooth surface after development of a long-term (7 days) salivary pellicle. However, this kind of protection was not observed with short-term pellicles (2 hours). Representative data of the experiments performed are shown in table 2. The results of other experiments indicated that the reduction in demineralization reported in Table 2 could not be attributed to a reduction in the initial rates of cell attachment. These observations are consistent with the idea that in the demineralization induced by cariogenic organisms, the pellicle is affecting the process in a similar fashion, as in the case of demineralization by acid buffers, i.e., by slowing ionic transport. Under oral conditions, however, with a mixed microflora, it is possible that the acquired pellicle may also affect the course of demineralization by selective colonization. Other investigatorsll have shown that the attachment of bacterial cells to apatite crystals and powdered enamel mineral is affected by saliva pretreatment of the apatitic surfaces; for some strains there is an enhancement, whereas there is a decrease for others. At present, the features of enamel demineralization have been well defined and a good understanding exists of the mechanisms involved. Furthermore, chemical and bacteriological models are available with which natural and induced factors affecting the demineralization process can be investigated. In this respect, the bacteriological model is particularly appealing because it makes possible the investigation of factors affecting the enamel mineral as well as those altering the actual source of acidity, i.e.,
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
899
Vol. 58(B)
the microbial metabolism.
Remineralization. A voluminous literature on enamel remineralization has been produced in the last 15 years. Significantly, the large majority of the reports pertain to in vitro work. However, there are reports on clinical studies12 showing that incipient lesions on smooth and proximal surfaces reverse spontaneously. Thus, white spots induced by suspension of oral hygiene and concomitant mouth rinses with sucrose solutions revert to sound enamel after discontinuation of the rinses, reinitiation of common hygiene practices, and the use of fluoride-containing rinses.13 Evidently, when the cariogenic challenge is minimized, the conditions in the oral environment bring about remineralization of incipient carious lesions. This result could be anticipated to a large extent; the levels of calcium and phosphate in whole saliva (and in plaque fluid) indicate supersaturation with respect to the enamel mineral at the usual pH values of saliva or resting plaque, and undersaturation with respect to the more acid calcium phosphate, dicalcium phosphate dihydrate, that could have formed during the demineralizing period3 However, inasmuch as the remineralization does not ordinarily proceed beyonrd the geometric surface of the tooth, it appears that the tooth-saliva interface possesses unique properties to which not enough attention has been paid in relation to remineralization. The remineralizing potential of saliva appears to be part of a built-in mechanism which, under some conditions, counteracts demineralization produced by cariogenic challenges. The effectiveness of such a mechanism, however, is limited, since caries in humans do occur. The possibility of improvement upon the natural remineralizing mechanism has been examined by several investigators using laboratory models in which conditions can be better controlled than in vivo. Typically, these models may involve the use of extracted teeth with natural lesions, with caries-like lesions induced by acid buffers, or pretreated with a dilute acid solution to produce "softened" enamel. These teeth are then exposed to remineralizing solutions of various compositions under different frequency regimes.
The treatment effectiveness is assessed by measuring the changes in some of the physical properties of the tissue exposed to the remineralizing medium. Common techniques for this assessment are optical and electron microscopy, microhardness, Compositional and microradiography. changes in the remineralizing solution have also been used to estimate the mineral deposition on demineralized enamel, although, in this case, the actual effect upon the enamel is, by force, conjectural. In general, the results of experimentation in this area indicate that repair of the demineralized tissue can be accomplished through the use of simple solutions containing calcium and phosphate ions at levels comparable to those present in whole saliva.14'15 The extent of the repair is usually not complete within the experimental times used. In some cases, deposition of mineral at the tooth surface has been reported, but in others there is convincing evidence that substantial mineral deposition can occur in the bulk of the lesion in the subsurface region.15 The presence of fluoride in the remineralizing solution at concentrations as low as 1 ppm increases markedly the enamel remineralization rates; this fmding is based both on microscopic observations1s and microhardness measurements.16 This phenomenon is similar to the reported effect that fluoride has on the kinetics of crystal growth of hydroxyapatite (HA); in this latter case, however, it appears that acceleration or retardation may occur depending upon the fluoride concentration.17 Fluoride may increase the rate of remineralization by preventing the formation of intermediate calcium phosphates more acid than apatites and by increasing the driving force for precipitation (supersaturation with respect to fluorapatite).
When solutions of variable composition have been used in enamel remineralization studies, it has been found that the rate of remineralization (assessed by rehardening) is proportional to the degree of supersaturation of the solution with respect to HA.16 This result is significant because it parallels the observations in in vitro systems in which the crystal growth rates of HA in seeded solutions are also proportional to the degree of supersaturation of the solution with respect to HA.17 Evidently some of the basic principles governing the rate of enamel remineralization can be studied through the
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
900
J Dent Res Special Issue B, 1 9 79
MORENO & ZAHRADNIK
TABLE 2 ENAMEL DEMINERALIZATION AFTER 5 DAY INCUBATION WITH IB1600
Control
Depth of lesion ( m) Maximum mineral loss (% Intact layer thickness (m)
2-hr Pellicle Pellicle
117 26 33
of laboratory models designed to study the precipitation or crystal growth of synthetic apatites. Furthermore, the effects of crystal growth accelerators or retardants can be better studied in the latter systems where accurate and precise observations are not hampered by variability of the enamel composition and its heterogeneity. A limitation of the models using seeded solutions is that the restrictions due to enamel structure can not be taken into account. Nevertheless, the systems with synthetic materials can be used to investigate remineralization problems related to specific factors present in the oral environment. For example, it has been reportedl5 that solutions containing calcium and phosphate at comparable levels of whole saliva are more effective remineralizing agents than the saliva itself. Probable explanations for this observation are the binding of calcium and/or phosphate by salivary constituents, and the presence of inhibitors for crystal growth or nucleation.1 9 In fact, we have recently found that very low concentrations of individual peptides and proteins isolated from human saliva greatly affect the rate of crystal growth in solutions seeded with HA. In one experiment, for example, the experimental solution (250 ml) has initial calcium and phosphorus concentrations of 1.1 x 10-3M and 6.4 x 10-4M, respectively, at a pH of 7.4; HA, 25 mg, was added and the changes in the solution compositions were determined at successive intervals thereafter. In figure 1 is plotted the concentration of calcium in solution versus the time after the addition of the seeds. From the figure, it can be seen that the presence of statherin19 (a tyrosine-rich acidic peptide isolated from parotid saliva, M.W. 5380) at a concentration as low as 0.4 ,M (its concentration in parotid saliva is in the order of 5 MM) brings about a substantial change in the kinetics of the process. At present, this phenomenon use
122 25 30
7-day Pellicle Control 103 32 17
Pellicle 103 18 29
be related to adsorption of the macromolecule onto active growth sites of the crystalline seeds.
appears to
Fig. 1 - Effect of statherin on the crystal growth of hydroxyapatite.
The usefulness of laboratory models involving synthetic apatites should not be underestimated, but their limitations should be clearly recognized. The repairing process requires the penetration of ionic species into the bulk of the lesion and not a simple precipitation on the enamel surface. This situation poses a significant problem, particularly in connection with possible clinical applications; by increasing the calcium and phosphorus concentrations in a remineralizing solution, the rate of mineral deposition may be increased, but the possibility of having precipitation on the surface of the tooth is also augmented. Surface deposition is not remineralization; in addition, its presence restricts the penetration required to remineralize the bulk of the lesion in the subsurface region. Therefore, the effectiveness of a remineralizing agent requires direct in vitro observations or measurements on the tissue itself. Furthermore, the models used in this connection should include features
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
901
Vol. 58(B) present in the oral environment that may play an important role in enamel remineralization. Thus, the acquired pellicle markedly decreases ionic transport, which should slow down remineralization. However, this integument may also have desirable properties; we have observed recently that the presence of a salivary pellicle prevents the deposition of mineral on the surface of extracted teeth when exposed to remineralizing solutions. The ultimate goal of investigations on enamel remineralization is a possible clinical application. It is then necessary to recognize that remineralization of enamel in the laboratory has been invariably accomplished under optimal conditions, i.e., calcium and phosphorus concentrations as high as the chemistry of calcium phosphates permits, high pH values, and almost total disregard for the factors that may favor the reverse process. There is no information, at present, on those factors determining the reversibility of an incipient carious lesion; in this context, it is pertinent to recall the demineralization stages described before in this paper. The marked increase in the pore volume accessible to water vapor that occurs during the demineralization process has been ascribed5 to an alteration of the organic matrix that, essentially, gives full accessibility to intraprismatic pores and marks the loss of order in the enamel structure. After this event, it may be extremely difficult to reconstruct the tissue with all its original properties. In other words, the alteration of the organic matrix may be the limiting factor for the reversibility of an incipient carious lesion. This possibility concerning reversibility of a lesion is based only on conjectural thought, but it deserves consideration and experimental effort. Similarly, the possibility of enhancing the natural remineralizing mechanism should be investigated. In fact, the effectiveness of fluoride mouth rinses in caries prevention may derive from such an enhancement. In this case, fluoride may have a multiple effect, e.g., decreasing the acid production of cariogenic microorganisms8 (so that the mineralizing potential of saliva becomes effective), increasing the natural rate of remineralization, and enhancing the stability of the enamel mineral. The clinical literature on the possible benefits of remineralizing solutions is very scanty. A recent report20 indicates that the
reversal rate of incipient lesions (induced by rinses with sucrose solutions without oral hygiene practices), was increased by the frequent use of a remineralizing solution containing calcium, phosphate and fluoride. The study20 was conducted with a small number (15) of subjects and the differences between the experimental and control groups were notvery great. Perhaps real advancements in this field are hampered by the limitations imposed by the chemistry of calcium phosphates (metastability of the solutions proposed) and the difficulty in separating the effects of remineralizing agents from those associated with the oral fluids. The lack of a marked success at the clinical level may imply that the simple approaches using solutions containing calcium, phosphorus, and fluoride are not adequate. A closer look at the natural remineralization process may yield useful information concerning factors present in the oral environment which are involved in remineralization. It is conceivable that organic or inorganic additives may accelerate the rate of mineral deposition at the needed location under oral conditions. REFERENCES 1. HOLLY, F. J., GRAY, J. A.: Arch. Oral Biol. 13:319-334 (1968). 2. MORENO, E. C., ZAHRADNIK, R. T.: J. Dent. Res. 53:226-235 (1974). 3. SILVERSTONE, L. M.: Caries Res. (Suppl 1), 11:59-84 (1977). 4. DARLING, A. I.: In: Mechanisms of Hard Tissue Destruction (SOGNNAES, R. F., ed) pp. 171-184. American Association for the Advancement of Science, Washington, 1963. 5. ZAHRADNIK, R. T., MORENO, E. C.: Arch. OralBio. 22:585-591 (1977). 6. ZAHRADNIK, R. T., PROPAS, D., MORENO, E. C.: J. Dent. Res. 56:1107-1110 (1977). 7. ZAHRADNIK, R. T., PROPAS, D., MORENO, E. C.: J. Dent. Res. In press. 8. JENKINS, G. N.: J. Dent. Res. 42:444452 (1963). 9. ZAHRADNIK, R. T., MORENO, E. C., BURKE, E. J.: J. Dent. Res. 55:664-670 (1976). 10. TURNER, E. P.: Dent. Practitioner 8:373382 (1958). 11. CLARK, W. B., BAMMANN, L. L., GIBBONS, R. J.: Infect. Immun. In press. 12. BACKER-DIRKS, O.: J. Dent. Res. 45:
503-511, (1966).
13. VON DER FEHR, F. R., LOE, H., THEILADE, E.: Caries Res. 4:131-148 (1970). 14. KOULOURIDES, T.: Ann. N. Y. Acad. Sci.
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
902
MORENO & ZAHRADNIK
153:84-101 (1968). 15. SILVERSTONE, L. M., POOLE, D. F. G.: Caries Res. 2: 87-96 (1968). 16. FEAGIN, R., PATEL, P. R., KOULOURIDES, T., PIGMAN, W.: Arch. Oral Bio. 16:535548 (1971). 17. MEYER, J. L., NANCOLLAS, G. H.: J. Dent.
J Dent Res Special Issue B, 1 9 79 Res. 51:1443-1450 (1972). 18. MORENO, E. C., ZAHRADNIK, R. T., GLAZMAN, A., HWU, R.: Calcif. Res. 24:47-57
(1977). 19. SCHLESINGER, D. H., HAY, D. I.: J. Biol. Chem. 252:1689-1695 (1977). 20. LEVINE, R. S.: Brit. Dent. J. 138:249-253
(1975).
Session VI Discussion Dr. Driessens: I was impressed, especially by your measurements of the radius of the pores. In our laboratory we determined in the past few years the diffusion coefficients for 45Ca2+, 86Rb+, 36C1-, 3H-sorbitol and 14C-glycerol. We found different results for the different particles, depending on both their size and charge. This means that there are certain interactions of the enamel membrane with the particles diffusing through the pores. The first effect is that of ion selectivity. We also could detect a certain molecular sieve effect, especially for sorbitol and glycerol as compared to the average for the ions. If we account for the differences in radius, and if we average over the ion selectivity effect, we end up with the same range of values for the diameters of the pores as you have found with the BET method. We have also carried out a computer simulation of the caries process. We first worked with Nernst-Planck flux equations for each of the particles, without making additional assumptions, and then tried to obtain a subsurface lesion. We did not succeed. We had to include an extra condition, an extra term giving consideration to a gradient in the rate of dissolution of the mineral. Otherwise, we did not get subsurface demineralization but etching. We found in our simulation program that changes in the ion selectivity as small as found in your experiments with pellicle adsorption did not influence the rate of demineralization more than a few percent, still resulting in etching. This might be important for the interpretation of your results, as you found subsurface demineralization after adsorption of pellicle. The only way to explain this result, it seems to me, is that the adsorption of pellicle changes the rate of dissolution of the mineral in the outer layer very considerably.
Dr. Moreno: That is essentially what we have shown. In other words, in the presence of a pellicle, you control the transport of matter from the tooth surface into the demineralizing medium. Dr. Driessens: Well, you gave me another impression, because in your presentation you said that the pellicle had a marked effect on the rate of diffusion which is different from the rate of dissolution of the mineral. Dr. Moreno: Well, that is correct. If the enamel mineral dissolves, its constituents in solution will have to move, through the interface, into the demineralizing medium; we maintain that the pellicle, because of its ionic permselective properties, slows down this process. Dr. Driessens: We found that you must take into account both the rate of diffusion and the rate of dissolution in order to simulate the caries process. Dr. Moreno: You mentioned something that is interesting, that in your simulation program, you couldn't account for your results by differences in the permselective properties of the enamel interface. I would suggest that you review your program because, experimentally, we can produce subsurface demineralization by the modification of these properties, i.e., by forming a salivary pellicle at the enamel surface. Dr. Driessens: We did simulate this formation of the intact layer, but then we had to assume either a gradient in the pore density, or a gradient in the solubility product, or a gradient in the dissolution rate constant. Dr. Silverstone: In one of your early tables, you showed the depth of lesion in respect to different aged pellicles, and you
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
903
Vol. 58(B) noted there was hardly any difference in depth. Dr. Moreno: That is correct. Dr. Silverstone: And you were somewhat disappointed. Can I suggest that this is because you are limiting yourself to what I consider to be a relatively crude technique, microradiography? When you're measuring the depth of the lesion, what in fact you're measuring is the depth of zone number three, the body of the lesion. So, in other words, you're not looking at the translucent zone or the dark zone, both of which together could be a hundred microns. This, in fact, could make quite a difference. Dr. Moreno: That is right. I could not detect those through this technique. Dr. Termine: Could you elucidate further on your studies of pellicle effects in remineralization, particularly in light of the very dramatic effects you obtained in the demineralization studies? Dr. Moreno: We can illustrate this point with the results obtained in kinetic studies of crystal growth using seeded solutions supersaturated with respect to hydroxyapatite; these results are shown in a graph included in the paper. It is clear that a concentration of statherin of 0.4 um brings about a substantial reduction in the precipitation rate of the seeded solution. The mean concentration of statherin in parotid saliva is about 4 um. Thus, concentrations about one-tenth of that in the glandular secretion reduce drastically the rate of crystal growth. This is a phenomenon that we have not elucidated completely but we feel that more consideration should be given to it in relation to enamel remineralization. Dr. Ten Bosch: You have been showing those results on a seven-day old pellicle. I think that is a very long time for a pellicle to get formed; do you have any idea how thick it is, and what material it is? Dr. Moreno: We frankly do not know, but we think that some structural changes in the pellicle occur during the seven-day period of development. We have conducted obvious experiments; for example, exposure of enamel to fresh saliva every hour for
seven consecutive times fails to develop a pellicle with the permselective properties obtained when the enamel is exposed for seven days using fresh saliva every 24 hours. Similar failure is observed when the enamel is exposed to saliva for 24 hours and then is kept for seven days in a moist atmosphere. So, some undetermined changes occur during the seven-day development period and perhaps people in the salivary field could shed some light on this. Dr. Weill: Speaking of remineralization, do you really mean that a new apatite is formed? If not, it is deposition of mineral and not remineralization. Further, I think that there are distinct differences between in vivo and in vitro experiments. If you look at human carious enamel you find that the core of the prism is invaded by stainable material that is quite different from the normal enamel matrix. Dr. Moreno: We don't know if the remineralization proceeds to the core of the prism or if it does, if it proceeds to form crystals with the same order that the crystallites had prior to demineralization. This is something that should be investigated. Dr, Nancollas: It would be very interesting to know whether the statherin does in fact affect the mechanism of mineralization, or whether it simply affects the rate. We found that there are very few active sites on crystals which are actually growing. Dr. Moreno: That is correct. Dr. Nancollas: We have done some rotating disc dissolution work involving laminar flow of both whole and white spot enamel. The decrease in rate in the latter experiments can be interpreted in terms of different diffusion rates through the surface phases. Dr. Moreno: The only answer I can give is that we have conducted this kind of experiment at different concentrations of statherin. Our results, so far, match very well with the adsorption studies that we have done with statherin on hydroxyapatite, so that a decrease in the initial rate of precipitation is proportional to the amount of statherin adsorbed on the apatite seeds.
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
Demineralization and Remineralization of Dental Enamel _____ E.C. Moreno and R.T. Zahradnik J DENT RES 1979 58: 896 DOI: 10.1177/00220345790580024301 The online version of this article can be found at: http://jdr.sagepub.com/content/58/2_suppl/896
Published by: http://www.sagepublications.com
On behalf of: International and American Associations for Dental Research
Additional services and information for Journal of Dental Research can be found at: Email Alerts: http://jdr.sagepub.com/cgi/alerts Subscriptions: http://jdr.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jdr.sagepub.com/content/58/2_suppl/896.refs.html
>> Version of Record - Feb 1, 1979 What is This?
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
Demineralization and Remineralization of Dental Enamel E. C. MORENO and R. T. ZAHRADNIK Forsyth Dental Center, 140 The Fenway, Boston, Massachusetts 02115, U.S.A. J Dent Res 58(B):896-902, March 1979 Experimentation in vitro using organic acid buffers calcium phosphates. This inhibition has been demonstrated for some macromoleas demineralizing media shows that caries-like lesions can be obtained which are very similar in cules at concentrations much lower than morphology and developmental stages to early those present in saliva. Consideration of oral lesions formed naturally under oral conditions. conditions in experimental design could The use of these chemical systems and of mechanis- enhance the possibility of developing retic models advanced to explain the unique histo- mineralizing solutions useful at the clinical logical features of incipient caries have yielded a good understanding of the processes involved in level. caries formation. The study of natural and induced factors influencing the demineralization process has been greatly facilitated by the use of bacterio- Demineralization. logical systems in which demineralization is proEnamel demineralization under oral conduced by direct colonization of cariogenic microis generally accepted to result from ditions organisms on the surfaces of extracted teeth. Comparison of results obtained with these latter acid conditions produced by the metabolism of cariogenic microorganisms in dental systems and with chemical systems has allowed us, .for example, to elucidate the mechanism by which plaque. However, the formation of a dental acquired salivary pellicles and fluoride topical caries is more complex than a simple dissolutions decrease the rate of enamel demineraliza- solution process of the hard tissue. The tion. The pellicle retards transport of matter across striking histological feature of the incipient the enamel surface, whereas the fluoride topical lesion is the presence of a relatively sound solutions decrease the cariogenicity of the colonizsurface layer of enamel overlying the deing bacteria.
Remineralization of incipient carious lesions occurs naturally in the oral environment when the cariogenic challenge is minimized. In vitro experimentation shows that remineralization can be attained by exposure of demineralized enamel to solutions containing calcium and phosphate ions. The remineralization rate appears to be proportional to the degree of supersaturation of the solutions with respect to hydroxyapatite and it is enhanced by fluoride at low concentrations. These observations are significant because they are very similar to those reported in relation to crystal growth rates in supersaturated solutions
seeded with hydroxyapatite crystals. In spite of the considerable volume of literature dealing with remineralization in vitro, there are not enough clinical studies to assess the possible benefits of synthetic remineralizing solutions. Furthermore, it seems that many laboratory studies have been conducted, paying little attention to specific oral factors which may affect remineralization, e.g., the presence of enamel pellicles and salivary macromolecules that actually inhibit the formation of basic
mineralized zone, i.e., the bulk of the mineral loss in the early stages of demineralization occurs at some distance from the enamel surface. For this reason the term "subsurface lesion" is used to describe this stage of the process. Another term widely used in this connection is "white spot", which describes the appearance of the affected enamel surface. Detailed histological descriptions of incipient caries lesions have been given by numerous investigators. However, comprehensive models explaining the controlling factors in the development of these features are relatively few. Some of these modelst were developed on the basis of diffusion theory under steady-state conditions and were used to relate the rate of enamel demineralization to factors such as pH, and acid concentrations in in vitro experimentation, using organic acid buffers as demineralizing media. The presence of the unaltered surface enamel is explained then by ascribing special properties to this enamel region, e.g., resistance to acid dissolution by adsorption of organic matter onto the enamel crystallites. Other models2 have been developed on the basis of the present
896 Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
Vol. 58(B)
897
knowledge of the chemistry of calcium phosphates. In these latter cases, the importance of kinetics is recognized, but the mechanism advanced depends on the driving forces operating on the system, i.e., degrees of saturation with respect to various calcium phosphates. The relatively sound enamel layer at the surface is then explained by a reprecipitation process occurring at the expense of the underlying enamel. The subsurface demineralization patterns produced in vitro by acid buffers are comparable to those found in natural lesions, although the morphologies of the demineralized zones are often different. In the artificial lesions the demineralization front is generally smooth and parallel to the enamel surface, whereas in the natural lesions this front displays a typical serrated morphology. However, if the demineralization with the acid buffer is controlled so as to cause a very slow process, the resulting morphology is quite comparable to that of a natural
lesion.3 According to microscopic studies, it appears that, in vivo, the first mineral lost in enamel demineralization is that present in the interprismatic regions;4 the dissolution proceeds then through crossstriations into the enamel prisms, and continues along the prism cores. A similar conclusion has been reached for the case of demineralization produced by acid buffers in vitro; the conclusion in the case based on porosity measurements and pore volume distributions obtained at various demineralization stages using water vapor adsorption techniques.5 With these techniques, however, increases in porosity have been observed that can not be attributed simply to mineral dissolution; they coincide with increases in both the specific surface area and the pore volume occupied by pores with the smaller radii in the bimodal distribution of dental enamel.5 This point will be discussed further in relation to enamel remineralization. The laboratory models using acid buffers have been very useful in studying the processes responsible for subsurface demineralization. Under oral conditions, however, enamel demineralization is affected by a number of factors, either of natural occurrence or specifically induced to check demineralization. In the latter case, it is not easy to determine whether demineralization is affected by changes in the physicochemical
properties of the enamel mineral or by modification in the microbial activity. For this reason, models have been developed that have a closer resemblance to the oral environment, i.e., artificial mouths. In these models demineralization is induced by acid-produced metabolically by microorganisms. A modification of this approach is a bacteriological model which involves the use of pure cultures of microorganisms.6 In this case colonization by the microoganisms is confined to a window of defined area on smooth surfaces of extracted teeth, the rest of the tooth surface being protected by dental wax. After a period of several days, depending on the bacterial strain used, subsurface enamel demineralization is observed and the caries-like features are quite similar to those obtained with acid buffers. We have reported6 results obtained with two strains of Streptococcus mutans. The combination of the chemical (acid buffers) and bacteriological models is a useful tool to investigate the mechanism by which induced modifications of the tooth surface are effective in reducing caries. One example concerns the effectiveness of fluoride topical solutions. It has been found7 that extracted teeth treated with acidulated phosphate-fluoride solutions (APF) or neutral sodium fluoride undergo demineralization by acid buffers to the same extent as untreated teeth. However, the same treatments result in a substantial decrease in demineralization, in comparison with untreated teeth, whenthe tooth surfaces are colonized by cariogenic microorganisms. Representative data illustrating this point are shown in table 1. The observed protection is lost when the treated tooth surfaces are exposed to a concentrated solution of KOH prior to microbial colonization. The alkaline treatment removes CaF2 formed during the topical treatment. Thus, the results indicate that the protection against demineralization by the microorganisms is related to the presence of CaF2 at the surface of the enamel. The presence of this salt likely affects both the rate of cell attachment7 and the glycolytic metabolism8 of the cariogenic bacteria. In the past, little consideration has been given to the presence of the acquired tooth pellicle in relation to enamel demineralization. Nevertheless, it is to this integument that the microbial cells attach and it is through this pellicle that acidity and dis-
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
898
MORENO & ZAHRADNIK
J Dent Res Special Issue B, 1 9 79
TABLE 1 DEMINERALIZATION OF APF-TREATED TEETH IN TWO IN VITRO SYSTEMS
Control
Depth of lesion (pm) Maximum mineral loss (%) Intact layer thickness (pum)
Lactic Acid* APF-1 hr
106 43 36
102 43.5 37
IB1600* Control APF-1 hr
130 45 40
102 8 38
*Exposure to lactate buffer for 72 hr. Duration of bacteriological test, 5 days.
solved enamel must diffuse. Thus, there are at least two ways in which this integument may affect caries formation: modifying cell attachment and imposing limitations on the kinetics of the demineralization process. The present consensus is that this pellicle is formed, at least in its initial stages, by selective adsorption of salivary peptides and glycoproteins. Acquired pellicles can be developed in vitro by exposing extracted teeth to saliva. We have reported9 on some of the electrochemical properties of these pellicles, i.e., a significant ionic permselectivity. This property has been related to the marked protection that acquired pellicles (developed in vitro) confer to enamel against demineralization by acid buffers.9 Furthermore, both the ionic permselectivity and the degree of enamel protection increase with the pellicle development time; in fact, these properties are observed only with pellicles developed for more than several days. These latter fmdings are consistent with the behavior of pellicles developed in situ in relation to enamel erosion by mineral acids.10 Although the changes in the pellicle properties with developing time are not fully understood at present, most probably they are related to changes in its composition and ultrastructure. In terms of the model previously proposed2 to explain the formation of a subsurface enamel lesion, the acquired pellicle affects the kinetics of the process without altering the driving forces involved to a significant extent. It is conceivable that an understanding of the pellicle formation and its properties at the molecular level may lead to the development of agents more effective than the natural pellicle to check enamel demineralization. Until recently, there was no information concerning the possible protective action of the acquired pellicle under oral conditions. Elucidation of this point using in vivo
conditions is practically impossible. However, the use of the bacteriological model has made possible some advances in this direction. We recently have found that a significant reduction in enamel demineralization by S. mutans (IB 1600) is attained when the organism is allowed to colonize the tooth surface after development of a long-term (7 days) salivary pellicle. However, this kind of protection was not observed with short-term pellicles (2 hours). Representative data of the experiments performed are shown in table 2. The results of other experiments indicated that the reduction in demineralization reported in Table 2 could not be attributed to a reduction in the initial rates of cell attachment. These observations are consistent with the idea that in the demineralization induced by cariogenic organisms, the pellicle is affecting the process in a similar fashion, as in the case of demineralization by acid buffers, i.e., by slowing ionic transport. Under oral conditions, however, with a mixed microflora, it is possible that the acquired pellicle may also affect the course of demineralization by selective colonization. Other investigatorsll have shown that the attachment of bacterial cells to apatite crystals and powdered enamel mineral is affected by saliva pretreatment of the apatitic surfaces; for some strains there is an enhancement, whereas there is a decrease for others. At present, the features of enamel demineralization have been well defined and a good understanding exists of the mechanisms involved. Furthermore, chemical and bacteriological models are available with which natural and induced factors affecting the demineralization process can be investigated. In this respect, the bacteriological model is particularly appealing because it makes possible the investigation of factors affecting the enamel mineral as well as those altering the actual source of acidity, i.e.,
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
899
Vol. 58(B)
the microbial metabolism.
Remineralization. A voluminous literature on enamel remineralization has been produced in the last 15 years. Significantly, the large majority of the reports pertain to in vitro work. However, there are reports on clinical studies12 showing that incipient lesions on smooth and proximal surfaces reverse spontaneously. Thus, white spots induced by suspension of oral hygiene and concomitant mouth rinses with sucrose solutions revert to sound enamel after discontinuation of the rinses, reinitiation of common hygiene practices, and the use of fluoride-containing rinses.13 Evidently, when the cariogenic challenge is minimized, the conditions in the oral environment bring about remineralization of incipient carious lesions. This result could be anticipated to a large extent; the levels of calcium and phosphate in whole saliva (and in plaque fluid) indicate supersaturation with respect to the enamel mineral at the usual pH values of saliva or resting plaque, and undersaturation with respect to the more acid calcium phosphate, dicalcium phosphate dihydrate, that could have formed during the demineralizing period3 However, inasmuch as the remineralization does not ordinarily proceed beyonrd the geometric surface of the tooth, it appears that the tooth-saliva interface possesses unique properties to which not enough attention has been paid in relation to remineralization. The remineralizing potential of saliva appears to be part of a built-in mechanism which, under some conditions, counteracts demineralization produced by cariogenic challenges. The effectiveness of such a mechanism, however, is limited, since caries in humans do occur. The possibility of improvement upon the natural remineralizing mechanism has been examined by several investigators using laboratory models in which conditions can be better controlled than in vivo. Typically, these models may involve the use of extracted teeth with natural lesions, with caries-like lesions induced by acid buffers, or pretreated with a dilute acid solution to produce "softened" enamel. These teeth are then exposed to remineralizing solutions of various compositions under different frequency regimes.
The treatment effectiveness is assessed by measuring the changes in some of the physical properties of the tissue exposed to the remineralizing medium. Common techniques for this assessment are optical and electron microscopy, microhardness, Compositional and microradiography. changes in the remineralizing solution have also been used to estimate the mineral deposition on demineralized enamel, although, in this case, the actual effect upon the enamel is, by force, conjectural. In general, the results of experimentation in this area indicate that repair of the demineralized tissue can be accomplished through the use of simple solutions containing calcium and phosphate ions at levels comparable to those present in whole saliva.14'15 The extent of the repair is usually not complete within the experimental times used. In some cases, deposition of mineral at the tooth surface has been reported, but in others there is convincing evidence that substantial mineral deposition can occur in the bulk of the lesion in the subsurface region.15 The presence of fluoride in the remineralizing solution at concentrations as low as 1 ppm increases markedly the enamel remineralization rates; this fmding is based both on microscopic observations1s and microhardness measurements.16 This phenomenon is similar to the reported effect that fluoride has on the kinetics of crystal growth of hydroxyapatite (HA); in this latter case, however, it appears that acceleration or retardation may occur depending upon the fluoride concentration.17 Fluoride may increase the rate of remineralization by preventing the formation of intermediate calcium phosphates more acid than apatites and by increasing the driving force for precipitation (supersaturation with respect to fluorapatite).
When solutions of variable composition have been used in enamel remineralization studies, it has been found that the rate of remineralization (assessed by rehardening) is proportional to the degree of supersaturation of the solution with respect to HA.16 This result is significant because it parallels the observations in in vitro systems in which the crystal growth rates of HA in seeded solutions are also proportional to the degree of supersaturation of the solution with respect to HA.17 Evidently some of the basic principles governing the rate of enamel remineralization can be studied through the
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
900
J Dent Res Special Issue B, 1 9 79
MORENO & ZAHRADNIK
TABLE 2 ENAMEL DEMINERALIZATION AFTER 5 DAY INCUBATION WITH IB1600
Control
Depth of lesion ( m) Maximum mineral loss (% Intact layer thickness (m)
2-hr Pellicle Pellicle
117 26 33
of laboratory models designed to study the precipitation or crystal growth of synthetic apatites. Furthermore, the effects of crystal growth accelerators or retardants can be better studied in the latter systems where accurate and precise observations are not hampered by variability of the enamel composition and its heterogeneity. A limitation of the models using seeded solutions is that the restrictions due to enamel structure can not be taken into account. Nevertheless, the systems with synthetic materials can be used to investigate remineralization problems related to specific factors present in the oral environment. For example, it has been reportedl5 that solutions containing calcium and phosphate at comparable levels of whole saliva are more effective remineralizing agents than the saliva itself. Probable explanations for this observation are the binding of calcium and/or phosphate by salivary constituents, and the presence of inhibitors for crystal growth or nucleation.1 9 In fact, we have recently found that very low concentrations of individual peptides and proteins isolated from human saliva greatly affect the rate of crystal growth in solutions seeded with HA. In one experiment, for example, the experimental solution (250 ml) has initial calcium and phosphorus concentrations of 1.1 x 10-3M and 6.4 x 10-4M, respectively, at a pH of 7.4; HA, 25 mg, was added and the changes in the solution compositions were determined at successive intervals thereafter. In figure 1 is plotted the concentration of calcium in solution versus the time after the addition of the seeds. From the figure, it can be seen that the presence of statherin19 (a tyrosine-rich acidic peptide isolated from parotid saliva, M.W. 5380) at a concentration as low as 0.4 ,M (its concentration in parotid saliva is in the order of 5 MM) brings about a substantial change in the kinetics of the process. At present, this phenomenon use
122 25 30
7-day Pellicle Control 103 32 17
Pellicle 103 18 29
be related to adsorption of the macromolecule onto active growth sites of the crystalline seeds.
appears to
Fig. 1 - Effect of statherin on the crystal growth of hydroxyapatite.
The usefulness of laboratory models involving synthetic apatites should not be underestimated, but their limitations should be clearly recognized. The repairing process requires the penetration of ionic species into the bulk of the lesion and not a simple precipitation on the enamel surface. This situation poses a significant problem, particularly in connection with possible clinical applications; by increasing the calcium and phosphorus concentrations in a remineralizing solution, the rate of mineral deposition may be increased, but the possibility of having precipitation on the surface of the tooth is also augmented. Surface deposition is not remineralization; in addition, its presence restricts the penetration required to remineralize the bulk of the lesion in the subsurface region. Therefore, the effectiveness of a remineralizing agent requires direct in vitro observations or measurements on the tissue itself. Furthermore, the models used in this connection should include features
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
901
Vol. 58(B) present in the oral environment that may play an important role in enamel remineralization. Thus, the acquired pellicle markedly decreases ionic transport, which should slow down remineralization. However, this integument may also have desirable properties; we have observed recently that the presence of a salivary pellicle prevents the deposition of mineral on the surface of extracted teeth when exposed to remineralizing solutions. The ultimate goal of investigations on enamel remineralization is a possible clinical application. It is then necessary to recognize that remineralization of enamel in the laboratory has been invariably accomplished under optimal conditions, i.e., calcium and phosphorus concentrations as high as the chemistry of calcium phosphates permits, high pH values, and almost total disregard for the factors that may favor the reverse process. There is no information, at present, on those factors determining the reversibility of an incipient carious lesion; in this context, it is pertinent to recall the demineralization stages described before in this paper. The marked increase in the pore volume accessible to water vapor that occurs during the demineralization process has been ascribed5 to an alteration of the organic matrix that, essentially, gives full accessibility to intraprismatic pores and marks the loss of order in the enamel structure. After this event, it may be extremely difficult to reconstruct the tissue with all its original properties. In other words, the alteration of the organic matrix may be the limiting factor for the reversibility of an incipient carious lesion. This possibility concerning reversibility of a lesion is based only on conjectural thought, but it deserves consideration and experimental effort. Similarly, the possibility of enhancing the natural remineralizing mechanism should be investigated. In fact, the effectiveness of fluoride mouth rinses in caries prevention may derive from such an enhancement. In this case, fluoride may have a multiple effect, e.g., decreasing the acid production of cariogenic microorganisms8 (so that the mineralizing potential of saliva becomes effective), increasing the natural rate of remineralization, and enhancing the stability of the enamel mineral. The clinical literature on the possible benefits of remineralizing solutions is very scanty. A recent report20 indicates that the
reversal rate of incipient lesions (induced by rinses with sucrose solutions without oral hygiene practices), was increased by the frequent use of a remineralizing solution containing calcium, phosphate and fluoride. The study20 was conducted with a small number (15) of subjects and the differences between the experimental and control groups were notvery great. Perhaps real advancements in this field are hampered by the limitations imposed by the chemistry of calcium phosphates (metastability of the solutions proposed) and the difficulty in separating the effects of remineralizing agents from those associated with the oral fluids. The lack of a marked success at the clinical level may imply that the simple approaches using solutions containing calcium, phosphorus, and fluoride are not adequate. A closer look at the natural remineralization process may yield useful information concerning factors present in the oral environment which are involved in remineralization. It is conceivable that organic or inorganic additives may accelerate the rate of mineral deposition at the needed location under oral conditions. REFERENCES 1. HOLLY, F. J., GRAY, J. A.: Arch. Oral Biol. 13:319-334 (1968). 2. MORENO, E. C., ZAHRADNIK, R. T.: J. Dent. Res. 53:226-235 (1974). 3. SILVERSTONE, L. M.: Caries Res. (Suppl 1), 11:59-84 (1977). 4. DARLING, A. I.: In: Mechanisms of Hard Tissue Destruction (SOGNNAES, R. F., ed) pp. 171-184. American Association for the Advancement of Science, Washington, 1963. 5. ZAHRADNIK, R. T., MORENO, E. C.: Arch. OralBio. 22:585-591 (1977). 6. ZAHRADNIK, R. T., PROPAS, D., MORENO, E. C.: J. Dent. Res. 56:1107-1110 (1977). 7. ZAHRADNIK, R. T., PROPAS, D., MORENO, E. C.: J. Dent. Res. In press. 8. JENKINS, G. N.: J. Dent. Res. 42:444452 (1963). 9. ZAHRADNIK, R. T., MORENO, E. C., BURKE, E. J.: J. Dent. Res. 55:664-670 (1976). 10. TURNER, E. P.: Dent. Practitioner 8:373382 (1958). 11. CLARK, W. B., BAMMANN, L. L., GIBBONS, R. J.: Infect. Immun. In press. 12. BACKER-DIRKS, O.: J. Dent. Res. 45:
503-511, (1966).
13. VON DER FEHR, F. R., LOE, H., THEILADE, E.: Caries Res. 4:131-148 (1970). 14. KOULOURIDES, T.: Ann. N. Y. Acad. Sci.
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
902
MORENO & ZAHRADNIK
153:84-101 (1968). 15. SILVERSTONE, L. M., POOLE, D. F. G.: Caries Res. 2: 87-96 (1968). 16. FEAGIN, R., PATEL, P. R., KOULOURIDES, T., PIGMAN, W.: Arch. Oral Bio. 16:535548 (1971). 17. MEYER, J. L., NANCOLLAS, G. H.: J. Dent.
J Dent Res Special Issue B, 1 9 79 Res. 51:1443-1450 (1972). 18. MORENO, E. C., ZAHRADNIK, R. T., GLAZMAN, A., HWU, R.: Calcif. Res. 24:47-57
(1977). 19. SCHLESINGER, D. H., HAY, D. I.: J. Biol. Chem. 252:1689-1695 (1977). 20. LEVINE, R. S.: Brit. Dent. J. 138:249-253
(1975).
Session VI Discussion Dr. Driessens: I was impressed, especially by your measurements of the radius of the pores. In our laboratory we determined in the past few years the diffusion coefficients for 45Ca2+, 86Rb+, 36C1-, 3H-sorbitol and 14C-glycerol. We found different results for the different particles, depending on both their size and charge. This means that there are certain interactions of the enamel membrane with the particles diffusing through the pores. The first effect is that of ion selectivity. We also could detect a certain molecular sieve effect, especially for sorbitol and glycerol as compared to the average for the ions. If we account for the differences in radius, and if we average over the ion selectivity effect, we end up with the same range of values for the diameters of the pores as you have found with the BET method. We have also carried out a computer simulation of the caries process. We first worked with Nernst-Planck flux equations for each of the particles, without making additional assumptions, and then tried to obtain a subsurface lesion. We did not succeed. We had to include an extra condition, an extra term giving consideration to a gradient in the rate of dissolution of the mineral. Otherwise, we did not get subsurface demineralization but etching. We found in our simulation program that changes in the ion selectivity as small as found in your experiments with pellicle adsorption did not influence the rate of demineralization more than a few percent, still resulting in etching. This might be important for the interpretation of your results, as you found subsurface demineralization after adsorption of pellicle. The only way to explain this result, it seems to me, is that the adsorption of pellicle changes the rate of dissolution of the mineral in the outer layer very considerably.
Dr. Moreno: That is essentially what we have shown. In other words, in the presence of a pellicle, you control the transport of matter from the tooth surface into the demineralizing medium. Dr. Driessens: Well, you gave me another impression, because in your presentation you said that the pellicle had a marked effect on the rate of diffusion which is different from the rate of dissolution of the mineral. Dr. Moreno: Well, that is correct. If the enamel mineral dissolves, its constituents in solution will have to move, through the interface, into the demineralizing medium; we maintain that the pellicle, because of its ionic permselective properties, slows down this process. Dr. Driessens: We found that you must take into account both the rate of diffusion and the rate of dissolution in order to simulate the caries process. Dr. Moreno: You mentioned something that is interesting, that in your simulation program, you couldn't account for your results by differences in the permselective properties of the enamel interface. I would suggest that you review your program because, experimentally, we can produce subsurface demineralization by the modification of these properties, i.e., by forming a salivary pellicle at the enamel surface. Dr. Driessens: We did simulate this formation of the intact layer, but then we had to assume either a gradient in the pore density, or a gradient in the solubility product, or a gradient in the dissolution rate constant. Dr. Silverstone: In one of your early tables, you showed the depth of lesion in respect to different aged pellicles, and you
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
903
Vol. 58(B) noted there was hardly any difference in depth. Dr. Moreno: That is correct. Dr. Silverstone: And you were somewhat disappointed. Can I suggest that this is because you are limiting yourself to what I consider to be a relatively crude technique, microradiography? When you're measuring the depth of the lesion, what in fact you're measuring is the depth of zone number three, the body of the lesion. So, in other words, you're not looking at the translucent zone or the dark zone, both of which together could be a hundred microns. This, in fact, could make quite a difference. Dr. Moreno: That is right. I could not detect those through this technique. Dr. Termine: Could you elucidate further on your studies of pellicle effects in remineralization, particularly in light of the very dramatic effects you obtained in the demineralization studies? Dr. Moreno: We can illustrate this point with the results obtained in kinetic studies of crystal growth using seeded solutions supersaturated with respect to hydroxyapatite; these results are shown in a graph included in the paper. It is clear that a concentration of statherin of 0.4 um brings about a substantial reduction in the precipitation rate of the seeded solution. The mean concentration of statherin in parotid saliva is about 4 um. Thus, concentrations about one-tenth of that in the glandular secretion reduce drastically the rate of crystal growth. This is a phenomenon that we have not elucidated completely but we feel that more consideration should be given to it in relation to enamel remineralization. Dr. Ten Bosch: You have been showing those results on a seven-day old pellicle. I think that is a very long time for a pellicle to get formed; do you have any idea how thick it is, and what material it is? Dr. Moreno: We frankly do not know, but we think that some structural changes in the pellicle occur during the seven-day period of development. We have conducted obvious experiments; for example, exposure of enamel to fresh saliva every hour for
seven consecutive times fails to develop a pellicle with the permselective properties obtained when the enamel is exposed for seven days using fresh saliva every 24 hours. Similar failure is observed when the enamel is exposed to saliva for 24 hours and then is kept for seven days in a moist atmosphere. So, some undetermined changes occur during the seven-day development period and perhaps people in the salivary field could shed some light on this. Dr. Weill: Speaking of remineralization, do you really mean that a new apatite is formed? If not, it is deposition of mineral and not remineralization. Further, I think that there are distinct differences between in vivo and in vitro experiments. If you look at human carious enamel you find that the core of the prism is invaded by stainable material that is quite different from the normal enamel matrix. Dr. Moreno: We don't know if the remineralization proceeds to the core of the prism or if it does, if it proceeds to form crystals with the same order that the crystallites had prior to demineralization. This is something that should be investigated. Dr, Nancollas: It would be very interesting to know whether the statherin does in fact affect the mechanism of mineralization, or whether it simply affects the rate. We found that there are very few active sites on crystals which are actually growing. Dr. Moreno: That is correct. Dr. Nancollas: We have done some rotating disc dissolution work involving laminar flow of both whole and white spot enamel. The decrease in rate in the latter experiments can be interpreted in terms of different diffusion rates through the surface phases. Dr. Moreno: The only answer I can give is that we have conducted this kind of experiment at different concentrations of statherin. Our results, so far, match very well with the adsorption studies that we have done with statherin on hydroxyapatite, so that a decrease in the initial rate of precipitation is proportional to the amount of statherin adsorbed on the apatite seeds.
Downloaded from jdr.sagepub.com at UNIVERSITE LAVAL on July 1, 2014 For personal use only. No other uses without permission.
