Physico-chemical Characterisation of the Processes Involved in Enamel Remineralisation by CPP-ACP

Casein phosphopeptides derived from tryptic digests of milk caseins spontaneously assemble with calcium and phosphate ions at high pH to form casein phosphopeptide-amorphous calcium phosphate complexes (CPP-ACP). These complexes have been shown to be able to repair lesions in tooth enamel (biohydroxyapatite – HA) both in vitro and in vivo (specifically white spot lesions in the early stages of tooth decay). In order to better understand the processes involved in enamel remineralisation, the chemical equilibria between the CPP and calcium and phosphate ions as a function of pH were investigated. Furthermore, a thin-enamel slab technique was developed with enhanced sensitivity to monitor the diffusion of radio-opaque ions into individual lesions over a period of days to weeks.

essentially a chemical process, at early stages of dental caries, the hydroxyapatite mineral loss is reversible. Any products that prevent enamel demineralisation and promote remineralisation are described as exhibiting anticaries activity. The principal components of dairy products associated with their anticariogenic activity are multi-phosphorylated caseins complexed with calcium and phosphate. Enzymic hydrolysis of caseins yields phosphopeptides known as casein phosphopeptides (CPP). These peptides, with their multiple phosphoseryl residues, bind relatively large quantities of calcium and phosphate ions in an amorphous, bioavailable form (Reynolds et al. 1995). The resulting complexes are known as casein phosphopeptideamorphous calcium phosphate (CPP-ACP). The two dominant, self-assembling peptides are β-CN(1-25) and α S1 -CN(59-79) forming 20-30% by mass of the total CPP ( Fig. 23.1). These bovine casein-derived peptides all contain the cluster sequence motif -(Ser(P)-) 3 (Glu-) 2 .
The aim of this study was to investigate the interactions between the peptides and crystalline and non-crystalline mineral components during the remineralisation process.

Ion-Binding Studies
To evaluate the influence of pH on the calcium and phosphate ion equilibria, solutions of CPP-ACP, α S1 -CN(59-79)-ACP, and β-CN(1-25) were subjected to pH titrations. Calcium and phosphate ion concentrations were determined using microfiltration using previously described protocols, (Cross et al. 2005) modified to use a Dionex ion analyser.

Nuclear Magnetic Resonance Studies
NMR spectra were acquired at 599.741 MHz on a Varian Unity Inova spectrometer as described previously (Cross et al. 2016). Solution-phase diffusion measurements were performed using the sLED experiment as described previously (Altieri et al. 1995).The amplitude of the NMR signal is a function of the applied magnetic field gradient and the Stokes-Einstein radius of the diffusing species (S ∝ exp (−αDG 2 ),where D is the diffusion coefficient, G is the magnetic field gradient, and α depends on experimental values. The diffusion coefficient is related to the hydrodynamic radius by D = k B T/6πρR where k B is the Boltzmann constant, T the absolute temperature, ρ the solution viscosity, and R the hydrodynamic radius of the spherical particles.

Remineralisation Studies
The novel technique utilised thin slabs of enamel (~300 μ thick) cut from human third molars that allowed sound mineral portions of the slabs to be used as controls in measuring the time course of remineralisation of artificial lesions. The technique is an extension of that previously described (Cochrane et al. 2008), with remineralisation of individual slabs being assessed after 0, 2, 3, 6, 12, 15, and 20 days immersion in a remineralisation solution at either pH 5·5 or pH 7·0. Diffusion of radio-opaque ions into the artificially prepared lesions was monitored by TWIM (Thomas et al. 2006). Acid-resistant nail polish was used to define the remineralisation zone and applied three times to prevent leakage during the soaking in the remineralisation solutions. Remineralising solutions consisted of 1% solutions of either CPP-ACP or β-CN(1-25)-ACP prepared at either pH 7.0 or 5.5 to compare the effects of neutral and acidic pH. Lesion-sections were subjected to transversal wavelength-independent microradiography (TWIM) at days 0, 2, 3, 6, 12, 15, and 20 to visualise the time dependence of mineral ion uptake during enamel remineralisation. Microradiographs acquired on day 0 were used as control images for each lesion-section. The lesion-sections remained soaked in the remineralisation solutions except when being X-rayed.
DOSY experiments were performed using the sLED technique to determine the relative rates of diffusion of the complexes using either the integrated aromatic or aliphatic signals of a β-CN(1-25)-ACP sample.
SDS-PAGE of CPP cross-linked using glutaraldehyde was conducted to analyse the multimerisation of CPP as described previously (Cross et al. 2016).

The CPP-ACP Complexes Exist in Equilibria with Both Bound and Free Calcium and Phosphate Ions
When prepared at pH 9, the CPP-ACP complexes have most of the calcium and phosphate peptide bound. However, these complexes exist in solution in equilibrium with free ionic calcium and phosphate. Figure 23.2 shows sigmoidal pH titration curves for β-CN(1-25)-ACP representative of results obtained in this study. The pK a values for Ca 2+ and Pi binding are 5.983 ± 0.038 and 6.302 ± 0.067, respectively. Further pH titrations were performed on 1% and 2% β-CN(1-25)-ACP complexes prepared with varying Ca/Pi ratio and varying peptide:Ca ratio. Plots of the bound calcium and phosphate concentrations against pH revealed sigmoidal curves whose shape remained independent of the Ca/Pi ratio. For the complexes prepared at ratios of Ca/Pi ranging from 1.6 to 1.51, and 12-15 Ca/peptide, the pH dependence The solid-line curves are fit to the Henderson-Hasselbalch formula yielding effective pK a values of 6.302 ± 0.067 for the phosphate ion curve and 5.983 ± 0.038 for the calcium ion curve. Note that at low pH, the peptides bind residual calcium ions but no phosphate ions of the bound calcium and phosphate were similar. Further analysis using α S1 -CN(59-79)-ACP revealed a well-defined phase of ACP stabilised by the CPP (Fig. 23.3ac). Figure 23.3d shows a representative powder X-ray diffraction pattern of CPP-ACP that is consistent with an amorphous phase of calcium phosphate.

The CPP-ACP Complexes Are Small Readily Diffusible Species
The sLED experiment yields a diffusion-dependent signal whose functional dependence on the applied, magnetic-field gradient is dependent on the hydrodynamic radius of the molecule being studied. Figure 23.4a shows a plot of sLED signal intensity against the applied magnetic field gradient. This provides the ratio of hydrodynamic radii of the β-CN(1-25)-ACP complex relative to that of water. The A plot of CPPbound inorganic phosphate against ACP ion activity product. This illustrates that only an amorphous calcium phosphate phase predicts a one-to-one functional dependence of calcium phosphate ion activity product and the activity of phosphate (shown here) or calcium. (c) A plot illustrating that the best fit between the ion activity product and the phosphate (or calcium) activity occurs with a slightly calcium-rich, non-stoichiometric ACP phase having the formula Ca 3.0425 (PO 4 ) 2 (OH) 0.085 . The lower curve is for calcium-deficient ACP phases and does not have a maximum at realistically achievable compositions. (d) X-ray powder diffraction image of a CPP-ACP sample demonstrating the broad peaks expected from an amorphous solid measured hydrodynamic radii of the β-CN(1-25)-ACP complex range from 1.53 ± 0.03 nm at pH 6 to 1.92 ± 0.08 nm at pH 9. The deviation from the fitted curves at high-magnetic-field-gradient values fits a two-component model that is consistent with the formation of aggregates.
SDS-PAGE of CPP cross-linked using glutaraldehyde, in the presence of either calcium ions (Fig. 23.4b) or calcium and phosphate ions (Cross et al. 2016), suggests that the complexes contain up to six CPP peptides. Figure 23.4c shows a model of the CPP-ACP complex consistent with the results of these experiments.  (Altieri et al. 1995) that uses magnetic field gradients to measure the relative rates of diffusion of various protonated species. The water signal (blue) can be compared with either the integrated aromatic (green) or aliphatic (red) signals of a CPP-ACP sample. The data from both peptide curves provided an internal consistency check. The Stokes-Einstein equation relates the hydrodynamic radius of a particle and its diffusion constant, allowing the determination of the hydrodynamic radius of the complexes given the known radius of water. The calculated hydrodynamic radii of the CPP-ACP complexes ranged from 1.53 ± 0.03 nm at pH 6 to 1.92 ± 0.08 nm at pH 9. The deviation from the fitted curves at high-magnetic-field-gradient values fits a two-component model consistent with the formation of aggregates. (b) A representative SDS-PAGE gel featuring the cross linking of the CPP using glutaraldehyde, observed in the presence of calcium ions. The multimerisation was also observed in the presence of calcium and phosphate ions suggesting that the complexes contain up to six CPP peptides. (c) A model of the CPP-ACP complex consistent with these and various other experiments. The peptide is depicted as sticks within a translucent van der Waals surface, and the ACP is shown as a pale grey van der Waals surface

Both CPP-ACP and β-CN(1-25)-ACP Complexes Release Mineral Ions that Remineralise Demineralised Enamel Lesions
Blocks of enamel from third molars with demineralised lesions were prepared that yielded multiple uniformly demineralised lesion-sections that would enable intertooth and intra-tooth comparisons (Fig. 23.5a). The sound enamel regions provided the experimental control regions for the experiments. Acid-resistant nail polish was used to define the remineralisation zone. Microradiography was used to visualise the time dependence of mineral ion uptake, in enamel remineralisation experiments. Figure 23.5b shows a sample X-ray micrograph with the demineralised zone (arrow) and the adjacent sound enamel (star). .5c shows a plot of the time-dependent, X-ray opacity of a specific enamel slab after demineralisation at day 0 and following remineralisation until day 20. The X-ray data showed a diffusion-dependent increase in electron density, interpreted as mineralisation.
A time-dependent uptake of mineral was observed in the presence of both CPP-ACP and β-CN(1-25)-ACP at both pH values. The calculated data from each sample was fitted to the time-dependent part of a diffusion equation of the form where 'k' represents the rate constant for diffusion and 't' represents the elapsed time taken in days. The Pearson correlation coefficients for the non-linear curve fit confirmed that the data were consistent with a simple diffusion model for remineralisation. Figure 23.6 illustrates time-dependent plots of the calculated extent of remineralisation (R f ) for four representative sections. The data is consistent with mineral ions from CPP-ACP and β-CN(1-25)-ACP diffusing into and interacting with the enamel crystals. A recent study of the interaction of CPP-ACP with either enamel or saliva-coated enamel (Huq et al. 2018) indicates that chemical A m is the area under the 'remineralised' enamel curve, A d is the area under the 'demineralised' enamel curve (day 0), and A s is the area under the sound enamel curve. Statistically significant differences in the extent of remineralisation are observed between different samples and may be due to biological differences: the lowest R f of 0.17 ± 0.04 (in the graph above) differs significantly from the maximum R f of 0.28 ± 0.04 at a p = 0.025 equilibrium is established within a few hours; thus, the rate-limiting step for enamel remineralisation is the rate of diffusion of ions into the subsurface enamel.

Discussion
In this study, a variety of methods have been used to characterise the complexes formed by the casein phosphopeptides with calcium and phosphate ions. CPP bind calcium and phosphate to form stable CPP-ACP complexes in alkaline solution. These studies show that the ratio of bound calcium to bound phosphate is constant and independent of pH in the range of pH 7-9. Furthermore, the ion activity product fits a single curve for a calcium-rich, non-stoichiometric calcium phosphate phase. DOSY experiments using the sLED sequence demonstrated that the complex has a small but significant variation in size with pH. These experiments further revealed the ability of the small complexes to aggregate as concentrations were increased. The model of the CPP-ACP complex with all amino acids in the peptides interacting with the ACP surface is consistent with earlier findings that the peptide length influences the extent of binding to calcium and phosphate ions (Cross et al. 2005).
The importance of small readily diffusible CPP-ACP complexes is confirmed by the experiments using thin slabs of human enamel with artificial lesions that mimic early carious lesions. To enable the time-dependent monitoring, the current remineralisation procedures (Shen et al. 2011) required extensive improvements. The use of the single section for all-time points required a thicker section to withstand repeated handling. To accommodate the increased thickness, TWIM was used instead of the commonly used transverse microradiography (TMR) technique, both methods being validated techniques for monitoring carious lesions (Thomas et al. 2006). In addition, the acid-resistant nail varnish was reapplied three times to prevent leakage during the soaking in remineralisation solutions. To improve the signal-to-noise ratio, an additional 500-μ-thick aluminium filter was used with an optimal X-ray tube voltage of 30 kV.
We observed statistically significant differences in the extent of remineralisation within the multiple lesion-sections derived from the same individual tooth. These differences were attributed to varying microporosities of the demineralised lesionsections of each individual tooth.
In conclusion, the CPP-ACP complexes were characterised to be a small readily diffusible species that can release the mineral ions on contact with demineralised enamel lesions. Furthermore both CPP-ACP and β-CN(1-25) complexes were able to remineralise within a few days in in vitro experiments.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.