1311
Electrophoresis 2012, 33, 1311–1321
Chris S. Ivanoff Timothy L. Hottel Franklin Garcia-Godoy College of Dentistry,The University of Tennessee Health Science Center, Memphis, TN, USA
Received September 23, 2011 Revised January 9, 2012 Accepted January 11, 2012
Research Article
Dielectrophoresis: A model to transport drugs directly into teeth The article describes an innovative delivery system based on the principles of dielectrophoresis to transport drugs directly into site-specific intraoral targets. The hypothesis that a drug can be driven into tooth enamel during the application of an applied electrical potential difference was tested by the authors in in vitro studies comparing dielectrophoresis to diffusion to transport carbamide peroxide and fluoride. The studies showed that these agents can be transported directly into teeth using an alternating current (AC) electric field more effectively than diffusion. It was found that a 20-min bleaching treatment on human teeth with dielectrophoresis increased carbamide peroxide absorption by 104% and, on average, improved the change in shade guide unit 14 times from 0.6 SGU to 9 SGU. After applying a 1.23% acidulated phosphate fluoride gel to bovine incisors for 20 min by dielectrophoresis or diffusion, analysis with wavelength dispersive spectrometry determined that dielectrophoresis doubled fluoride uptake in the superficial layers compared to diffusion, and drove the fluoride significantly deeper into enamel with an uptake 600% higher than diffusion at 50 m depth. Finally, dielectrophoresis promises to be a viable model that can potentially be used clinically to deliver other targeted drugs of variable molecular weight and structure. Keywords: Dielectrophoresis / Diffusion / Drug transport
1 Introduction An innovative technology that could potentially transport drugs “directly” into teeth using an alternating current (AC) electric field is being evaluated. This electrochemical delivery approach utilizes the principles of dielectrophoresis (DEP) and is not iontophoresis (IP). The advantages of delivering antibiotics, anesthetics, and antiinflammatory drugs directly to site-specific intraoral targets would be a breakthrough in dentistry that would potentially eliminate systemic side effects and risks commonly associated with oral drug delivery. It also implies the additional benefit of efficient and targeted drug delivery that is pain-free. 1.1 Dielectrophoresis The general principles of DEP were first outlined by Pohl in 1951 [1–3]. Much like electrophoresis (EP), DEP describes
Correspondence: Dr. Chris S. Ivanoff, Department of Bioscience Research, College of Dentistry, University of Tennessee Health Science Center, 875 Union Avenue, Memphis, TN 38163, USA E-mail:
[email protected] Fax: (901) 448-2744
Abbreviations: CP, carbamide peroxide; DEA, dielectric analysis; DEP, dielectrophoresis; EP, electrophoresis; HP, hydrogen peroxide; IDE, interdigitated (array) electrode; IP, iontophoresis; LOC, lab-on-a chip; PCB, printed circuit board; SGU, shade guide units C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI 10.1002/elps.201100505
the movement of particles under the influence of applied electric fields. DEP is the electrokinetic motion of dielectrically polarized particles in a nonuniform electrical field due to the unbalanced force of the electrical field on the particle’s induced dipole moment. The phenomenon of DEP occurs both in AC and direct current (DC) electric fields and can be applied both to charged and neutral particles. In the case of DEP, the development of nonuniform electric fields will cause a net force on any polarizable object, charged, or neutral. However, the strength of the force depends strongly on the medium and particles’ electrical properties, on the particles’ shape and size, as well as on the frequency of the applied electric field. DEP exploits the differences in particle dielectric properties to allow their manipulation and characterization. Particles can be trapped or moved between regions of high or low electric fields due to the polarization effects in nonuniform electric fields. By varying the applied electric field frequency, as well as the type of current (DC or AC), the magnitude and direction of the dielectrophoretic force on the particles can be controlled, thus providing the means for their manipulation, separation, or orientation [4]. DEP techniques have been used to manipulate cells, to concentrate a single cell type from a heterogeneous mixture, or alternatively, separate different cell types [5]. Washizu and Kurosawa showed it was possible to use DEP to manipulate and stretch DNA molecules [6]. There are reports of the
Colour Online: See the article online to view Figs. 1 to 5 in colour. www.electrophoresis-journal.com
1312
C.S. Ivanoff et al.
Electrophoresis 2012, 33, 1311–1321
dielectrophoretic manipulation of other nano-sized particles, such as proteins, organelles, and viruses as well [7]. Nanodielectrophoresis technology has been integrated in lab-on-a chip (LOC) systems and is currently being used to assemble nanocircuits [4]. LOC devices have drawn considerable attention in DEP manipulation of uncharged dielectric biological nanoparticles (molecules, protein, DNA, etc.) suspended in fluidic channels in the presence of an AC electrical field [8–10]. When considering the anatomy of tooth enamel and the enamel interstitial space as a DEP model comprised of many microscopic fluidic channels, the authors hypothesize that the dielectric properties of drugs can be manipulated by an AC field to enhance drug transport into the interstitial fluid of enamel. Enamel forms a fine network potential for diffusions in the inter- and intrarod prism, where electric resistivity drops rapidly when enamel is saturated in physiologic saline [11]. Metastable excess of water in enamel contributes in a large part to the electrical conductivity of enamel. Most of the water content in the intercrystalline spaces of enamel is presumed to exist as free water and only a small amount is present in form of hydroxyl groups [12]. Furthermore, water desorption from micropores is not easy, thus providing a network of fluidic channels through which DEP-mediated drug transport could potentially occur. The advent of microscale electrodes has provided the means to generate large gradients in electric field intensity and, consequently, DEP forces suitable for electrokinetic translation of microscopic particles. Depending on the relative dielectric permittivity () of the particle with respect to the suspending fluid, translation can occur in a region in the direction of the electric field gradient (positive DEP or pDEP) or in a region in the direction opposite to it (negative DEP or nDEP) [4,7]. Variation in electrode geometry changes the pattern of the nonuniform electric field which can significantly affect the electrokinetic motion of particles in suspending fluids [4,7,13,14]. Different electrode geometries can include parabolic, castellated, elliptic, and array geometries using parallel rectangular tracks or its triangular variations.
1.2 DEP fundamentals The dielectrophoretic force, FDEP , exerted by the field on a polarizable (dielectric) particle, such as a drug compound in a surrounding medium, may be approximated by the equation: 2 FDEP = 2r 3 m Re[ f CM ]∇ E rms
(1)
where m is the permittivity of the suspending medium; 2 is the root mean square value of the gradient of the ∇ E rms squared electric field; Re[fCM ] is the real part of the “Clausius– Mossotti” factor given by the equation: ∗p − ∗m f CM = Re ∗ (2) p + 2∗m
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
where ∗p and ∗m are the complex permittivities of the particle and the suspending medium, respectively, defined as ∗ = − ( j /), where and are the conductivity and angular frequency of the applied electric field, respectively, and j = √ −1 [4, 7, 13]. The force may be positive or negative, depending on the relative permittivity values of the particle and the surrounding medium. The frequency dependence of Re[fCM ] indicates the force acting on the particle varies with the frequency. Depending on the relative polarizability of the particle with respect to the surrounding medium, the particle will be induced to move either toward a region where the electrical field gradients are the strongest (Re[fCM ] > 0) (positive DEP), or toward a region where the electrical field gradients are the weakest (Re[fCM ] < 0) (negative DEP). The equation applies to both AC and DC fields [4, 7, 13]. With regard to the electrokinetic translation of a drug into enamel to which an AC current is applied, a nonpolar molecule will be acted upon in an AC electric field by opposing field vectors. In an AC field, field vectors change direction with frequency characterized by the frequency of the AC current supplied to the area. The magnitude of field vectors and, hence, the strength of the electric field can be manipulated by modulating the amplitude of the current. Field vectors can attract or repel drug molecules, causing them to orient in the field, and the sum effect of such forces on every local polarizable site within the structure of each molecule will cause a net dipole in the molecule. These dipoles may affect the movement of the molecule in a nonuniform electric field and can, in turn, be manipulated to enhance molecular transport into enamel [4, 7, 13]. 1.3 DEP drug delivery Drug transport into tooth enamel can occur by simple diffusion. By applying an electric AC current, however, drugs can be driven more efficiently, but depends on the unique transport behavior of each individual drug. The DEP system used in the present article uses a diffusion cell that comprises one plate electrode with two interdigitated electrode arrays and applies an AC current determined by two optimal operating frequencies to induce a nonuniform electric field and temporary polarization of a drug to enhance its mobility. The optimal frequencies are determined individually for each drug by dielectric analysis (DEA), which measures drug and dipole behavior. As the AC electrical signaling device continually cycles between a critical high- and low-frequency signal unique to each drug, the low-frequency wave aligns or orients the molecules over the target site, while the high-frequency wave motivates the molecules into the enamel [15–17]. 1.4 DEP in dentistry This drug delivery approach has never been attempted in dentistry until recently, the investigators used DEP to deliver carbamide peroxide (CP) into human tooth enamel [18], and fluoride into bovine tooth enamel [19]. The results showed a 104% www.electrophoresis-journal.com
General
Electrophoresis 2012, 33, 1311–1321
increase in CP penetration into human enamel compared to the conventional diffusion delivery method and 1400% enhanced whitening effect on enamel after one 20-min application of a topical CP bleaching gel [18]. Analysis with wavelength dispersive spectrometry (WDS) showed that after one 20-min application of 1.23% APF gel to bovine enamel, DEP enhanced penetration and increased uptake of fluoride on average by 600% at the depth of 50 m. The amount of fluoride delivered at 50 m was equivalent to a life-time exposure of prophylactic fluoride [19], adding further clinical relevance to the development of a DEP technique as a viable delivery model for dentistry. If results are reproduced clinically, DEP could potentially deliver targeted doses of fluoride directly into tooth enamel and eliminate side effects and risks thought to be associated with over-fluoridated water supplies, extend the efficacy of fluoride treatments, leading to improved remineralization and disease prevention. The long-term goal of the investigators is to use dielectrophoretic technology to administer anesthetics, analgesics, antiinflammatory agents, and antibiotics directly into teeth and make drug delivery in dentistry safer and more efficient.
2 Materials and methods
1313
establish a field between component electrodes or between both IDE electrodes. Logic controls the four circuits. Optimal operating frequencies are determined by DEA following procedures used by Lvovich et al. [15] and Singh et al. [16]. A low-frequency signal in the range of 0.1–100 Hz orients the drug in relation to the electric field generated by the IDEs. A high-frequency AC signal in the range of 100– 20,000 Hz motivates the drug toward and into the tissue. The AC signals are supplied to the circuits in a successive manner or simultaneously. The current generated is about 0.01–0.03 mA. The electric field strength is typically 1–4 V/mm.
2.2 Dielectric analysis The TA Instrument Dielectric Analyzer 2970 (DEA 2970; TA Instruments) measures the capacitive and conductive properties of the drug over a wide range of frequencies. The capacitive nature of a material is its ability to store electrical charge, and the conductive nature is its ability to transfer an electric charge. These electrical properties are related to molecular activity, allowing for molecular mobility of drugs and polymers.
2.1.1 DEP delivery system (electrical equipment) 2.2.1 DEA experimental procedure The apparatus includes a nonuniform electrode (Buckeye Dental, Beachwood, OH, USA; Nottingham-Spirk Design Associates, Cleveland, OH, USA), an AC signal generator (CH Industries Inc., Austin, TX, USA), and logic configured to control the electrical signal source (TA Instruments, New Castle, DE, USA). The signaling device is a portable potentiostat and operates on a 9 V battery. The diffusion cell consists of one nonuniform plate electrode (30 mm × 40 mm × 20 m) which is separated by a rubber spacer from two interdigitated array electrodes (IDEs), made from flexible printed circuit board (PCB) material, each connected to the signal generator. The basic design is depicted by Fig. 1E. The first IDE (30 mm × 40 mm × 30 m) consisting of two complementary gold-plated copper “comb” electrode components, is separated from a second identical IDE by an insulating layer containing interstitial areas that overlap those of both IDEs. The IDE component electrodes are coplanar with fingers from one electrode interleaving with its complimentary component. The fingers (14 pairs/electrode) are separated by intervals of 1 mm containing interstitial spaces 0.5 mm wide, thus providing sufficient passage to allow a therapeutic agent to pass through. The overall dimensions of the IDE arrangement are 30 mm × 40 mm × 120 m.
2.1.2 AC signal characteristics The electrode assembly is powered by four circuits that selectively apply a signal to either the first or second IDEs to C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
A 5 mg sample of the drug is deposited on a single surface gold ceramic diffusion cell for the analysis. The entire arrangement is placed in the DEA furnace chamber and brought to 37⬚C. The AC source is cycled through a range of frequencies from 0.1 Hz to 100,000 Hz. The induced current and phasing in the sample and permittivity information at each frequency is monitored and collected. The characteristic frequencies are identified by plotting the log of conductivity against the log of frequency. An optimal motivating frequency is then selected from the plot as a high operating frequency value between 100 Hz and 20,000 Hz, where conductivity is relatively high and constant, or the impedance is low and constant. Similarly, an optimal low (orienting) operating frequency is selected as a frequency value between 0.1 Hz and 100 Hz, where capacitance is relatively high and constant. The signal is then applied to the IDE assembly by the signal generator which is programmed to continually cycle the two frequencies. The laboratory application of the electrode is depicted in Fig. 1A.
2.3 CP experimental design This in vitro study compared the absorption and whitening effect of CP on human enamel after applying a bleaching gel for 20-min by DEP or diffusion. Forty freshly extracted human teeth without detectable caries or restorations were stored in distilled water at 4⬚C and used within 1 month of extraction. Two different bleaching gels, Plus White 5 Minute www.electrophoresis-journal.com
1314
C.S. Ivanoff et al.
Electrophoresis 2012, 33, 1311–1321
using the DEP diffusion cell. The teeth were rinsed twice in distilled water and dried. The outer surface of the teeth was covered with two layers of clear nail varnish to fill any defects. A class V cavity measuring 2 mm deep was created above the cementum-enamel junction by drill press using a 1/16 diamond bur (Diamond Drill Bit & Tool, Omaha, NE, USA). Teeth were placed in 50-mL beakers containing 3 mL distilled water with class V cavities below the water level to allow diffusion from the cavity to the water acting as a receiving medium. The total of the receiving medium was removed at 1 h (2 mL), 24 h, and 48 h (3 mL), and immediately replaced. The amount of H2 O2 that diffused from the dentin was measured by a colorimetric oxidation–reduction reaction kit (CHEMetrics, Inc., Calverton, VA, USA). HP concentration was measured by UV-Vis spectroscopy at 328 nm. The analysis was repeated in a second round using the remaining 24 four teeth. Eighteen teeth were randomly assigned (nine teeth in each group) to either the DEP or the diffusion group. Six additional teeth were divided into two control (drilled and not drilled) groups.
Figure 1. (A) A specimen placed on the IDE transport electrode. (B) Specimens for the dielectrophoresis and diffusion groups were cut from the same tooth. The treatment windows were demarcated with nail polish. Areas under the nail polish were used for baseline fluoride measurement. (C) After treatment, each specimen was sectioned longitudinally through the window area. (D) Sectioned specimen, embedded and polished, ready for wavelength dispersive spectrometry analysis. Fluoride and calcium were determined at various depths within the treatment window and the baseline area. (E) IDE electrode. (F) IDE is planar and comprises two component electrodes with long finger projections that interleave. Interdigit space is about 1 mm. Interstitial space is about 0.5 mm.
Speed Whitening Gel (CCA Industries, East Rutherford, NJ, USA) and 35% Opalescence PF Gel (Ultradent, South Jordan, UT, USA), were tested to evaluate the versatility of DEP drug delivery. An AC signal consisting of frequencies 0.1 Hz and 1200 Hz was applied to the DEP group for 20 min. The experiment was conducted at room temperature.
2.3.1 Comparison of absorption of hydrogen peroxide applied by DEP or diffusion Quantitative comparison of hydrogen peroxide (H2 O2 , HP) absorption was conducted after applying an over-the-counter 35% HP whitening gel (Plus White 5 Minute Speed Whitening Gel) to 30 extracted human teeth [20, 21] by conventional diffusion and DEP. Six teeth were randomly assigned to the diffusion experimental group (three teeth) or DEP experimental group (three teeth) for an initial run. The diffusion group received: 66 ± 1 mg 35% HP gel applied evenly to the tooth surface for 20 min. The DEP group received: 66 ± 1 mg 35% HP gel applied evenly to the tooth surface for 20 min C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.3.2 Comparison of shade changes in stained teeth treated with DEP or diffusion A 35% CP whitening gel (35% Opalescence PF gel) was applied to 10 extracted human teeth stained by immersion in a black tea solution for 48 h. The teeth were randomly assigned to the 20-min DEP or diffusion treatment group; whitening was evaluated by a dental spectrophotometer and macrophotography. Each tooth was divided in half from root to crown. The left half was sealed in taut Parafilm wax to prevent whitening. Opalesence gel (30 ± 5 mg) was applied to the exposed half of the tooth surface. In the DEP group, the teeth were subjected to an AC electrical field at frequencies of 0.1 Hz and 1200 Hz for 20 min. In the diffusion group, the gel was placed on the tooth and allowed to sit on the surface for 20 min. After 20 min, all teeth from both groups were wiped down with nonlinting tissue, rinsed twice in distilled water, and the Parafilm wax was removed. A Shade-X dental spectrophotometer (X-Rite, Grand Rapids, MI, USA) was used to measure shade changes. Pretest measurements were taken prior to the study; posttest measurements were taken from both the exposed (treated) right side and the waxed (untreated) left side. Triplicate measurements were taken in two shade guides, Vita Classical (Vident, Brea, CA, USA) and Chromascop (Ivoclar Vivadent, Amherst, NY, USA), with each change in shade equal to one shade guide unit (SGU). Data interpretation and statistical analysis were completed using the statistical software program R by The R Foundation for Statistical Computing and Microsoft Excel 2007. The two treatment methods (diffusion and DEP) were analyzed by betweengroup comparisons of average absorptions and a single-factor ANOVA. www.electrophoresis-journal.com
General
Electrophoresis 2012, 33, 1311–1321
2.4 Experimental design fluoride study This laboratory study compared DEP to diffusion after applying a 1.23% APF gel to 16 bovine incisor specimens (n = 16) for 20 min. Optimal high (5000 Hz) and low (10 Hz) frequencies were chosen from a Debye plot following procedures used by Lvovich et al. [15] and Singh et al. [16]. However, our pilot study indicated that 400 Hz was more effective for the enamel. Therefore, an extra run with low frequency of 10 Hz and high frequency of 400 Hz was added. Testing was performed for 20 min at 37⬚C and 100% humidity.
1315
ride concentration was averaged from both halves. Fluoride uptake was calculated by subtracting the baseline fluoride concentration from the treated site. 2.4.3 Statistical analysis One-way ANOVA followed by Student-Newman-Keuls post hoc was used to compare the fluoride concentrations between the baseline, diffusion, and DEP groups at each depth (significance level 0.05). Paired t-test was used to compare the fluoride uptake between the diffusion and DEP groups at each depth (significance level 0.05).
2.4.1 Tooth preparation and fluoride treatment Eight bovine incisors were cut longitudinally in half for a total of 16 specimens (n = 16). Both halves were coated with nail polish, except for treatment windows on labial enamel (Fig. 1B). The specimens were stored in normal saline solution at 37⬚C before treatment. Acidulated phosphate fluoride gel (PediaGel 1.23% Acidulated Phosphate Fluoride Topical Fluoride Gel; Preventive Technologies Inc., Indian Trail, NC, USA) was applied onto the treatment window of both halves. One half was subjected to passive diffusion. The DEP group specimens were wrapped in the diffusion cell and subjected to the AC current with 10 Hz and 5000 Hz at 37⬚C and 100% humidity for 20 min and wiped with Kimwipe. The specimens were then exposed to a “dry” run for an additional 20 min at 10Hz and 400 Hz. The passive diffusion specimens were kept at 37⬚C and 100% humidity for 20 min and wiped with Kimwipe.
2.4.2 Fluoride content analysis The specimens were cut through the window area (Fig. 1C), embedded in acrylic resin, and polished. The sectioned surface (Fig. 1D) was coated with carbon film, approximately 200 A˚ thick and subjected to WDS analysis, using a JEOL JXA-8600 electron probe micro-analyzer (JEOL, Akishima, Tokyo, Japan) with 15 nA probe current and 15 kV accelerating voltage. Quantitative analyses of calcium and fluoride were carried out at 10, 20, 50, 100, and 200 m from the enamel surface. Four measurements with 10 m spot size were made to obtain an average value at each depth. The quantitative analyses were done in conjunction with atomic number, absorption, and fluorescence matrix correction and calibrated standards, which determined the absolute elemental concentration (weight%) of the elements at designated spots. The standard for both calcium and fluoride was a fluorapatite (Durango, Mexico). Two areas were analyzed in each specimen: within the treatment window to determine fluoride content in the treated enamel, and under the nail polish to determine the baseline fluoride content (Fig. 1D). Fluoride concentration (ppm) at each measurement site was calculated based on weight% of fluoride and calcium oxide, assuming 37 weight% calcium content in enamel [22]. Baseline fluo C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3 Results and discussion 3.1 HP and CP results 3.1.1 Absorption results The receiving medium (distilled water) was removed for analysis and replaced at 1 h, 24 h, and 48 h. The majority of the HP eluted from the teeth into the water between 1 h and 24 h and was captured in the 24-h samples. The 1-h and 48-h samples contained HP concentrations below the threshold of reliable detection (0.1 PPM) and were excluded from analysis. A summary of the data is presented in Table 1A. The analysis found significant differences between both groups with relative percent errors of 3% or less (a single outlier had an relative percent error (RPE) of 12%). The average absorbance for the DEP group in round 1 was 79% greater than the diffusion group. The average absorbance for the DEP group in round 2 was 130% greater than the diffusion group. A single-factor ANOVA found a statistically significant difference between the diffusion and DEP groups with p = 0.01 confidence. 3.1.2 Shade change results Pre- and posttest spectrophotometer shade measurements are depicted in Table 1B. These measurements represent the average of the three readings taken for each sample with the error of measurement ±0.82 SGU. Given the imposed linearity in the shade measurement, it may also be useful to describe the error of measurement in terms of repeatability. In every set of triplicate measurements, at least two of the three measurements yielded the same shade. In cases where the third measurement differed, it was typically within the same shade group with a single unit difference in hue (a set of measurements reading A2, A2, A3, for example). The average change in SGU was 0.6 for the diffusion group, well under the error of measurement of 0.82 SGU. The average change in SGU for the DEP group was 9, significantly above the error of measurement and 14 times or 1400% greater than the diffusion group average. A single-factor ANOVA found a statistically significant difference between the diffusion and DEP treatment groups (p < 0.001). www.electrophoresis-journal.com
1316
C.S. Ivanoff et al.
Electrophoresis 2012, 33, 1311–1321
Table 1. Comparison of hydrogen peroxide absorbances and spectrophotometric shade changes after applying bleaching gel with DEP or diffusion
A. UV-visible spectroscopy absrobances at = 328 nm. Round 1 Group samples (n = 1)
Sample no. 1
(Three readings per sample)
1
2
3
Average
SD
error
Diffusion A Diffusion B Diffusion C DEP A DEP B DEP C
0.236 0.170 0.150 0.191 0.240 0.497
0.229 0.164 0.129 0.187 0.240 0.500
0.227 0.157 0.122 0.192 0.234 0.495
0.231 0.164 0.122 0.190 0.238 0.497
0.0047 0.0065 0.0146 0.0026 0.0035 0.0025
2.05 3.98 11.94 1.39 1.46 0.51
Round 2 Group samples (n = 3)
Sample no. 1
(Three readings per sample) Diffusion undrilled DEP undrilled Diffusion group A Diffusion group B Diffusion group C DEP group A DEP group B DEP group C
1 0.014 0.043 −0.012 0.025 0.032 0.038 0.021 0.033
Rel %
Sample no. 2
2 0.014 0.040 −0.012 0.022 0.030 0.039 0.016 0.032
3 0.015 0.040 −0.012 0.021 0.030 0.039 0.015 0.031
1
2
0.015 0.038 −0.012 0.022 0.029 0.038 0.016 0.032
0.015 0.037 −0.012 0.023 0.029 0.036 0.016 0.032
Sample no. 3 3 0.016 0.038 −0.012 0.023 0.029 0.036 0.015 0.032
1 0.014 0.038 −0.012 0.022 0.029 0.036 0.015 0.033
Rel %
2
3
0.015 0.038 −0.012 0.021 0.029 0.037 0.016 0.030
Average
0.014 0.037 −0.013 0.022 0.029 0.037 0.017 0.031
SD
0.015 0.039 −0.012 0.022 0.029 0.037 0.016 0.032
error
0.0006 0.0022 0.0002 0.0011 0.0009 0.0010 0.0016 0.0010
4.2 5.6 2.0 4.8 3.1 2.7 10.0 3.2
Comparison of averages Round 1 Round 2 Average
Diffusion
Dielectrophoresis
%Difference
0.170 0.014 0.092
0.274 0.031 0.153
79 130 104
B. Summary of pre- and posttest spectrophotometer results Diffusion
Dielectrophoresis
Vita classical Pre
Post
A2 B2 A2 A2 A2
A2 A2 A1 A1 A2
Chromascop ⌬ (sgu) 0 −2 3 3 0
Vita classical
Chromascop
Pre
Post
⌬ (sgu)
Pre
Post
⌬ (sgu)
Pre
Post
⌬ (sgu)
220 BL4 220 BL4 BL4
220 BL4 220 BL4 BL4
0 0 0 0 0
C4 C3 C4 C1 A2
A1 A1 A1 A1 A1
14 12 14 5 3
410 340 340 410 BL4
110 120 120 120 BL1
12 10 10 11 3
3.1.3 Summary DEP significantly enhanced the effectiveness of an in vitro 20-min whitening treatment using a 35% CP gel. The study provided physiological evidence of statistically significant differences in HP absorption and between the two treatment methods revealed by between-group comparisons of average absorptions and a single-factor ANOVA. The DEP treatment group absorbed more HP without dilution of the formula or reduction in oxidation of absorbed HP and on average, whitened the samples better than diffusion during the allotted treatment time. DEP improved HP absorption an average of 104% over conventional diffusion. The average change in SGU was 0.6 for the diffusion
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
group and 9 SGU for the DEP group, an improvement of 1400%. 3.2 Fluoride study results Spatial distributions of calcium and fluoride from two treatment areas (diffusion and DEP) with 1 m step size are shown in Fig. 2, indicating higher fluoride content in the first 50 m of the DEP group. Quantification of the fluoride concentration (Fig. 3A) shows that dielectrophoretic transport delivered fluoride up to 50 m deep, whereas the conventional fluoride application via the passive diffusion process effectively delivered fluoride to 20 m depth (p < 0.05). Compared to passive diffusion, the fluoride uptake in enamel was www.electrophoresis-journal.com
Electrophoresis 2012, 33, 1311–1321
Figure 2. Wavelength dispersive spectrometry map analysis of fluoride (F) and calcium (Ca) for dielectrophoresis and diffusion. The color bar index indicates relative concentrations in arbitrary units within each panel.
significantly higher in the DEP group at 10, 20, and 50 m depths (p < 0.05) (Fig. 3B). DEP doubled fluoride uptake in the superficial layers compared to passive diffusion, and drove the fluoride significantly deeper with an uptake six times higher than diffusion at 50 m depth. Absorption and penetration were evaluated by WDS. The results showed that fluoride concentrations in the diffusion group were significantly higher than baseline readings at 10 and 20 m depths; fluoride concentration in dielectrophoretically treated teeth was significantly higher than the diffusion group at 10, 20, and 50 m (ANOVA/Student-Newman-Keuls post hoc, p = 0.05). Significantly higher fluoride uptake was found with DEP compared to passive diffusion at 10, 20, and 50 m depths (paired t-test, p = 0.05). DEP doubled fluoride uptake in the superficial layers compared to passive
General
1317
diffusion, and drove the fluoride significantly deeper with an uptake six times higher than diffusion at 50 m depth. DEP enhanced penetration and increased uptake of fluoride on average by 600% at 50 m. The amount of fluoride delivered at 50 m was equivalent to a life-time exposure of prophylactic fluoride. Other electrochemical delivery systems have occasionally been tested on tooth enamel, especially for fluoride application. Until now the results have been disappointing. Although it has been found that EP can significantly increase the fluoride content in superficial enamel [23, 24], the advantage over passive diffusion has been inconclusive. While Gedalia reported higher fluoride uptake with IP [24], Kim found no significant difference [25], and still Lee reported better results with passive diffusion [26]. Based on results of Gedalia [24], IP would enhance fluoride uptake by enamel no more than 60% higher than topical application alone. In the current study, the DEP method doubled fluoride uptake in the superficial layers compared to passive diffusion, and drove the fluoride significantly deeper with an uptake six times higher than diffusion at 50 m depth. Comparison of enamel fluoride concentration between different studies must be carried out with caution since outcomes may depend on factors such as species, tooth stage, and previous fluoride exposure. Nevertheless, the baseline fluoride concentrations in this study were in the same range as values found in the literature. The literature reports that at 10 m depth fluoride concentrations range from approximately 170 ppm in porcine enamel to 1000 ppm in human enamel for a low-fluoride area [27, 28]. At 20 and 100 m depth, reported values range between approximately 100–400 and 30–100 ppm, respectively [28, 29]. The diffusion group in the experiments showed a fluoride uptake of 1175 ppm at 10 m depth, which decreased to 758 ppm at 20 m depth. This is comparable with values reported by Wei and Hattab, who measured 1196–1982 ppm at 10 m and 565–1240 ppm at 15 m in human premolars after 4 min gel or foam application [30]. However, fluoride uptake values in the literature vary widely and lower values have also been found [29, 31]. The baseline and passive diffusion results obtained in this study can, therefore, be considered to be within the same range reported in the literature. Figure 3. (A) Fluoride concentrations (mean and standard deviation; ppm) in enamel at 10, 20, 50, 100, and 200 m depths. Different uppercase letters denote significant differences between groups at each depth; different lowercase letters indicate differences within groups across the depths (ANOVA/Student-Newman-Keuls post hoc; significance level 0.05). (B) Fluoride uptake (mean and standard deviation; ppm) in enamel at 10, 20, 50, 100, and 200 m depths. Different letters denote significant differences in uptake values between dielectrophoresis and diffusion groups at each depth (paired t-test; p < 0.05).
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
1318
C.S. Ivanoff et al.
Using the DEP technique, fluoride uptake was substantially increased over passive diffusion, and significantly higher than both the 351 to 409 ppm fluoride uptake with IP within the first 50 m enamel layer, as well as the 113 to 178 ppm uptake by the 50–100 m enamel layer [24]. The levels of fluoride content after DEP are comparable to, or higher than, concentrations found in enamel after prolonged fluoride exposure in areas with optimally or naturally fluoridated water [32, 33] or in mild fluorosis [34]. The substantial difference in fluoride uptake compared to passive diffusion or IP methods suggests that DEP can significantly increase the transport over any existing delivery method. It is conceivable that the efficacy of DEP in enamel can be further improved. Since using DEP to deliver fluoride in enamel has never been attempted before, the determination of low and high frequencies was initially based on protocols developed for transdermal [15] and transcleral [16] applications. However, initial tests by these investigators found that, although the 5000 Hz high frequency was effective in increasing the fluoride content in the first 20 m of the enamel surface, it did not transport the fluoride deeper into the enamel than passive diffusion. The additional dry run with a reduced motivating frequency of 400 Hz, however, caused a significant increase in fluoride content; not only at 20 m depth, but also up to 50 m. Considering that the fluoride content at 20 m still increased after the dry run, it is speculated that the initial high frequency of 5000 Hz facilitated a significant deposit of fluoride into the superficial enamel surface from the fluoride gel, while the additional reduced frequency of 400 Hz drove the fluoride deeper into the enamel. More research is needed to better understand the optimal conditions and variables that affect the efficacy of DEP. However, the study confirmed the hypothesis that DEP could transport more fluoride into enamel, resulting in deeper penetration than the diffusion process. The difference of up to a sixfold increase is highly significant when put in the context of existing active delivery methods. Unlike IP, which is not selective about which charged particles are transported and may inadvertently draw positive ions such as calcium out of the tooth structure [23], DEP can selectively transport particles depending on the applied frequencies. Furthermore, unlike IP which uses a DC that has the potential side effect of generating heat that could cause tissue damage, DEP may be safer since it uses AC [15], and, thus, more flexible in modulating electromotive forces. Fluoride could theoretically be transported into tooth enamel with DEP to create a protected enamel layer of 50 m thick in one application. Uptake of fluoride with DEP is significantly better than diffusion and enhances penetration and absorbed concentration. On average, absorbance for DEP was 600% greater than the diffusion group at 50 m. Current topical treatments are readily soluble in saliva, quickly dissolving and exhausting their efficacy within 24 h. The authors hypothesize that dielectrophoretic transport will enhance fluoride delivery into enamel to extend the efficacy window of in-office treatments, promote remineralization, and prevent decay. Further study should evaluate the clinical effectiveness of DEP in enhancing C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2012, 33, 1311–1321
the remineralizing effects of fluoride on enamel, as well as to investigate more extensively conditions and factors, such as temperature, humidity, salinity, and frequency, that may influence the permeability of enamel.
3.3 Discussion While the principles of DEP transport may be difficult to grasp, the clinical importance of this technology is not. The authors have demonstrated proof-of-concept that it is possible to drive drugs directly into tooth structure without relying on diffusion and, therefore, conceptually possible to adapt this technology to treat dental infection, inflammation, and pain directly. It is here where the main focus of this research effort lies. Virtually, every drug prescribed by dentists is ingested through the mouth. The oral systemic route is often slow, inefficient, and presents possible side effects. Targeted intraoral drug delivery with DEP could potentially overcome these disadvantages using the principles illustrated in Fig. 4. Accordingly, an AC signal applying a nonuniform electric field to a drug will induce a dipole and generate an electrical field gradient that provides an electromotive force on the drug which varies in magnitude and direction with applied frequency and strength. DEP should motivate any polarizable chemical compound deposited on teeth or their supporting structures, including those that are difficult to polarize because of the absence of free charges, random distribution of charge, or large molecular size. Under optimal conditions where the selected frequencies ensure high and constant conductivity, the electromotive force would be sufficient to orient and preconcentrate the drug on the enamel surface and then motivate the drug into the tooth. Nonpolar drugs deposited on tooth enamel should be polarizable as they would on any other membrane, and the direction and force of their movement should be controlled by the applied AC electric fields. For intraoral use, a more flexible PCB electrode needs to be developed. Based on the investigators’ findings, a model for in-office bleaching treatment trays has been proposed (Fig. 5C) where the flexible perforated interdigitated electrode (Fig. 5A) is embedded in a polymeric dental tray (Fig. 5B). The tray consists of an absorbent layer to store a preloaded drug formulation inside the electrode tray; and incorporates an interchangeable terminal for connection to an electrical signal. The electrodes, in turn, would be connected to a portable signal generator that runs on a 9 V battery. The PCB is comprised of grooves between the electrode digits to permit drug transfer. A threelayer interdigitated electrode layout with perforated slots, 0.5 mm wide, duplicates the electronic design of the diffusion cell used in the experiments. Each layer is composed of two interdigitated electrodes approximately 0.5 cm wide and 2 cm long (Fig. 5D). The goal is to develop a DEP delivery model to obviate the use of needles in dentistry, as well as to administer antibiotics, analgesics, and antiinflammatory agents directly into teeth and their supporting periodontal structures to treat www.electrophoresis-journal.com
Electrophoresis 2012, 33, 1311–1321
General
1319
Figure 4. Principles of intraoral DEP drug transport. (A) No applied field E = 0 V/mm. (B) Drug molecule excitation by applied low-frequency AC signal (Hz). (C) Drug molecule excitation by applied low-frequency AC signal (Hz). (D) Drug molecules oriented by resonance AC electric field. (E) Drug molecules directed toward enamel as molecules become excited by high-frequency AC (and DC) electric field. (F) Drug molecules pushed and pulled by E = 1 V/mm and frequency greater than critical frequency (fc) enter enamel.
hyperemia, pulpitis, and periodontal disease. The study supports the conceptualization that DEP can potentially achieve this goal.
4 Conclusion Our in vitro studies have shown proof-of-concept that DEP could be implemented in a practical dental device to enhance in-office bleaching and fluoride treatments. The results suggest that this technology could lend itself to a variety of other clinical applications. Preclinical studies are currently underway to evaluate the ability of this model to promote the absorption of fluoride into human enamel. An in vivo clinical bleaching trial is currently also in development. If the results are repeated in human trials, DEP technology could potentially out-
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
perform current bleaching methods with shorter treatment times and less patient risk, while eliminating the damaging heat and intense chemical exposure associated with current treatment methods. More important, however, is the benefit to humanity, if results of our fluoride studies are reproduced clinically. Considering underdeveloped communities across the globe that lack adequately fluoridated drinking water, the clinical impact could be profoundly far-reaching. DEP could potentially deliver in one single targeted dose, an amount of prophylactic fluoride equivalent to a life-time exposure, directly and deeply into tooth enamel, while eliminating additional side-effects and risks thought to be associated with overfluoridated water supplies, leading to improved remineralization and disease prevention.
www.electrophoresis-journal.com
1320
C.S. Ivanoff et al.
Electrophoresis 2012, 33, 1311–1321
Figure 5. Proposed intraoral DEP delivery system with flexible PCB electrode and polymeric tray. (A) IDE electrode array. (B) Polymeric tray with embedded IDE. (C) Conceptualized model for clinical DEP bleaching. (D) Schematic of IDE.
In conclusion, DEP technology promises to advance the science of drug delivery in dentistry by: extending drug transport to a specific set of intraoral biological targets; eliminating the need for reformulation; improving efficacy of delivery; and negating many of the safety risks associated with the oralsystemic delivery route. This drug transport model promises to be an effective tool that could revolutionize drug delivery in dentistry. The authors thank Buckeye Dental for the use of its IDE electrode. The study was partially supported by a grant from the University of Tennessee College of Dentistry Dental Alumni Endowment Funds The authors have declared no conflict of interest.
[8] Khoshmanesh, K., Nahavandi, S., Baratchi, S., Mitchell, A., Kalantar-zadeh, K., Biosens. Bioelectron. 2011, 26, 1800–1814. [9] Seger, U., Gawad, S., Johann, R., Bertsch, A., Renaud, P., Lab. Chip. 2004, 4, 148–151. [10] Medoro, G., Manaresi, N., Leonardi, A., Altomare, L., Tartagni, M., Guerrieri, R., IEEE Sens. J. 2003, 3, 317– 325. [11] Yukikazi, H., Kawaguchi, M., Egashira, S., Hayashi, Y., Connect Tissue Res. 1998, 38, 53–57. [12] Dibdin, G. H., Arch. Oral. Biol. 1972, 17, 433–437. [13] Goddard, W. A., in: Goddard, W. A., Brenner, D. W., Lyshevski, S. E., Iafrate, G. J., (Eds.), Handbook of Nanoscience, Engineering, and Technology, 2nd ed. CRC Press, Boca Raton, FL 2007, pp. 16.2–16.5. [14] Hasan, R. S. M., Khurma, A., J. Sens. 2011, 2011 Article ID 204767, doi:10.1155/2011/204767.
5 References [1] Pohl, H. A., J. Appl. Phys. 1951, 22(7), 869–871. [2] Pohl, H. A., J. Appl. Phys. 1958, 29(8), 1182–1188. [3] Pohl, H., Dielectrophoresis, Cambridge University Press, Cambridge, UK 1978.
[15] Lvovich, V.F., Matthews, E., Riga, A. T., Kaza, L., J. Control Release 2010, 145, 134–140. [16] Singh, R. P., Matthews, M. E., Kaufman, M., Riga, A., Br. J. Ophthamol. 2010, 94, 170–173.
[4] Burke, P. J., in: Nalwa, H. S. (Ed.), Encyclopedia of Nanoscience and Nanotechnology, American Scientific, Stevenson Ranch, CA, USA 2004, 6, 623–641.
[17] Venumuddala, H. R., Study of Drug Delivery Behavior Through Bio-Membranes Using Thermal and BioAnalytical Techniques, M. Sc. Chemistry Thesis #SI420, Cleveland State University, Cleveland, OH, USA 2010.
[5] Holmes, D., Morgan, H., Eur. Cell Mater. 2002, 4, 120– 122.
[18] Ivanoff, C. S., Hottel, T. L., Garcia-Godoy, F., Riga, A. T., Am. J. Dent. 2011, 24, 259–263.
[6] Washizu, M., Kurosawa, O., Arai, I., Suzuki, S., Shimamoto, N., IEEE Trans Indust. App. 1995, 31, 447– 456.
[19] Ivanoff, C. S., Hottel, T. L., Tantbirojn, D., Versluis, A., Garcia-Godoy, F., Am. J. Dent. 2011, 341–345.
[7] Morgan H., Green, N., AC Electrokinetics: Colloids and Nano-Particles, Research Studies Press Ltd., Hertfordshire, UK, 2003.
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[20] Lee, G. P., Lee M. Y., Lum, S. O., Poh, R. S., Lim, K. C., Int. Endod. J. 2004, 37, 500–506. [21] Camps, J., de Franceschi, H., Idir, F., Roland, C., About, I., J. Endod. 2007, 33, 455–459.
www.electrophoresis-journal.com
Electrophoresis 2012, 33, 1311–1321
General
1321
[22] Retief, D. H., Cleaton-Jones, P. E., Turkstra, J., De Wet, W. J., Arch. Oral. Biol. 1971, 16, 1257–1267.
[29] Mellberg, J. R., Nicholson, C. R., Franchi, G. J., Englander, H. R., Mosley, G. W., J. Dent. Res. 1977, 56, 716–721.
[23] Wagner, M. J., Weil, T. M., J. Dent. Res. 1966, 45, 1563.
[30] Wei, S. H. Y., Hattab, F. N., Pediatr. Dent. 1988, 10, 111– 114.
[24] Gedalia, I., Weinman, J., Hermel, J., Feit, D., J. Dent. Res. 1970, 49, 1555. [25] Kim, H. E., Kwon, H. K., Kim, B. I., J. Oral. Rehabil. 2009, 36, 770–775. [26] Lee, Y. E., Baek, H. J., Choi, Y. H., Jeong, S. H., Park, Y. D., Song, K. B., J. Dent. 2010, 38, 166–175.
[31] Dijkman, A. G., Tak, J., Arends, J., Caries. Res. 1982, 16, 197–200. [32] Nasir, H. I., Retief, D. H., Jamison, H. C., Community Dent. Oral. Epidemiol. 1985, 13, 65–67.
[27] Grobler, S. R., Joubert, J. J., Arch. Oral. Biol. 1988, 33, 627–630.
[33] Li, J., Nakagaki, H., Tsuboi, S., Kato, S., Huang, S., Mukai, M., Robinson, C., Strong, M., Arch. Oral. Biol. 1994, 39, 727–731.
[28] Richards, A., Coote, G. E., Pearce, E. I. F., J. Dent. Res. 1994, 73, 644–651.
[34] Richards, A., Fejerskov, O., Bealum, V., Adv. Dent. Res. 1989, 3, 147–153.
C 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
