Article pubs.acs.org/Langmuir
Functional Polymeric Systems as Delivery Vehicles for Methylene Blue in Photodynamic Therapy Mariana V. Junqueira,† Fernanda B. Borghi-Pangoni,† Sabrina B. S. Ferreira,† Bruno R. Rabello,‡ Noboru Hioka,‡ and Marcos L. Bruschi*,† †
Postgraduate Program in Pharmaceutical Sciences, Laboratory of Research and Development of Drug Delivery Systems, Department of Pharmacy, and ‡Postgraduate Program in Chemistry, Department of Chemistry, State University of Maringá, Maringá, Paraná, Brazil ABSTRACT: Antibiotic-resistant microorganisms have become a global concern, and the search for alternative therapies is very important. Photodynamic therapy (PDT) consists of the use of a nontoxic photosensitizer (PS), light, and oxygen. This combination produces reactive oxygen species and singlet oxygen, which can alter cellular structures. Methylene blue (MB) is a substance from the phenothiazine class often used as a PS. In this work, to facilitate the PS contact within the wounds, we have used Design of Experiments 23 plus central point to develop functional polymeric systems. The formulations were composed by poloxamer 407 [15.0, 17.5, or 20.0% (w/w)], Carbopol 934P [0.15, 0.20, or 0.25% (w/w)], and MB [0.25, 0.50, or 0.75% (w/w)]. The sol−gel transition temperature, flow rheometry, in vitro MB release, and ex vivo study of MB cutaneous permeation and retention were investigated. Moreover, the evaluation of photodynamic activity was also analyzed by in vitro degradation of tryptophan by singlet oxygen and using Artemia salina. The determination of the gelation temperature displayed values within the range of 25−37 °C, and the systems with better characteristics were subjected to rheological analysis and in vitro release profiling. The 20/0.15/0.25 formulation showed the best release profile (42.57% at 24 h). This system displayed no significant skin permeation (0.38% at 24 h), and the photooxidation of tryptophan test showed the production of reactive species of oxygen. The toxicity test using A. salina revealed that the MB associated with the light increased the mortality rate by 61.29%. Therefore, investigating the PDT efficacy of the functional polymeric system containing MB will be necessary in the future.
■
INTRODUCTION Subtherapeutic use of antimicrobials and their overuse have allowed the emergence of antibiotic-resistant microorganisms, which has become a global concern motivating the search for alternative therapies such as photodynamic therapy (PDT). PDT consists of the use of a nontoxic chemical compound (usually pigment or dye with no dark toxicity) known as a photosensitizer (PS), irradiated with an appropriate wavelength of light (in the presence of oxygen), resulting in the generation of reactive oxygen species (ROS) and singlet oxygen (1O2), which induce toxicity.1 This formation occurs by two distinct mechanisms. Type I involves the transfer of hydrogen atoms or electrons of PS to oxygen, resulting in ROS, and type II involves the transfer of energy forming 1O2. For a molecule having short half-lives (0.01−0.04 μs) and a size range of up to 0.02 μm, 1O2 is capable of interacting only with molecules near and connected to PS, mainly affecting amino acids, unsaturated lipids, proteins, and nucleic acids, with consequent damage to cell integrity and permeability.2,3 The most common PSs are from the phenothiazine class, such as methylene blue (MB). MB monomers have absorbance spectra at 664 nm and dimers spectra at 590 nm. These absorptions ensure an efficient light penetration of tissue.4 Moreover, they may be involved in different kinds of photo© XXXX American Chemical Society
dynamic reactions, and their concentration influences the mechanism and type of ROS produced. The MB monomer is responsible for producing 1O2 by photochemical pathway type II, energy transferred to the oxygen molecule from the triplet excited state MB (3MB+). On the other hand, at higher MB concentrations, the MB dimer [(MB)22+] produces semireduced MB radicals (MB*), by redox suppression of excited ground state dimers, which are oxidized to produce superoxides (photochemical pathway type I).5 The MB selectivity for microbial cells exists because of the electrostatic interaction between the positive charges of PS and negative charges of the cells’ surface, resulting in cellular damage.3,6 Many researchers have shown the efficacy of an aqueous MB solution is associated with PDT in the killing of microorganisms, even vancomycin-resistant Enterococci and methicillin-resistant Staphylococcus aureus, and treatment of cutaneous infections caused by S. aureus, Staphylococcus epidermidis, Streptococcus pyogenes, Corynebacterium minutissimum, Proprionibacterium acnes, and Escherichia coli.7−11 It also demonstrated effectiveness on sepsis control due to burns.12 In addition, the treatment of Received: June 3, 2015 Revised: December 14, 2015
A
DOI: 10.1021/acs.langmuir.5b02039 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
(Diadema, Brazil), and triethanolamine, purchased from Galena (Campinas, Brazil), was used as a neutralizing agent. All other chemicals were purchased from Merck (Darmstadt, Germany) and were of AnalaR, or equivalent, quality. Experimental Design. A full factorial design 23 was employed to determine the influence of P407 (X1), C934P (X2), and MB (X3) concentrations that have a significant influence on the study response. Each factor was set at one of two levels, low (−) or high (+), as shown in Table 1. Two central points were added to detect curvature and errors associated with isolated effects or the interactions between them.
infections on skin, nails, and mucous membranes caused by Candida albicans has been described.3,13,14 Other authors mention the death of herpes simplex virus type I, eradication of genital herpes,3,15 epithelialization and regression of leishmaniasis injuries,16,17 and wound healing.18 However, in all these studies, MB was carried in an aqueous liquid dispersion (aqueous solution). This dosage form has some disadvantages, the main one being the inability to allow good retention on the application site and runoff. For PDT, it is essential that PS remains for a sufficient time at the action site to suffer from light irradiation and produce satisfactory amounts of ROS and 1O2, and not affect adjacent cells.2 This problem can be resolved using polymers that have appropriate rheological and mechanical characteristics when in contact at body temperature, besides allowing the systems to attach to the skin and/or mucosa.19 Functional polymeric systems can be obtained by association of thermoresponsive and bioadhesive polymers, allowing the best adhesiveness and a prolonged residence time when compared with those of conventional dosage forms.20 This improved intimate contact between the formulation and the tissue results in a high concentration of active agent at the application site. Furthermore, as the systems are formed in situ, they are easier to administer and cause less irritation and pain, which leads to better adherence to treatment by patients.21−23 Therefore, functional polymeric systems with thermoresponsive and bioadhesive properties can permit the PS delivery to the site of administration, followed by its activation by the application of light. These systems can be composed of blends containing different ratios of polymers, which can be developed using Quality by Design (QbD). This tool comprises a riskbased, proactive, and scientific approach that seeks to explain the entirety of the phenomena for the development of methodologies and formulations.24 Therefore, alteration of critical process and system variables occurs to produce products of the desired quality.25,25 In the QbD approach, statistical design as design of experiments (DoE) has been widely applied. DoE allows one to plan and conduct experiments with deliberate alterations at input variables of a process or system, allowing identification of the reasons for changes in final results. This is essential for determining which variables more influence a process or system, as well, define the input parameters for the answers that are closer to the desired results, and, besides that, minimize variability in the response and in the influence of uncontrollable factors.26 In this sense, many studies have been conducted to understand interactions between variables to obtain better results.27 In this Article, we investigate the use of poloxamer 407 (P407) and Carbopol 934P (C934P) as a binary mixture to develop functional platforms for MB delivery in PDT. In particular, we use an experimental design to improve the development of the systems, which are assessed regarding their sol−gel transition temperature and rheological continuous flow properties. In addition, we determine the MB release profile of the system and its cutaneous permeation and retention. Finally, the photodynamic activity of the formulations was also studied.
■
Table 1. Matrix of Factorial Design 23 for Systems Containing Methylene Blue and Values for the Low and High Levels of Each Variable independent variables standard run
X1
X2
X3
1 2 3 4 5 6 7 8 9 10
−1 −1 +1 +1 −1 −1 +1 +1 0 0
−1 −1 −1 −1 +1 +1 −1 −1 0 0 level
−1 +1 −1 +1 −1 +1 −1 +1 0 0
factor
−1
0
+1
X1 (P407, %, w/w) X2 (C934P, %, w/w) X3 (MB, %, w/w)
15.0 0.15 0.25
17.5 0.20 0.50
20.0 0.25 0.75
Preparation of Thermoresponsive and Bioadhesive Formulations. C934P [0.15, 0.20, or 0.25% (w/w)] was dispersed in purified water by being mechanically stirred until the compound was completely dispersed. After this process, P407 [15.0, 17.5, or 20.0% (w/w)] was added and kept in contact with the preparation for 12 h to ensure complete hydration. Systems were then stirred, to ensure complete mixing of two polymers, neutralized with triethanolamine, and stored at 4 °C for 24 h. MB was added to the formulation at a level of 0.25, 0.50, or 0.75% (w/w) with stirring. All samples are stored in amber jars at 4 °C for at least 24 h before further analysis.28 Determination of the Sol−Gel Transition Temperature. To determine the sol−gel transition temperature (Tsol/gel) of polymeric systems, a MARS II instrument (Thermo-Haake, Thermo Fisher Scientific Inc., Newington, Germany) was utilized in oscillatory mode with a temperature ramp, in conjunction with parallel steel cone−plate geometry (60 mm, cone code L09006 C60/1° Ti L, separated by a gap of 0.052 mm). Samples were carefully applied to the lower plate, ensuring that formulation shearing was minimized, and allowed to equilibrate for at least 1 min before each analysis. Temperature sweep analysis was performed over the temperature range from 5.0 to 60.0 °C with a rate of heating of 10 °C/min and a frequency of 1.0 Hz. The storage modulus (G′), loss modulus (G″), dynamic viscosity (η′), and loss tangent (tan δ) were calculated using RheoWin version 4.10.0000 (Haake). The temperature at which the G′ was halfway between the values for the solution and the gel was called Tsol/gel. Tsol/gel was calculated for all formulations in which the η′ increased with an increasing temperature, and at least five replicates of each system were examined.20,28 Continuous Shear (flow) Rheometry. The continuous shear analysis of systems was performed at 25.0 and 37.0 °C, using the same rheometer previously described, in flow mode, using the cone−plate as described above. Samples were carefully applied to the lower plate, ensuring that formulation shearing was minimized, and allowed to equilibrate for at least 1 min prior to analysis. Upward and downward
MATERIALS AND METHODS
Materials. MB, mucin (from porcine stomach, type II crude), and tryptophan were purchased from Sigma-Aldrich (St. Louis, MO). P407 and C934P were purchased from BASF (Ludwisghaften, Germany) and from B. F. Goodrich (Brecksville, OH), respectively. Monobasic sodium phosphate and sodium hydroxide were purchased from Synth B
DOI: 10.1021/acs.langmuir.5b02039 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir flow curves for each formulation were measured over shear rates ranging from 10 to 2000 s−1. The shearing rate was increased and decreased over a period of 150 s, remaining in the upper limit for 10 s before decreasing. Flow shear properties of at least five replicates were determined, and the upward flow curves were modeled using the Ostwald−de-Waele (Power Law) and Herschel−Bulkley equation:20,28,29
σ = kγ n
(1)
σ = σ0 + kγ n
(2)
factor, V is the volume of Franz’s cell, and Qreal,t−1 is the accumulated permeated amount at time t − 1. The amount of MB permeated through the skin was divided by the skin area, and this value was plotted as a function of time (micrograms per square centimeter).34 Evaluation of Photodynamic Activity. In Vitro Degradation of Tryptophan by Singlet Oxygen. The polymeric systems were analyzed for photogeneration of singlet oxygen by assessing the reaction of the PS with tryptophan, a strong antioxidant. To 0.50 mL of the formulation was added 0.18 mg of tryptophan (1.76 × 10−3 mol/L). The resulting solution was illuminated by six LED light (600−700 nm, total power of 10 mW) at 17 °C to ensure the fluidity of the system. For each measurement, 32 μL of the illuminated solution was diluted with 3 mL of Mcllvaine buffer (0.015 mol/L, pH 7.25), and the tryptophan fluorescence intensity (λexc = 280 nm; λem = 300−500 nm) was monitored by a Cary Eclipse fluorimeter (Varian, Victoria, Australia) at 0, 10, 50, 140, 290, 460, and 1400 min. The result is the mean of three replicate determinations.2 In Vivo by Artemia salina. A. salina cysts were hatched in saline water (0.1 mol/L NaCl) at 25 °C in a recipient constructed of two interconnected parts, one part being kept in the dark (dark chamber, where cysts were deposited) and the other side being illuminated (illuminated chamber) with a lamp (7 W) to allow migration of the nauplii by phototropism. After 48 h, the nauplii were transferred to Petri dish containing 1.0 or 2.0 mL of each sample to be tested and kept in contact for 60 min before the mortality rate was calculated. To evaluate the toxicity of polymer systems with MB, tests were done in the presence and absence of LED light irradiation for 30 min after the contact time (600−700 nm, measured power of 5 mW). Control tests were performed with LED lights and polymer systems in the absence of MB and MB solution (with and without irradiation). After this time, the counts of living and dead nauplii were recorded. The results are the means of at least three replicate determinations.10 Statistical Analysis. The effects of polymer and MB concentrations on gelation temperature were statistically evaluated using a DoE. For this, a polynomial model that correlates the input variables and the response is further described by eq 5.
where σ is the shear stress (pascals), k is the consistency index [(pascals)n], γ is the rate of shear (inverse seconds), and n is the flow behavior index (dimensionless). In Vitro MB Release from the Systems. In vitro MB release was assessed (at least in three replicates) using a modified Franz’s cell apparatus, containing a cellulose acetate membrane. Purified water (50 mL) at 37 °C was utilized as a release medium with constant magnetic stirring. Exactly 1.0 mL of sample was placed on the membrane, ensuring the sink conditions. At predetermined time intervals (30 min and 1, 2, 3, 4, 6, 8, 12, and 24 h), aliquots (3.0 mL) of the dissolution fluid were collected, with replacement of the release medium. The MB concentration was analyzed by spectrophotometry (λ = 664 nm).29,30 To investigate the mechanism of release of MB from polymeric systems and MB solution [0.25% (w/w)], the data generated from these release studies were fitted to the general equation using logarithmic transformations and least-squares regression analysis29−32
Mt n kt M∞
(3)
where Mt/M∞ is the fraction of released drug, t is the time of release, k is the kinetic constant of incorporation of the structural and geometric characteristics of the release device, and n is the exponent that might indicate the mechanism of drug release. Ex Vivo Study of MB Cutaneous Permeation and Retention. Preparation of Pig Skin. Select skin samples were taken from white, young, and freshly slaughtered pigs (from a slaughterhouse authorized by the Ministry of Agriculture for human consumption). After the pigs had been cleaned, the subcutaneous fat was carefully removed and a skin square sample from the central region of the dorsal side of the auricle was excised using scissors and a surgery scalpel; samples with wounds, warts, or hematomas were not used. The pig skin was stored at −18 °C. On the experimental day, the skin was unfrozen at room temperature.33 Permeation Study. The permeation studies were performed using Franz’s cell apparatus. Pig skin samples were placed between the donor and receptor chambers of the cell with the dermal side in contact with the receptor medium. The receptor chamber was filled with phosphate buffer (pH 7.4) and kept at 37.0 °C. Two milliliters of each system was applied to the epidermis and carefully covered to avoid evaporation, as well as the receptor sampling side arm opening. Samples of 0.60 mL were withdrawn from the receptor medium (with replacement of the same volume) at 30 min and 1, 2, 4, 6, 12, and 24 h. Sink conditions were maintained in all cases, and the absence of air bubbles was checked after each replacement. The MB concentration was analyzed by spectrophotometry (λ = 664 nm) and determined on the basis of the available area for skin permeation, and these values were plotted as a function of time (micrograms per square centimeter).33,34 Retention Study. After the end of the permeation study, the amount of MB present in the skin was also evaluated. First, the stratum corneum (SC) was then tape-stripped 40 times with Scoth 3M. MB was extracted from the tape strips using methanol. The permeation area of the epidermis (without SC) and the dermis was then retired with scissors, and the MB content was extracted with methanol. The aliquot was analyzed by spectrophotometry (λ = 664 nm) and calculated according to eq 4.
⎛ Cmeasured, t ⎞ Q real = ⎜ ⎟V + ⎝ ⎠ D
∑ Q real,t− 1
y = b0 + b1X1 + b2X 2 + b3X3 + b12X1X 2 + b13X1X3 + b23X 2X3 (5) where y is the response, b0 is the arithmetic mean response, b1−b3 are the estimated coefficients for X1−X3, respectively, and b12−b23 are the estimated coefficients for interaction terms. The paired Student’s t test was used to determine if the dynamic viscosity of the formulations increased significantly with an increase in temperature (gelation). The toxicity test was evaluated by three-way analysis of variance (ANOVA). In all cases of ANOVA, post hoc comparisons of the means of individual groups were performed using Tukey’s Honestly Significant Different test. In all tests, a p value of 1 indicates kinetic type super case II (extreme form of transport, with abrupt solvent input presser the core of the systems).40,41 The n value for the solution aqueous was near 0.5, indicating a behavior governed by Fickian diffusion (Table 2). Systems 17.5/ 0.20/0.50 and 20/0.15/0.75 display values of n near of 1.0, indicating that these formulations presented a behavior predominantly dependent on polymer chain relaxation, which agrees with the polymer matrix that is composed of 98.87 and 99.25% (w/w) poloxamer. Polymeric association is an endothermic process that allows the restoration of the free hydrogen bonding structure of water restructuring the micellization process. Thus, C934P displayed a litter influence of kinetics of MB release. However, even with the same percentage of P407, system 20/0.15/0.025 displays an n value of >1 (1.1301 ± 0.0096), probably because of the low MB concentration in this formulation compared with the others. Increasing the MB concentration (0.50 and 0.75%) resulted in an increase in the absolute concentration of MB released but decreased the percentage release. However, the highest concentration of P407 reduced the level of MB release, possibly because of the presence of a large number of polymeric chains. This way, the systems with the highest concentration of MB did not show the highest level of release. Moreover, the higher MB concentration produces self-aggregates that block the membrane, preventing their transference to the release medium. The times required to release 10, 30, and 50% of the original concentration of MB were calculated (Table 2). System 20/0.15/ 0.75 had the slowest release and the lowest percentage, particularly for the 50% where the time required for release was 3−4 times greater than for the other systems. However, system 20/0.15/0.25 exhibited the quicker release, with 0.6% of MB released in just 30 min, and did not display self-aggregates. This concentration is considered to be enough for PDT.3 On the basis of the result obtained, the 20/0.15/0.25 formulations exhibited the faster release and for this reason were submitted to drug cutaneous permeation and retention, generation of ROS, and toxicity with A. salina. MB Cutaneous Permeation and Retention. The permeation of a substance into the skin is dependent on its physicochemical characteristics and behavior when incorporated into a system, and the skin properties, especially the stratum corneum (SC), the main barrier against the entry of several substances. In this specific formulation, it is necessary that the MB has a low flow and does not reach the bloodstream.42 Even though the MB concentration in the systems was close with saturation, there was a small displacement to the release medium. This explained by the fact that in vitro drug permeation from polymeric systems depends on the concentration and physicochemical characteristics of the polymers. According to Escobar-Chavéz et al.,43 the level of penetration from P407 systems is very low, and thus, the release is prolonged. This is probably related to the water channels formed by the micelles, which occurs during the diffusion of the actives. Another factor, which can justify the low level of permeation, is the hydrophilic properties of MB. The skin has a hydrophobic barrier (SC), which hinders the passage of substances with an affinity for water. However, when the SC is missing or broken, the extent of skin permeation is greater. After 24 h, the SC was separated from the epidermis (without SC) and dermis and the value of MB for each layer was calculated (Figure 5). The results confirmed that there is a barrier preventing the passage of the active through the skin and
recovering the original rheological proprieties that it possessed before administration.28 To allow statistical comparisons of the effects of each polymer and MB on the flow properties of the systems, the upcurve of the rheogram was mathematically defined using eq 1 from the consistency index and flow behavior index. The consistency index values of systems 17.5/0.20/0.50, 20/0.15/0.25, and 20/ 0.15/0.75 at 25 °C were 3.41 (±0.38), 154.71 (±6.93), and 49.21 (±6.32) and at 37 °C were 160.25 (±10.07), 242.18 (±15.93), and 229.78 (±2.60), respectively. For the flow behavior index, the results were 0.63 (±0.01) and 0.12 (±0.01) to the central point, 0.08 (±0.01) and 0.07 (±0.01) to systems with 0.25% MB, and 0.32 (±0.02) and 0.06 (±0.01) to systems with 0.75% MB at 25 and 37 °C, respectively. These results showed the rheological behavior was influenced by the temperature and concentration of P407 and MB. In Vitro Release Studies. Systems composed of poloxamer 407 and carbomer are commonly used to control drug release.28,29 Moreover, for PDT it is necessary that the PS be available to receive light radiation for full activity. The in vitro MB release profile provides a prior understanding of the behavior of the formulations. The utilization of a support membrane is common in this test. Thus, it is necessary to evaluate and consider the influence of this membrane on the release profile.39 For this was used an aqueous solution of 0.25% (w/w) MB to evaluate possible interference. The aqueous solution of MB showed a first-order release, with total release (100%) occurring in ∼3 h, with 93.43% already released in 2 h. This happens because the MB has to diffuse through the membrane before contacting the release medium. In contrast, the systems presented prolonged release of PS. The amount of MB released in 3 h was 3.41, 5.06, and 7.66% and in 24 h was 16.79, 30.73, and 42.57% for 20/0.15/0.75, 17.5/ 0.20/0.50, and 20/0.15/0.25 systems (P407/C934P/MB), respectively (Figure 4).
Figure 4. In vitro release profile of MB from a 0.25% (w/w) aqueous solution for systems 17.5/0.20/0.50, 20/0.15/0.25, and 20/0.15/0.75 (P407/C934P/MB). Standard deviations are smaller than the symbols. In all cases, the coefficient of variation of replicate analyses was
