Chitosan microparticles for the controlled delivery of fluoride

journal of dentistry 40 (2012) 229–240

Available online at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/jden

Chitosan microparticles for the controlled delivery of fluoride Gemma M. Keegan a, John D. Smart b, Matthew J. Ingram b, Lara-Marie Barnes b,*, Gary R. Burnett c, Gareth D. Rees b,d a

Chiesi Ltd., Chippenham, Wiltshire SN14 0AB, UK School of Pharmacy and Biomolecular Sciences, University of Brighton, Brighton BN2 4GJ, UK c GlaxoSmithKline Consumer Healthcare, Weybridge, Surrey KT13 0DE, UK d Genesis Oral Bioscience, Horley, Surrey RH6 7AN, UK b

article info

abstract

Article history:

Objectives: To manufacture and characterise chitosan/fluoride microparticles prepared by

Received 18 July 2011

spray drying and assess their utility as controlled release vehicles for fluoride.

Received in revised form

Methods: Microparticles were manufactured from dispersions containing 1.0% and 2.0% (w/

14 December 2011

v) chitosan and 0.20% or 0.40% (w/v) NaF in the absence/presence of glutaraldehyde. Particle

Accepted 15 December 2011

size distributions were determined using laser diffraction; fluoride loading and release were determined by ion-selective electrode. Release profiles were studied in isotonic media (pH 5.5) over 360 min; microparticles exhibiting greatest cumulative fluoride release were

Keywords:

further evaluated at pH 4.0 and 7.0. Particle morphology was investigated using environ-

Chitosan

mental scanning electron microscopy. Bioadhesion parameters were determined with a

Microparticles

texture-probe analyser.

Bioadhesion

Results: Microparticles exhibited low polydispersity and volume mean diameters (VMDs)

Fluoride

<6 mm. VMDs increased on doubling the chitosan/fluoride concentrations but were largely

Controlled release

independent of glutaraldehyde concentration. Recovered yields were inversely proportional to dispersion viscosity due to compromised fluid atomisation; adding NaF reduced viscosity and improved yields. Best-case entrapment efficiency and NaF loading were 84.1% and 14%, respectively. Release profiles were biphasic, releasing 40–60% of the total fluoride during the first 600 s, followed by a prolonged release phase extending out to 6 h. Incorporation of 0.40% NaF to the 2.0% chitosan dispersion yielded microparticles with reduced bioadhesive parameters (Fmax and WOA) versus the chitosan-only control whilst retaining significant bioadhesive potential. Conclusions: Bioadhesive chitosan/fluoride microparticles manufactured using a spray-drying protocol have been extensively characterised and further opportunity for optimisation identified. These microparticles may provide a means of increasing fluoride uptake from oral care products to provide increased protection against caries, however further work is required to demonstrate this principle in vivo. Clinical significance: Spray-drying is a low-cost route for the manufacture of bioadhesive chitosan/fluoride microparticles which can be exploited as controlled fluoride release agents to aid fluoride retention in the oral cavity. The potential exists to optimise release profiles to suit the delivery format thereby maximising the cariostatic benefits. # 2012 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +44 1273 643918; fax: +44 1273 642674. E-mail addresses: [email protected], [email protected] (L.-M. Barnes). 0300-5712/$ – see front matter # 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jdent.2011.12.012

230

1.

journal of dentistry 40 (2012) 229–240

Introduction

Dental caries is one of the most prevalent chronic diseases affecting western populations.1 In the presence of a plaque biofilm, demineralisation of the dental hard tissues occur when microbially derived organic acids, such as lactic acid, react with calcium hydroxyapatite (Ca10(PO4)6(OH)2), the principal mineral component of teeth. This results in loss of calcium and phosphate ions which subsequently diffuse out of the tooth. Demineralisation by plaque acids is stimulated by ingestion of fermentable carbohydrate, most notably sucrose, and is accompanied by a clear measurable fall in plaque pH.2 In healthy individuals the acids produced are cleared relatively rapidly through the action of saliva.3 Saliva production is stimulated by eating and drinking, and as saliva is supersaturated with respect to calcium and phosphate, remineralisation is thermodynamically favoured as the pH returns to normal.2,3 This continuous cycle of demineralisation followed by remineralisation is a natural process and provided there is no net mineral loss over an extended period, the dentition should remain healthy. However, prolonged periods of low pH within the saliva and dental plaque not only shift the equilibrium in favour of demineralisation, but also result in a shift in microbial balance in favour of acidogenic and acidophilic bacteria such as Streptococcus mutans, which possess several virulence factors important in the development of caries including the production of bioadhesive extracellular polysaccharides.4 Frequent consumption of a diet high in fermentable sugars is a significant risk factor in the development of dental caries, together with poor oral hygiene and numerous physical and biological factors such as salivary composition and flow.1,4 The addition of fluoride to public water supplies and oral healthcare products is known to exert an anti-caries effect and has resulted in significant improvements in the incidence of dental caries.5,6 Fluoride reduces the solubility of enamel through the substitution of hydroxyl ions to form fluorapatite (Ca10(PO4)6F2) or the partially substituted fluorhydroxyapatite (Ca10(PO4)6(OH)2 xFx); both have a lower solubility than hydroxyapatite at a given pH. Furthermore, fluoride may modulate the growth of oral bacteria through several different mechanisms that are dependent on low environmental pH. These include the sensitisation of biofilms to acid damage, reduction of acid production through the inhibition of enzymes involved in glycolysis, reduction of IgA1 protease synthesis, the disruption of bacterial adherence and the reduction in extracellular polysaccharide production.7–12 The incorporation of fluoride into dental materials has been shown to inhibit the growth of relevant bacteria.12,13 Moreover, such materials can be ‘‘recharged’’ with fluoride,14 and the continuous presence of even low concentrations of fluoride in the fluid phase surrounding the teeth is considered to confer significant cariostatic benefits.15,16 The use of oral care products containing fluoridated actives such as sodium fluoride (NaF), amine fluoride (AmF2) and stannous fluoride (SnF2) result in elevated fluoride concentrations both in saliva and plaque fluid for up to several hours. Sodium monofluorophosphate (NaMFP), a fluorinated anti-caries active, is also

widely used, however phosphatases present in saliva and dental plaque are required to release ionic fluoride. The baseline levels of fluoride in saliva of subjects using fluoridated dentifrices is typically double that of subjects using fluoride-free dentifrices.17 Sustained release fluoride devices, such as buccoadhesive tablets,18,19 and dental restoratives, including glass ionomer cements,20 help maintain low but significantly elevated fluoride concentrations in saliva and dental plaque.15,19,20 However, mucoadhesive tablets are often poorly tolerated by consumers due to the relatively short duration of adhesion to the oral mucosa, their size, and occasionally associated irritancy effects,21 whilst dental restoratives require professional intervention. As a consequence, the number of users of the aforementioned sustained fluoride delivery formats is negligible in comparison to those employing over the counter oral care products such as dentifrices and mouth rinses. The use of single high dose applications of fluoride, such as those found in a number of prescription oral gels, provides an alternative approach. The low pH of some of these formulations, coupled with their high fluoride concentrations, typically 12,500 mg g 1 (ppm), results in the formation of surface deposits of CaF2-like material. These CaF2-like deposits are more acid labile and thermodynamically less stable than fluorhydroxyapatite, and have been shown to be rapidly dissolved by saliva.22 As a result, the fluoride is ingested and repeated application of the high fluoride gel is required if this exogenous mineral layer is to be reestablished,22 precluding their use in anything other than the short term. Improving fluoride delivery from dentifrice and mouth rinse formats that are applied frequently and over the long term consequently offer the greatest opportunity to drive globally reaching improvements in caries protection at a cost that is within the reach of most populations in the developed and developing world, and that does not involve the contentious issue of fluoridating the water supply itself. It is notable that there appears to be a role for high fluoride dentifrices and gels in the control of dental erosion where the severity of the challenge is commonly 1–2 orders of magnitude greater than that of a caries challenge, and the anti-erosion benefit appears to be dose dependent with respect to fluoride. A significant number of in vitro and in situ studies have reported such an effect.23–25 Chitosan is a copolymer of glucosamine and N-acetyl glucosamine produced from the partial deacetylation of chitin.26 Chitosan exhibits moderate to good mucoadhesion, and in vivo has been shown to be retained on the oral mucosa for several hours, even when delivered in its fully hydrated form as a simple aqueous rinse.27 In addition, chitosan exhibits antibacterial activity against a range of oral pathogens through the disruption of the cell membrane,28,29 and reduces plaque formation following a chitosan-rinsing regime.30,31 A recent in vitro study also suggests it may have utility as an inhibitor of enamel demineralisation where its protective effect was attributed to binding and formation of a barrier layer.32 Chitosan is widely used as a dietary supplement and has a favourable toxicological profile.33 Chitosan microparticles have previously been investigated for local delivery to the oral cavity of therapeutics including tetracycline,34 chlorhexidine,35 and triclosan.36 These studies

231

journal of dentistry 40 (2012) 229–240

have employed manufacturing techniques including ionotropic gelation, and various emulsion polymerisations protocols where the active of interest is hydrophobic in order to form drug-loaded chitosan microparticles. However, such processes are expensive to scale up and, in the case of drug actives such as ionic fluorides designed for relatively low-cost fastmoving consumer products such as dentifrices, prohibitively costly. The aim of this work was to manufacture and characterise chitosan microparticles containing NaF using a spray drying route previously reported for the matrix microencapsulation of metoclopramide, a highly water soluble anti-emetic.37 In the present study, particle size distribution, fluoride loading, in vitro fluoride release, particle morphology, and in vitro bioadhesion of discs prepared from compressed chitosan/ fluoride microparticles to a model mucosal surface were assessed. The effect of glutaraldehyde and the use of different chitosan concentrations on particle size, fluoride loading and release were also determined. Chitosan microparticles formed using glutaraldehyde as a crosslinker are reportedly nontoxic.38 The null hypotheses tested in this study, using in vitro models, were (a) that the concentrations of glutaraldehyde and chitosan used in the manufacture of microparticles would not influence particle size, fluoride loading or fluoride release and (b) that the use of glutaraldehyde as a cross-linker and the incorporation of fluoride would not influence bioadhesive properties.

2.

Materials and methods

2.1.

Materials

Chitosan (HCMF; Lot GR23238901) with a molecular weight of 50–1000 kDa was used in this study (kindly donated by Cognis Deutschland GmbH Company, Illertissen, Germany). Sodium acetate trihydrate, ethylenediaminetetraacetic acid (EDTA), tetrasodium salt dihydrate, sodium dihydrogen phosphate monohydrate and NaF were all of ACS reagent grade (Sigma– Aldrich Company Ltd., Gillingham, UK). All other chemicals used in this investigation were purchased from Fisher Scientific UK Ltd. (Loughborough, UK). All chemicals were used as received without further treatment unless otherwise stated.

2.2.

Microparticle manufacture

The composition of the various aqueous dispersions used to produce the chitosan microparticles is described in Table 1. A single batch was prepared in each case, sufficient to provide material for their physical characterisation and subsequent release studies. Chitosan was first solubilised in 1.0% (v/v) acetic acid (40 mL), sealed, and stirred overnight using a magnetic stirrer at 300 rpm (Stuart, Hotplate Stirrer CB192: Fisher Scientific). Solutions of NaF were then prepared in deionised (DI) water (10 mL) and added drop wise over a period of ca. 60 s to the aqueous chitosan dispersion to give a final volume of 50 mL. Microparticles were prepared by spray drying the dispersion using a bench-top mini spray drier (Buchi B290; Buchi UK Ltd., Oldham, UK) fitted with a 0.7 mm

Table 1 – Compositional details of the dispersions used to prepare the spray dried chitosan-based microparticles used in these studies; final dispersion volume = 50 mL. Further detail regarding spray drying conditions is provided in Section 2.2. Composition/ sample ID 1 2 3 4 5 6 Control

Chitosan (g)

Glutaraldehyde (g)

Sodium fluoride (g)

0.50 0.50 0.50 1.0 1.0 1.0 1.0

0 0.0005 0.025 0 0.001 0.05 0

0.10 0.10 0.10 0.20 0.20 0.20 0

nozzle, in accordance with the previously reported method.35 Each dispersion was fed to the nozzle using a peristaltic pump at a flow rate of 8 mL min 1. Drying conditions employed a spray flow rate of 35 m3 h 1, compressed air flow rate of 600 Nl h 1, inlet air temperature of 130 8C and outlet temperature of 65 8C. The dry powder product was then collected in a receiver bottle and stored in a desiccator until use.

2.3.

Microparticle characterisation

2.3.1.

Particle size distribution

Spray dried microparticles (10 mg) were accurately weighed into a 30 mL glass bottle using a precision balance (MettlerToledo Ltd., Leicester, UK). Samples were then suspended in carrier fluid (20 mL) containing 1.0% (v/v) Tween 20 in methanol, and sonicated for 2.5 min to promote formation of a homogeneous dispersion. Particle size distributions were determined using a particle sizer (Series 2600; Malvern Instruments Ltd., Malvern, UK) with an MS1 small volume cell system and MS23 measurement cell operating on Malvern B.0D software. Background scattering of the laser was determined prior to measurement of particle size with 80 mL of carrier fluid. No significant swelling was observed on suspension of the microparticles in the methanolic carrier over the duration of the measurement period.

2.3.2.

Environmental scanning electron microscopy (ESEM)

The morphology of the microparticles within and between batches was evaluated using ESEM. Dry microparticles were applied to an adhesive-coated SEM stub and the excess removed by gentle tapping. Specimens were then viewed using a Zeiss EVO 50EP operating at 10–12.5 keV with Windows XP based SmartSEM control software.

2.3.3.

Fluoride analysis

Fluoride ion activity was determined using a fluoride ion selective electrode (ISE) (Radiometer Analytical; Lyon, France) and silver/silver chloride reference electrode (Fisher Scientific). All samples and calibrants were measured in an ionic strength adjustment buffer (ISAB) consisting of 2.0% (w/v) EDTA, 2.0% (w/v) sodium acetate and 0.30% (w/v) sodium hydroxide at pH 7.25. A linear, five point calibration curve of electrode potential against Log10 of the fluoride-ion concentration was constructed between 1.0 and 50.0 mg mL 1 of fluoride as sodium fluoride, with the electrode conditioned as

232

journal of dentistry 40 (2012) 229–240

per manufacturer’s instructions prior to use. All measurements were performed in stirred solutions with the temperature maintained at 1 8C.

2.3.4.

Determination of total fluoride loading

Chitosan–fluoride and chitosan-only microparticles (10 mg) were hydrated in 8 mL of DI water in a 50 mL polycarbonate centrifuge tube (Fisher Scientific) for 30 min at room temperature. Following the addition of concentrated HCl (2 mL), samples were placed on an orbital shaker (Denley MaxiMix; Thermo Fisher Scientific, Basingstoke, UK) at 200 rpm for 60 min at 32 8C. Subsequently, ISAB buffer (40 mL) was added and the mixture agitated for a further 30 min, after which each sample was centrifuged at 1300  g for 15 min to collect insoluble chitosan as a pellet. The supernatant was decanted and analysed for fluoride as described in the preceding section.

2.3.5.

In vitro fluoride release

An isotonic solution was prepared from 0.0229% (w/v) calcium chloride, 0.8913% (w/v) sodium chloride and DI water then adjusted to pH 5.5 with 0.10 mol dm 3 HCl to a final volume of 1.0 dm 3. Chitosan–fluoride microparticles (40 mg) were accurately weighed and suspended in 10 mL of the isotonic solution. The suspension was then placed on an orbital shaker at 200 rpm and incubated at 32 8C for 360 min. The amount of fluoride released from the microparticles was determined after 10, 30, 60, 120, 240 and 360 min. Each tube was centrifuged at 6000  g for 10 min (Sorvall RC-5B; Thermo Fisher Scientific) and test samples (7.5 mL) removed using a pipette and replaced with fresh isotonic solution. Concentrated HCl (2.0 mL) and DI water (0.50 mL) were added to each 7.5 mL sample and the solution stirred for 60 min. The acidtreated samples were then diluted to 50 mL with ISAB buffer and the fluoride concentration determined as previously described. Mean fluoride release at each time point was determined from triplicate analyses; the mean cumulative fluoride released at each sampling point was calculated and expressed in terms of percentage of the total fluoride present based on the known fluoride loadings. The batch of microparticles that released most fluoride over the 360 min study period was subject to additional release studies in isotonic media adjusted to pH 4.0 and 7.0, respectively.

2.3.6.

Texture probe analysis

Microparticles (50 mg) prepared from 2.0% (w/v) chitosan dispersions were transferred into an 8 mm disc assembly unit (Specac Ltd., Slough, UK) and 2 tonnes of pressure applied for 10 s in order to generate discs suitable for bioadhesion analysis using a texture probe analyser. Discs were immediately transferred to a desiccator until required. Porcine oesophagi were obtained from a UK Food Safety Agency-approved abattoir (P.C. Turner, Farnborough, UK) and stored in saline solution (0.90% (w/v) sodium chloride) during transportation. Oesophageal epithelial tissue was separated from the surrounding smooth muscle within 2 h of collection and flash frozen in liquid nitrogen to prevent the formation of large ice crystals. Thereafter, tissue was stored at 40 8C and defrosted in saline solution at room temperature prior to use. Tissue sections (30 mm  60 mm) were excised and fixed to a

Plexiglas support covered with black electrical tape, using cyanoacrylate glue (Loctite III; Henkel Ltd., Hemel Hempstead, UK). Artificial saliva (1 mL, pH 7.0) comprising 5 mmol dm 3 sodium bicarbonate, 7.36 mmol dm 3 sodium chloride, 20 mmol dm 3 potassium chloride, 6.6 mmol dm 3 sodium dihydrogen phosphate monohydrate and 1.5 mmol dm 3 calcium chloride dihydrate was added to the tissue surface prior to testing to further moisten the mucosal epithelium. The test disc was fixed onto a movable probe using a 4 mm diameter circle of double-sided tape (3M; Bracknell, UK). Maximum force of detachment (Fmax) and work of adhesion (WOA) were determined using a TA.XT plus Texture Analyser with a 5 kg load cell (Stable Micro Systems, Godalming, UK). The test conditions comprised: pre-test probe lowering speed, 0.6 mm s 1; test speed, 0.1 mm s 1; applied force, 0.2 N; contact time, 300 s; post-test probe removal speed, 0.1 mm s 1; return distance, 5.0 mm; trigger force, 0.05 N. Data were acquired and expressed as force–distance plots using v.2.0.0.3 texture exponent software (Stable Micro Systems).

2.3.7.

Statistical analysis

Particle size analysis and determinations of fluoride loading produced data that fitted the normal distribution, confirmed by the Kolmogorov–Smirnoff test (SPSS v.14; IBM) and were analysed using the one-way analysis of variance (ANOVA) test (SPSS v.14: IBM) or unpaired two-sample t-test assuming unequal variance (Microsoft Excel 2003) to determine significant differences at the 5% probability level. Due to unequal variance encountered during texture probe analysis, significant differences between multiple data sets were determined using a Kruskal–Wallis ANOVA test (SPSS v.14; IBM). Subsequent significant differences occurring between each paired location were analysed using the Mann–Whitney U test (SPSS v.14; IBM). Both were analysed at the 5% probability level.

3.

Results

3.1.

Microparticle manufacture

In the present investigation the mass ratio of polymer to fluoride was fixed at 5:1 and two chitosan concentrations (1.0% and 2.0% (w/v)) were selected for investigation. Glutaraldehyde, where used, was added to the dispersions as a crosslinker at concentrations of 0.10% or 5.0% of the dry weight of chitosan. Microparticles prepared in the absence of glutaraldehyde generated white flocculated powders. Where significant deposition of solids occurred within the spray cylinder during manufacture, powder yields were correspondingly reduced. Increasing the chitosan concentration to 2.0% reduced the yield, whilst the addition of glutaraldehyde further suppressed the yield and conferred a roseate hue to the recovered powder. A repeat synthesis of microparticles from dispersions containing 2.0% chitosan only (control) and 2.0% chitosan plus 0.40% NaF (Sample 4) resulted in recovered yields of 19% and 42%, respectively, thus the addition of NaF to the chitosan dispersion more than doubled the yield. Visual inspection of the dispersions prior to spray drying revealed a markedly lower viscosity in the case of the dispersion

233

journal of dentistry 40 (2012) 229–240

Table 2 – Particle size distribution of spray-dried microparticles determined by laser diffraction (n = 4), expressed as volume mean diameter, D (mm). Sample ID

Volume mean diameter (mm  SD)

Volume percentiles mean (SD) D[v, 10]

1 2 3 4 5 6 Control

4.37 3.49 3.83 5.31 5.35 4.99 3.63

(2.27) (0.39) (0.70) (0.53) (0.16) (0.04) (0.28)

0.88 0.99 1.14 1.08 1.08 1.11 0.85

containing NaF. This is almost certainly attributable to electrostatic screening effects and has the practical effect of facilitating atomisation thereby increasing the efficiency and rate of drying. Higher viscosity dispersions give rise to larger droplets that may fail to dry completely, and are deposited instead on the walls of the spray drier.

3.2.

Particle size distribution

The results presented in Table 2 show that all microparticles derived from spray drying chitosan-based dispersions exhibited volume mean diameters (VMD) <6 mm. Increasing the chitosan concentration from 1.0% to 2.0% effected a predictable increase in VMD due to the corresponding increase in the total solid content of the starting dispersions. For Samples 5 and 6, this difference was significant ( p  0.01), however the increase in VMD was directional only for Sample 4 ( p > 0.05) due to the unexpectedly high variance associated with the VMD of Sample 1 microparticles. Inspection of the values for D[v,90] presented in Table 2 clearly show that microparticles prepared from dispersions containing 2.0% chitosan exhibited significantly greater polydispersity than those prepared from dispersions containing 1.0% chitosan ( p < 0.05). The inclusion of glutaraldehyde as a cross-linking agent had no significant impact on size distribution at either of the two selected chitosan concentrations ( p > 0.05). Control microparticles prepared from 2.0% chitosan dispersions in the absence of both NaF and glutaraldehyde exhibited a VMD of

Table 3 – Fluoride content of chitosan–fluoride microparticles (n = 6). Sample ID 1 2 3 4 5 6 a

Mean F content (mg mg 1)  SD 59.2 56.9 54.3 61.7 62.8 60.5

(1.7) (0.0) (1.2) (0.0) (1.0) (0.0)

Theoretical F contenta (mg mg 1)

Entrapment efficiencyb (%)

76.5 74.9 73.5 75.7 74.9 71.9

77.4 76.0 73.9 81.5 83.8 84.1

Theoretical contents are calculated based on the accurately recorded weights of NaF, chitosan and glutaraldehyde (where present) for each dispersion. b Entrapment efficiency (%): mean actual F content/theoretical F content  100.

(0.04) (0.05) (0.07) (0.66) (0.02) (0.01) (0.02)

D[v, 50] 2.44 2.45 3.16 3.34 3.34 3.23 2.48

(0.08) (0.02) (0.62) (0.02) (0.01) (0.04) (0.06)

D[v, 90] 6.41 5.56 6.94 10.53 10.58 9.90 6.97

(0.50) (0.16) (1.36) (0.73) (0.19) (0.12) (0.52)

3.6  0.28 mm. As expected, control microparticles exhibited significantly smaller VMD versus Samples 4–6 ( p < 0.0001).

3.3.

Fluoride loading

The fluoride contents of the spray dried microparticles are shown in Table 3. As expected, no fluoride was detected in the control microparticles prepared using the chitosan-only dispersion. The fluoride content and entrapment efficiency of spray dried microparticles were significantly enhanced when the starting concentration of chitosan in the original dispersion was increased from 1.0% to 2.0% ( p < 0.05). The theoretical fluoride contents are calculated assuming complete uptake of the dispensed NaF by the accurately weighed chitosan and (where appropriate) glutaraldehyde components. The addition of glutaraldehyde had no significant effect on the fluoride content and entrapment efficiency ( p > 0.05).

3.4.

Fluoride release profiles

The rate of fluoride release from the chitosan microparticles was measured over a period of 360 min in pH 5.5 isotonic media under sink conditions as described in Section 2.3.5. Due to the progressive replacement of the dissolution medium with fresh media as sampling progressed, monitoring of fluoride release beyond 360 min was effectively prohibited as analyte concentrations were below the detectable limit of the ISE. The cumulative percentage release of fluoride from microparticles prepared from dispersions containing 1.0% and 2.0% chitosan are shown in Figs. 1 and 2, respectively. Microparticles exhibited broadly similar biphasic release profiles characterised by an initial burst release during which 40–60% of the associated fluoride had been released by the 10 min time point, the first of the seven sampling points. For the microparticle series prepared from dispersions containing 1.0% chitosan, Samples 1–3 had released 40%, 59% and 43% of the total fluoride, respectively, at 10 min. Fluoride release from Samples 1 and 3 were not significantly different ( p > 0.05), however fluoride release was significantly greater from Sample 2 versus Samples 1 and 3 ( p < 0.001). At the 10 min time point microparticle samples 4–6, prepared from dispersions containing 2.0% chitosan, had released 48%, 52% and 58% of the total fluoride, respectively. Fluoride release from Samples 4 and 5 at 10 min was not significantly different ( p > 0.05), however fluoride release from Sample 6 was significantly greater at 10 min compared to Sample 4; fluoride

journal of dentistry 40 (2012) 229–240

85

85

75

75

Fluoride release (%)

Fluoride release (%)

234

65

55

45

65

55

45

35 0

60

120

180

240

300

35

360

0

60

120

Time (min)

180

240

300

360

Time (min)

Fig. 1 – Cumulative mean percentage fluoride release from spray dried chitosan microparticles into an isotonic solution containing NaCl and CaCl2 adjusted to pH 5.5 (n = 3, WSD). Microparticle Samples 1–3 were prepared from dispersions containing 1.0% chitosan plus 0.20% NaF (^); 1.0% chitosan, 0.0010% glutaraldehyde and 0.20% NaF(&); 1.0% chitosan, 0.050% glutaraldehyde and 0.20% NaF(~).

Fig. 2 – Cumulative mean percentage fluoride release from spray dried chitosan microparticles into an isotonic solution containing NaCl and CaCl2 adjusted to pH 5.5 (n = 3, WSD). Microparticle Samples 4–6 were prepared from dispersions containing 2.0% chitosan plus 0.40% NaF (^); 2.0% chitosan, 0.0020% glutaraldehyde and 0.40% NaF(&); 2.0% chitosan, 0.10% glutaraldehyde and 0.40% NaF(~).

release from Sample 6 was directionally but not significantly greater than that from Sample 5. The initial burst phase was followed by a much longer release phase, as evidenced by the data summarised in Figs. 1 and 2. Microparticles prepared from dispersions containing 1.0% chitosan and 0.0010% glutaraldehyde (Sample 2) exhibited the greatest cumulative fluoride release, with 77.4  3.9% total fluoride released after 360 min. For the period 10–360 min, microparticles prepared in the absence of glutaraldehyde (Samples 1 and 4) released the largest total amount of fluoride; specifically, 20% and 26%, respectively. The final cumulative fluoride release after 360 min was 59.7  3.8% and 73.6  1.1%, respectively. Microparticles prepared from dispersions containing 0.050% glutaraldehyde exhibited the lowest cumulative fluoride release at 360 min, irrespective of the chitosan concentration used in the initial dispersion. Although the fluoride release profiles from the various microparticle batches were not dramatically different, there were clear differences in swelling behaviour on exposure to the isotonic release media. Thus microparticles prepared in the absence of glutaraldehyde (Samples 1 and 4) swelled relatively quickly, forming individual hydrogels with diameters many times greater than their anhydrous precursors. Microparticles from Samples 2 and 5 also swelled as they hydrated, however not to the extent observed for the Samples 1 and 4. As expected, microparticles from Samples 3 and 6 exhibited minimal swelling due to the high cross-linking density. The data shown in Figs. 1 and 2 suggest that significant amounts of fluoride remain associated with the hydrated microparticles at the end of the 360 min study period. Confirmatory analyses of the residual fluoride were therefore undertaken using the methodology described in Section 2.3.4. Summing the cumulative fluoride release at 360 min and the residual fluoride content determined poststudy, ca. 90% of the calculated total fluoride load was released

or recovered from microparticles prepared from dispersions in which glutaraldehyde was absent or that contained 0.0010% glutaraldehyde (Samples 1 and 4; 2 and 5, respectively) This compares to a figure of ca. 82% for microparticles prepared from chitosan dispersions containing 0.050% glutaraldehyde (Samples 3 and 6). Microparticles prepared from a dispersion containing 2.0% chitosan and 0.40% NaF (Sample 4) were chosen to conduct additional fluoride release studies in isotonic media at pH 4.0 and 7.0. The aforementioned pHs, along with pH 5.5 were considered to be physiologically relevant to the oral cavity, and representative of pHs associated with a dietary acid 85

Fluoride release (%)

75

65

55

45

35 0

60

120

180

240

300

360

Time (min) Fig. 3 – Cumulative mean percentage release of fluoride from spray-dried chitosan microparticles into isotonic solutions containing NaCl and CaCl2 (n = 3, WSD), at pH 4.0 (^), 5.5 (&) and 7.0 (~). Microparticles were prepared from a glutaraldehyde-free dispersion containing 2.0% chitosan and 0.40% NaF.

journal of dentistry 40 (2012) 229–240

235

challenge (pH 4.0), a plaque acid challenge (pH 5.5) and unstimulated saliva (pH 7.0). The selection of microparticles from Sample 4 was driven by the observation in Fig. 2 that these microparticles had released the greatest cumulative weight of fluoride at the end of the 360 min study period (ca. 0.83 mg from 40 mg of microparticles), and had in addition achieved the maximum fluoride release during the prolonged release phase represented by the period 10–360 min (26% of total load, i.e. ca. 0.29 mg). The overlaid release profiles are shown as a function of pH in Fig. 3; cumulative fluoride release at 360 min was not significantly different for any of the three groups, nor were significant differences in fluoride release apparent at the 10 min time point following the initial burst release period ( p < 0.05).

3.5.

Microparticle morphology

Environmental scanning electron microscopy was used to examine the morphology of the spray dried microparticles prepared from dispersions containing 2.0% chitosan. A micrograph of the chitosan-only control sample is shown in Fig. 4. The particles were effectively spherical and exhibited varying degrees of agglomeration and surface indentation, the latter forming during spray drying as sub-surface cavitations collapse. Microparticles prepared from dispersions containing NaF (Samples 4–6) are shown in Fig. 5a–c, respectively. The addition of NaF and glutaraldehyde had minimal effect on the morphology of isolated microparticles, albeit there is a suggestion of increased agglomeration in the case of microparticles prepared from the dispersion containing 0.10% glutaraldehyde (Fig. 5c). There was no evidence of formation of crystalline NaF on the surface of the microparticle, nor indeed any evidence of unassociated crystalline material, both of which have previously been observed when such dispersions are dried at room temperature under vacuum (data not shown).

3.6.

Bioadhesion to porcine oesophageal mucosa

Although microparticles prepared from 1.0% to 2.0% chitosan dispersions exhibited broadly similar fluoride release profiles

Fig. 4 – Control microparticles prepared by spray drying a 2.0% chitosan dispersion in the absence of glutaraldehyde and NaF.

Fig. 5 – Spray dried microparticle Samples 4–6 prepared from a 2.0% chitosan dispersion containing 0.40% NaF plus (a) no added glutaraldehyde, (b) 0.0020% glutaraldehyde and (c) 0.10% glutaraldehyde, respectively.

in isotonic media, the latter exhibited significantly higher entrapment efficiencies (Table 3) and was available in greater quantities. Consequently, bioadhesion measurements were limited to compressed discs prepared from microparticles derived from dispersions containing 2.0% chitosan. The values of Fmax (mN) and WOA (mN mm 1) determined for the various chitosan discs adhered to oesophageal mucosae are given in Fig. 6. As is the norm, an ethyl cellulose disc was employed as the negative control and elicited no measurable bioadhesion, thereby confirming the validity of the test method. Discs prepared from the chitosan-only control microparticles were employed as a benchmark standard in order to assess the effects of NaF and glutaraldehyde on bioadhesive performance. A clear and marked decrease in both Fmax and WOA of ca. 80% and 70%,

236

journal of dentistry 40 (2012) 229–240

350

Fmax (mN)

300

WOA (mN/mm)

250 200 150 100 50 0 Control sample

Sample 4

Sample 5

Sample 6

Fig. 6 – Bioadhesion parameters of compressed discs formed from spray dried microparticles prepared from dispersions containing 2.0% chitosan (control sample), 2.0% chitosan plus 0.40% NaF (Sample 4); 2.0% chitosan, 0.0020% glutaraldehyde and 0.40% NaF (Sample 5); 2.0% chitosan, 0.10% glutaraldehyde and 0.40% NaF (Sample 6). Maximum force of detachment (Fmax) and work of adhesion (WOA) were determined using texture probe analysis and porcine oesophageal mucosa hydrated with artificial saliva (n = 6, WSD).

respectively, was observed for discs based on fluoridecontaining microparticles relative to the benchmark control; the differences were highly significant in all cases ( p < 0.001). Mean Fmax and WOA values were directionally greater for the discs prepared from the glutaraldehyde-free microparticles (Sample 4) compared to those prepared using glutaraldehyde as the cross-linker (Samples 5 and 6). However, with the sole exception of the discs prepared from Sample 5 where the mean WOA was significantly lower than that recorded for discs based on Sample 4 microparticles containing chitosan and fluoride only ( p < 0.05), differences were not significant ( p > 0.05).

4.

Discussion

The primary aim of this study was to undertake an in vitro proof of principle investigation to determine whether it was possible to employ spray drying as a low cost technique to manufacture bioadhesive chitosan/fluoride matrix microparticles with the potential to increase the retention of fluoride in the oral cavity. Microparticle manufacture using spray drying provides a low cost route for production of large quantities of matrix microparticles with utility as controlled release vehicles. Moreover, there is a precedent for the use of spray drying to produce bioadhesive chitosan microparticles and nanoparticles for drug delivery,39,40 however incorporation of fluoride as the active of choice has not previously been investigated. A previous study has reported that inclusion of glutaraldehyde as a cross-linker at concentrations representing >15% of the dried weight of chitosan has been shown to modulate metoclopramide release in vitro.37 In the present study, glutaraldehyde was employed as a crosslinker in the original dispersions at concentrations in the range 0.0010–0.10% (w/v). Glutaraldehyde is highly reactive due to the aldehyde

functions at either end of the molecule which serve to tie together secondary amine groups present in the chitosan chain via an enamine bridge. As the glutaraldehyde is present at much lower concentrations than the chitosan, it is considered to be completely reacted prior to spray drying. Any unreacted glutaraldehyde would be vented during spray drying along with the water component, the latter being present in vast excess. Batch-to-batch variability was not specifically investigated as a variable in this exploratory study, thus it is assumed that the isolated materials are representative of those that would be obtained were each synthesis to be run on multiple occasions. The inclusion of glutaraldehyde had minimal effect on entrapment efficiency and drug loading, both of which were deemed satisfactory for an exploratory study of this type. This was particularly true in the case of Samples 4–6 where entrapment efficiencies were ca. 82–84%, and sodium fluoride loadings ca. 14% (w/w). Glutaraldehyde addition had minimal effect on the size distribution of the recovered anhydrous microparticles; chitosan and fluoride concentrations were the primary determinants in this regard. However, it was noted that the rate and extent of swelling in the release media was markedly reduced for microparticles containing the higher glutaraldehyde concentration (Samples 3 and 6) compared to those formed from glutaraldehyde-free dispersions (Samples 1 and 4) and to a lesser extent those formed or in the presence of low concentrations of glutaraldehyde (Samples 2 and 5). The differences in swelling behaviour are directly attributable to crosslinking density, most notably in Samples 3 and 6. The burst phase of the fluoride release profiles suggest that fluoride situated at or near the particle surface is rapidly lost to the dissolution medium, a process promoted by the initially rapid hydration of the spray dried particle surface. The remaining fluoride is released over a much longer period of time, which in part reflects the reduction in hydration rate in the particle interior due to formation of a gel phase on the outer surface, but also reflects the greater diffusion path length as in the case of Samples 1, 2, 4 and 5, the hydrating particles swell to diameters an order of magnitude or more relative to the original anhydrous microparticle. In the case of the more highly cross-linked microparticles (Samples 3 and 6), hydration is significantly inhibited and cumulative fluoride release into pH 5.5 isotonic media at 360 min is lowest for these samples at 53% and 61%, respectively. Although hydration of Samples 3 and 6 is markedly reduced in comparison to Samples 1, 2, 4 and 5, the diffusion path length for fluoride egress is very much lower in the former case thus the two effects can be considered compensatory. Although the elemental distribution of fluoride through individual microparticles has not been examined in the present study, it is likely that the use of spray drying as the production method, together with the high mobility of the fluoride ion, results in an inhomogeneous distribution of fluoride within the microparticle matrix. When atomised droplets come into contact with the drying medium during spray drying, evaporation occurs quickly at the surface. This results in the formation of saturated vapour film. Moisture is then removed from the particle as it passes through the drying chamber. The surface indentations evident on particles prepared using this technique is the result of this drying

journal of dentistry 40 (2012) 229–240

process and has been observed previously when spray drying chitosan dispersions.37,41,42 As a result, the highly mobile fluoride ion is likely to be drawn with the solvent to the particle surface. This effect would result in increased local concentrations at the surface and sub-surface, thereby contributing to the characteristic initial burst phase observed on exposure of the microparticles to the isotonic release media. The use of secondary ion mass spectrometry (SIMS) to interrogate fluoride distribution within individual sectioned microparticles would be a useful technique to test this hypothesis. The particle sizes achieved in this study are well suited for local delivery of fluoride to the oral cavity since they are sufficiently small to avoid concerns over mouth-feel, and whilst larger in their hydrated state, the resulting microscopic hydrogels are compressible and unlikely to be ‘‘sensed’’ by the user. Increasing the chitosan concentration in the aqueous phase resulted in significant increases in particle size. This is primarily attributable to the increased solid-phase content of the original dispersion, but is also affected by the consequent increase in viscosity of the dispersion, which adversely affects fluid atomisation and thus increases the mean droplet size in the atomised spray. Increasing dispersion viscosity also reduces the recovered yield as the accompanying loss in atomisation efficiency encourages formation of large slowdrying droplets which coat the spray dryer cylinder, rather than travelling through the drier into the receiver. Low yields have been attributed by some researchers to powder adhesion to the cyclone, a conduit to the final receiver that collects dried particles that have passed through the drying cylinder, and the loss of the smallest and lightest particles through the exhaust of the spray dryer apparatus.35,42 Although some cyclone deposition did occur in this study, its extent was negligible in comparison to the amount of material coating the spray cylinder. The observation in Section 3.1 that increasing the ionic strength of the chitosan dispersions by the addition of NaF led to reduced viscosity may prove serendipitous. The behaviour is due to relaxation of the extended polymer conformation that occurs in the presence of salts.26 Further increases in NaF at fixed chitosan concentration may therefore improve entrapment efficiency and would certainly be expected to enhance drug loading and yield, as well as facilitating study of the effects of employing a broader concentration range of cross-linker. The use of chitosan with a lower molecular weight would also be expected to drive significant improvements in powder yield as the viscosity of the resulting dispersion would be inherently lower. In addition, increasing the inlet air temperature may provide a useful means of enhancing the drying rate and reducing liquid deposition within the spray cylinder.39 The substantivity of the fluoride ion when delivered from dentifrices and rinses is generally recognised to be poor, although the high reactivity of fluoride in the presence of calcium aids its retention in the form of fluoridated mineral reservoirs.43,44 Indeed, a recent in vivo fluoride clearance study conducted using a range of commercial fluoride rinses highlighted the fact that the residual salivary fluoride concentration 15 min post-use is <3% of that present immediately following rinsing.44 The production of anhydrous chitosan/fluoride microparticles offers significant advantages

237

over alternative hydrated delivery formats such as chitosancontaining rinses and dentifrices. Most important of these is the enhanced bioadhesion forces generated when an anhydrous charged polymer interacts with a hydrated mucosal surface or pellicle-coated substrate.45–48 The initial adhesive interaction is driven by hydration of polymer in the microparticle surface by water present in the fluid phase coating the oral mucosa and dentition. This has the effect of imparting mobility to the polymer chains present on the outer surface of the microparticle, thereby facilitating interpenetration, entanglement and electrostatic interaction with, for example, salivary mucins.49 Progressive hydration of the chitosan/ fluoride microparticles is one of a number of parameters that affect the subsequent release profile. When bioadhesive nonaqueous polymer gels or microparticles become fully hydrated, they form a slippery mucilage that exhibits minimal bioadhesion resulting in rapid shear-mediated displacement from the surface.46–48 As regards the potential utility of the chitosan/fluoride microparticles prepared and characterised in the present study, the microparticles could be formulated into an anhydrous dentifrice from which the release of fluoride would be initiated by the addition of water. We have previously demonstrated controlled release of triclosan in vitro from similarly formulated chitosan microparticles.36 Employing the microparticles isolated from Compositions 4– 6 where the NaF loading was ca. 14%, a 1500 mg g 1 dentifrice could be formulated by the addition of the chitosan/fluoride microparticles at a concentration of just 2.4% (w/w). Provision of a retained, controlled-release source of fluoride is likely to be beneficial as an adjunctive treatment to overthe-counter dentifrices and rinses in subjects with compromised saliva flow who are particularly susceptible to caries and erosion. Given the generally poor substantivity of fluoride delivered from conventional paste and rinse formats,43,44 attractive alternative delivery formats could include aerosolised chitosan/fluoride microparticles. Aerosol-based devices are a convenient means of ensuring product sterility and stability, and are well suited for the hygroscopic microparticle-based powders manufactured in the present study. Moreover, the anhydrous microparticles would be applied directly to the hydrated oral surfaces, maximising bioadhesive potential, inhibiting clearance and potentially providing controlled release of low concentrations of fluoride over a period of several hours following a single use. Based on current metered-dose aerosol systems, a single actuation delivering 10 mg of chitosan/fluoride microparticles isolated from Compositions 4–6 would introduce ca. 630 mg of fluoride into the oral cavity. The observation in the present study that extended fluoride release from mucoadhesive microparticles was observable in vitro for up to 360 min is an important prerequisite to successfully translating the technology into a clinical environ. Using mucoadhesive microparticle systems to achieve sustained delivery of fluoride to the oral cavity that results in elevated salivary fluoride concentrations over a period of hours rather than minutes may provide clinically meaningful benefits for end users both in terms of caries control and the management of dental erosion, though this remains to be established.

238

journal of dentistry 40 (2012) 229–240

Unlike negatively charged bioadhesive polymers such as carboxymethyl cellulose, xanthan, alginates and those based on polyacrylic acid, chitosan is cationic and consequently exhibits enhanced electrostatic interaction with the net negatively charged saliva-derived viscoelastic mucin gel coating the oral soft tissues.49 Furthermore, the small sizerange of the chitosan microparticles may facilitate their retention within interdental sites conferring a degree of protection from mechanical displacement. Potential indications for an aerosol format would include its use in the prophylaxis and treatment of dental caries, particularly root caries, in dry mouth patients with compromised saliva flow. This format would also be an attractive option for inhibiting dental erosion. A third potential format has already been utilised as part of the present study, and that is the use of compressed chitosan/fluoride discs that may be placed in the buccal pouch as a controlled release device. The compressed discs using to evaluate bioadhesive parameters described in Section 3.6 are effectively simple prototype buccal discs. The drawback of buccal discs for controlled release of therapeutics is that their comfort and acceptability is based to a significant extent on their thickness and diameter. Disc frangibility is also an issue, however specific investigation of such variables was considered beyond the scope of the present study. In common with the proposed aerosol delivery format, likely indications for buccal discs would include their use in the prophylaxis and treatment of caries, particularly root caries, in dry mouth patients with compromised saliva flow, as well as dental erosion. It should also be noted that the preparation of mucoadhesive buccal discs from the cationic chitosan polymer offers an advantage over those based on carboxylate-based polymers since there is no risk of mucosal irritancy associated with the development of local acidic pHs forming at the site of attachment, a problem of particular relevance to subjects with compromised saliva flow. A key observation impacting on the potential utility of the chitosan/fluoride microparticles was that it was the incorporation of NaF rather than the glutaraldehyde cross-linker that elicited the unexpected reduction in the bioadhesive parameters Fmax and WOA compared to the chitosan-only control. At this stage, any explanation can only be speculative, however such behaviour would be consistent with the previously discussed hypothesis that higher concentrations of fluoride are located in the surface and sub-surface region of the microparticle. If so, the early stages of microparticle hydration would result in high local concentrations of sodium and fluoride ions at the adhesion interface resulting in significant electrostatic screening and a reduction in the magnitude of the adhesive interaction. As previously suggested, application of SIMS would be of particular value in establishing whether the elemental distribution of fluoride within the microparticles, and would in addition provide a valuable tool for monitoring spray drying optimisation in future studies. Strategically, in the case of fluoride/chitosan microparticles designed for incorporation into anhydrous dentifrices, the biphasic release profile observed in the present study is beneficial since a significant proportion of the available fluoride is released during the burst phase. This is important because any anti-caries claim associated with the resulting novel dentifrice requires that it exhibits equivalence or

superiority to an FDA-recognised standard dentifrice containing NaF. This is a requirement of the relevant monograph, and typically involves in vitro testing to determine enamel fluoride uptake over a 30 min period from an aqueous slurry of the test dentifrice.50 Alternative delivery formats such as the previously described aerosolised chitosan/fluoride microparticles or buccal tablets would likely benefit from fluoride release profiles where the burst release phase is suppressed, and bioadhesion potentially enhanced. This may be achievable through reducing the solubility of the NaF in the initial dispersion used to prepare the microparticles. Alternatively, it may be preferable to change the source of fluoride employed in order to modify the inherent solubility, migration during the drying stage of microparticle formation, and fluoride egress during the release phase. On the basis of the results obtained, we reject our first null hypothesis as although glutaraldehyde had a minimal effect on entrapment efficiency and fluoride loading, it did influence fluoride release. Chitosan concentration was found to influence particle size, fluoride content and entrapment efficiency. The additional hypothesis is also rejected as the inclusion of fluoride into chitosan microparticles was found to reduce bioadhesive properties. Glutaraldehyde did not markedly influence the bioadhesive properties of the microparticles.

5.

Conclusion

The present in vitro proof of principle study was successful in manufacturing bioadhesive chitosan/fluoride microparticles using a low cost spray drying technique. Fluoride loadings and entrapment efficiencies were satisfactory for an exploratory study of this type, and extended fluoride release was observable in vitro for up to 360 min. Empirical observations suggest that recovered yields were inversely proportional to the viscosity of the original dispersion; thus yields were improved by increasing the ionic strength through the addition of higher concentrations of NaF. The characteristic initial burst phase of the release profile, coupled with the marked decrease in mucoadhesive parameters associated with discs prepared from microparticles isolated from dispersions containing 2.0% chitosan and 0.40% NaF, suggest that fluoride is not uniformly distributed through the microparticle. Based on the results obtained, the isolated chitosan/fluoride microparticles have potential utility as vehicles to enhance fluoride retention and promote its controlled fluoride delivery in the oral cavity from a variety of oral care formats. Future work should focus on refinement of the manufacturing protocol synthesis including spray drying conditions to enhance recovered yields, modulate the kinetics of the burst release phase, and ensure batch-to-batch reproducibility. The clinical relevance needs to be evaluated by running a proof of principal clinical study in which the utility of chitosan/fluoride microparticles to provide elevated salivary fluoride concentrations versus a suitable control is established.

Conflict of interest statement L-MB, MJI and JDS are members of the School of Pharmacy & Biomolecular Sciences and employees of the University of

journal of dentistry 40 (2012) 229–240

Brighton, the grant holding institution where the large majority of this research was performed. GMK conducted this research in part fulfilment of her doctorate, awarded in 2008. GMK is currently an employee of Chiesi Ltd. GRB is an employee of GSK, and GDR was an employee of GSK at the time this research was conducted. GDR is currently Director of Genesis Oral Bioscience, and holds an Honorary Senior Lectureship at the University of Brighton.

Acknowledgements The authors would like to thank GlaxoSmithKline (GSK) and the BBSRC for co-funding a PhD CASE studentship for GMK.

references

1. Selwitz RH, Ismail AI, Pitts NB. Dental caries. Lancet 2007;369:51–9. 2. Lingstro¨m P, van Ruyven FO, van Houte J, Kent R. The pH in dental plaque in its relation to early enamel caries and dental plaque flora in humans. Journal of Dental Research 2000;79:770–7. 3. Stookey GK. The effect of saliva on dental caries. Journal of the American Dental Association 2008;139:11S–7S. 4. Hicks J, Garcia-Godoy F, Flaitz C. Biological factors in dental caries: role of saliva and dental plaque in the dynamic process of demineralisation and remineralisation (part 1). Journal of Clinical and Pediatric Dentistry 2003;28:47–52. 5. Petersson GH, Bratthall D. The caries decline: a review of reviews. European Journal of Oral Biology 1996;104:436–43. 6. Pizzo G, Piscopo MR, Pizzo I, Giuliana G. Community water fluoridation and caries prevention: a critical review. Clinical Oral Investigations 2007;11:189–93. 7. Marquis R, Clock SA, Mota-Meira M. Fluoride and organic weak acids as modulators of microbial physiology. FEMS Microbiology Reviews 2003;26:493–510. 8. Balzar Ekenba¨ck S, Linden LE, Sund M-L, Lonnies H. Effect of fluoride on glucose incorporation and metabolism in biofilm cells of Streptococcus mutans. European Journal of Oral Science 2001;109:182–6. 9. Zameck RL, Tianoff N. Effects of NaF and SnF2 on growth, acid and glucan production of several oral streptococci. Archives of Oral Biology 1987;32:807–10. 10. Duguid R, Senior BW. The effect of fluoride on Streptococcus sanguis 7683 IgA1 protease production and activity. Archives of Oral Biology 1997;42:821–5. 11. Cox SD, Lassiter MO, Taylor KG, Doyle RJ. Fluoride inhibits the glucan-binding lectin of Streptococcus sobrinus. FEMS Microbiology Letters 1994;123:331–4. 12. Pandit S, Kim GR, Lee MH, Jeon JG. Evaluation of Streptococcus mutans biofilms formed on fluoride releasing and non fluoride releasing resin composites. Journal of Dentistry 2011;39:780–7. 13. Naorungroj S, Wei HH, Arnold RR, Swift Jr EJ, Walter R. Antibacterial surface properties of fluoride-containing resin-based sealants. Journal of Dentistry 2010; 38:387–91. 14. Lim BS, Lee SJ, Lim YJ, Ahn SJ. Effects of periodic fluoride treatment on fluoride ion release from fresh orthodontic adhesives. Journal of Dentistry 2011;39:788–94. 15. ten Cate JM. Current concepts on the theories of the mechanism of action of fluoride. Acta Odontologica Scandinavica 1999;57:325–9.

239

16. Margolis HC, Moreno EC, Murphy BJ. Effect of low levels of fluoride in solution on enamel demineralization in vitro. Journal of Dental Research 1986;65:23–9. 17. Zero DT, Fu J, Espeland MA, Featherstone JDB. Comparison of fluoride concentrations in unstimulated whole saliva following the use of a fluoride dentifrice and a fluoride rinse. Journal of Dental Research 1988;67:1257–62. 18. Diarra M, Pourroy G, Boymond C, Muster D. Fluoride controlled release tablets for intrabuccal use. Biomaterials 2003;24:1293–300. 19. Owens TS, Dansereau RJ, Sakr A. Development and evaluation of extended release bioadhesive sodium fluoride tablets. International Journal of Pharmacy 2005;288:109–22. 20. Wiegand A, Buchalla W, Attin T. Review on fluoridereleasing restorative materials – fluoride release and uptake characteristics, antibacterial activity and influence on caries formation. Dental Materials 2007;23:343–62. 21. Vivien-Castioni N, Gurny R, Baehni P, Kaltsatos V. Salivary fluoride concentrations following applications of bioadhesive tablets and mouthrinses. European Journal of Pharmaceutics and Biopharmaceutics 2000;49:27–33. 22. Hellwig E, Polydorou O, Lussi A, Kielbassa AM, Altenburger MJ. The influence of saliva on the dissolution of calcium fluoride after application of different fluoride gels in vitro. Quintessence International 2010;47:773–7. 23. Ganss C, Klimek J, Brune V, Schu¨rmann A. Effects of two fluoridation measures on erosion progression in human enamel and dentine in situ. Caries Research 2004;38:561–6. 24. Venasakulchai A, Williams NA, Gracia LH, Rees GD. A comparative evaluation of fluoridated and non-fluoridated mouthrinses using a 5-day cycling enamel erosion model. Journal of Dentistry 2010;38(Suppl. 3):S21–9. 25. Ren YF, Liu X, Fadel N, Malmstrom H, Barnes V, Xu T. Preventive effects of dentifrice containing 5000 ppm fluoride against dental erosion in situ. Journal of Dentistry 2011;39:672–8. 26. Singla AK, Chawla M. Chitosan: some pharmacological and biological aspects – an update. Journal of Pharmacy and Pharmacology 2001;53:1047–67. 27. Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. A direct-staining method to evaluate the mucoadhesion of polymers from aqueous dispersion. Journal of Controlled Release 2001;77:1–6. 28. Liu H, Du Y, Wang X, Sun L. Chitosan kills bacteria through cell membrane damage. International Journal of Food Microbiology 2004;95:147–55. 29. Helander IM, Nurmiaho-Lassila E-L, Ahvenainen R, Rhoades J, Roller S. Chitosan disrupts the barrier properties of the outer membrane of gram-negative bacteria. International Journal of Food Microbiology 2001;71:235–44. 30. Sano H, Shibasaki KI, Matsukubo T, Takaesu Y. Effect of chitosan rinsing on reduction of dental plaque formation. The Bulletin of Tokyo Dental College 2003;44:9–16. 31. Bae K, Jun EJ, Lee SM, Paik DI, Kim JB. Effect of water-soluble reduced chitosan on Streptococcus mutans, plaque regrowth and plaque viability. Clinical Oral Investigations 2006;10:102–7. 32. Arnaud TM, de Barros Neto B, Diniz FB. Chitosan effect on enamel demineralization: an in vitro evaluation. Journal of Dentistry 2010;38:848–52. 33. Kean T, Thanou M. Biodegradation, biodistribution and toxicity of chitosan. Advanced Drug Delivery Reviews 2010;62:3–11. 34. Govender S, Pillay V, Chetty DJ, Essack SY, Dangor CM, Govender T. Optimisation and characterisation of bioadhesive controlled release tetracycline microspheres. International Journal of Pharmaceutics 2005;306:24–40. 35. Giunchedi P, Juliano C, Gavini E, Cossu M, Sorrenti M. Formulation and in vivo evaluation of chlorhexidine buccal tablets prepared using drug-loaded chitosan microspheres.

240

36.

37.

38.

39.

40.

41.

42.

journal of dentistry 40 (2012) 229–240

European Journal of Pharmaceutics and Biopharmaceutics 2002;53:233–9. Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. Polymeric microspheres for drug delivery to the oral cavity: an in vitro evaluation of mucoadhesive potential. Journal of Pharmaceutical Sciences 2003;92:1614–23. Ganza-Gonzalez A, Anguiano-Igea S, Otero-Espinar FJ, Blanco Mendez J. Chitosan and chondroitin microspheres for oral administration controlled release of metoclopramide. European Journal of Pharmaceutics and Biopharmaceutics 1999;48:149–55. D’Ayla GG, Malinconico M, Laurienzo P. Marine-derived polysaccharides for biomedical applications: chemical modification approaches. Molecules 2008;13:2069–106. He P, Davis SS, Illum L. Chitosan microspheres prepared by spray drying. International Journal of Pharmaceutics 1999;187:53–65. Agnihotri SA, Mallikarjuna NN, Aminabhavi TM. Recent advances on chitosan-based micro- and nanoparticles in drug delivery. Journal of Controlled Release 2004;100:5–28. Martinac A, Filipovic-Grcic J, Voinovich D, Perissutti B, Franceschins E. Development and bioadhesive properties of chitosan–ethylcellulose microspheres for nasal delivery. International Journal of Pharmacy 2005;291:69–77. Corrigan DO, Healy AM, Corrigan OI. Preparation and release of salbutamol from chitosan and chitosan co-spray dried compacts and mulitparticulates. European Journal of Pharmaceutics and Biopharmaceutics 2006;62:295–305.

43. Issa AI, Toumba KJ. Oral fluoride retention in saliva following toothbrushing with child and adult dentifrices with and without water rinsing. Caries Research 2004; 38:15–9. 44. Mason SC, Shirodaria S, Sufi F, Rees GD, Birkhed D. Evaluation of salivary fluoride retention from a new high fluoride mouthrinse. Journal of Dentistry 2010;38:30–6. 45. Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. Polymeric microspheres for drug delivery to the oral cavity: an in vitro evaluation of mucoadhesive potential. Journal of Pharmaceutical Science 2003;92:1614–23. 46. Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. In situ evaluation of drug-loaded microspheres on a mucosal surface under dynamic test conditions. International Journal of Pharmaceutics 2004;276:51–8. 47. Zaman MA, Martin GP, Rees GD. Mucoadhesion, hydration and rheological properties of non-aqueous delivery systems (NADS) for the oral cavity. Journal of Dentistry 2008;36: 351–9. 48. Zaman MA, Martin GP, Rees GD. Bioadhesion and retention of non-aqueous delivery systems in a dental hard tissue model. Journal of Dentistry 2010;38:757–64. 49. Vaghani SS, Patel MM. Hydrogels based on interpenetrating network of chitosan and polyvinyl pyrrolidone for pH sensitive delivery of repaglinide. Current Drug Discovery Technologies 2011;8:126–35. 50. Anti-caries drug products for over-the-counter human use: final monograph. Federal Register 1995;60:52474–510.