Gel-like TPGS-Based Microemulsions for Imiquimod Dermal Delivery

Aug 21, 2017 - ... a 13 megapixels camera (Samsung Galaxy S4, Seoul, South Korea). ... The skin was excised from the outer part of pig ears within 3 h...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/molecularpharmaceutics

Gel-like TPGS-Based Microemulsions for Imiquimod Dermal Delivery: Role of Mesostructure on the Uptake and Distribution into the Skin Isabella Telò,† Elena Del Favero,‡ Laura Cantù,‡ Noemi Frattini,† Silvia Pescina,† Cristina Padula,† Patrizia Santi,† Fabio Sonvico,† and Sara Nicoli*,† †

Food and Drug Department, University of Parma, Parco Area delle Scienze 27/A, 43124 Parma, Italy Department of Medical Biotechnologies and Translational Medicine, LITA, University of Milan, Via F.lli Cervi, 93, 20090 Segrate, Italy



S Supporting Information *

ABSTRACT: The aim of this work was to develop an innovative microemulsion with gel-like properties for the cutaneous delivery of imiquimod, an immunostimulant drug employed for the treatment of cutaneous infections and neoplastic conditions. A pseudoternary phase diagram was built using a 1/1 TPGS (D-α-tocopheryl polyethylene glycol 1000 succinate)/Transcutol mixture as surfactant system, and oleic acid as oil phase. Eight microemulsionsselected from the 1.25/8.75 oil/surfactants ratio, along the water dilution line (from 20 to 56% w/w)were characterized in terms of rheological behavior, optical properties via polarized microscopy, and supramolecular structure using X-ray scattering. Then, these formulations were loaded with imiquimod and the uptake and distribution into the skin was evaluated on full-thickness porcine skin. X-ray scattering experiments revealed the presence of disconnected drops in the case of microemulsion with 20% water content. Diluting the system up to 48% water content, the structure turned into an interconnected lamellar microemulsion, reaching a proper disconnected lamellar structure for the highest water percentages (52−56%). Upon water addition, also the rheological properties changed from nearly Newtonian fluids to gellike structures, displaying the maximum of viscosity for the 48% water content. Skin uptake experiments demonstrated that formulation viscosity, drug loading, and surfactant concentration did not play an important role on imiquimod uptake into the skin, while the skin penetration was related instead to the microemulsion mesostructure. In fact, drug uptake became enhanced by locally lamellar interconnected structures, while it was reduced in the presence of disconnected structures, either drops or proper lamellae. Finally, the data demonstrated that mesostructure also affects the drug distribution between the epidermis and dermis. In particular, a significantly higher dermal accumulation was found when disconnected lamellar structures are present, suggesting the possibility of tuning both drug delivery and localization into the skin by modifying microemulsions composition. KEYWORDS: microemulsion, imiquimod, skin delivery, TPGS (D-α-tocopheryl polyethylene glycol 1000 succinate), Transcutol, X-ray scattering, viscosity, rheological properties



INTRODUCTION

During the last decades, microemulsions (MEs) have become increasingly popular for dermal and transdermal applications, because of their potential to enhance drug permeation in comparison with conventional formulations. One of the more interesting aspects is their capability to solubilize highly hydrophobic drugs, increasing their concentration and enhancing skin uptake.7,8 However, a limitation of ME’s use for topical delivery is given by the low viscosity of these vehicles, that makes skin application complicated. Indeed, rheological properties play a very important role in dermal delivery, first of all because the higher viscosity prolonges the retention on the skin surface; secondly, because it could slow

Imiquimod (IMQ, Figure 1) is an immunostimulant drug approved for the treatment of anogenital warts, actinic keratosis, and superficial basal cell carcinoma. It is also under investigation for the treatment of other skin cancers,1,2 keloids, and hypertrophic scars 3 and as an adjuvant for skin vaccination.4,5 Given the potential of this drug, identifying a formulation able to increase and tune imiquimod drug delivery to the skin represents an important undertaking. Actually, this is a challenge, due to the low solubility of IMQ in most pharmaceutical solvents and to its poor penetration properties into the skin. The difficulties associated with IMQ formulation are highlighted by the low pharmaceutical quality evidenced for generic products marketed in South America and China in comparison to the innovator product, resulting from the modifications introduced in formula.6 © 2017 American Chemical Society

Received: Revised: Accepted: Published: 3281

April 28, 2017 August 14, 2017 August 21, 2017 August 21, 2017 DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

Figure 1. Pseudoternary phase diagram of the surfactant/cosurfactant (smix)−oil−water system. The oil phase is oleic acid; smix is a mixture of TPGS/Transcutol 1/1 (w/v). The white region indicates transparent formulations, the black region indicates viscous formulations; the overlapping domains (clear and highly viscous formulations, a and b regions) are represented in gray. In the remaining region, coarse turbid emulsions or phaseseparated systems were observed. The blue squares indicate the composition of the formulations further characterized, obtained along the water dilution line of 1.25/8.75 oil/smix ratio. As an example, ME20 is indicated in the diagram; the dotted blue arrows permit visualization of its composition on the diagram’s axis. The percentage composition by weight of the selected MEs is reported in the table (density of oleic acid = 0.895 g/mL; density of Transcutol = 0.99 g/mL). In the figure, the chemical structures of imiquimod and excipients used for ME preparation are also shown.

down water evaporation9 and the consequent formulation changes occurring after skin application;10 finally, because the formulation rheological properties can also impact on patients’ acceptability.11 For these reasons, and due to their great potential, microemulsions have been studied for the preparation of gel-like formulations. Microemulsion-based gels are prepared either by directly adding the thickening agent to the ME12,13 or by mixing the ME with an already prepared gel.8,14 However, both these methods can impact on the ME structure,12,15 possibly modifying ME solubilizing power16 and decreasing its performance.17 As an example, the addition of 1% polymers (either carboxymethylcellulose, hydroxypropyl cellulose, Pluronic F127, or Pluronic acid F68) to an isopropyl myristate based ME significanty reduced the permeation of capsaicine across rat skin.18 An alternative method to increase ME viscosity is to take advantage of the capability of certain ME excipients, when combined in specific ratios, to give viscous systems, often associated with the formation of lamellar structures. Using this strategy, no gelifying agents are needed, the formulation is easier to prepare, and the presence of lamellar structures can, in principle, protect drug from degradation, increase skin hydration, and enhance drug uptake into the skin.19−22 D-α-Tocopheryl polyethylene glycol 1000 succinate (TPGS) is a water-soluble derivative of vitamin E, formed by the esterification of vitamin E succinate with polyethylene glycol 1000 (Figure 1). The presence of a lipophilic tail (tocopheryl succinate) and a hydrophilic polar head (polyethylene glycol 1000) confers surfactant properties to this molecule. TPGS has been extensively used in pharmaceutical technology23,24 for the formulation of microemulsions,25−27 liquid crystalline systems,28,29 and a self-microemulsifying drug delivery system (SMEDDS), where the contact with water originates lamellar systems.30 TPGS-based microemulsions for dermal delivery have rarely been evaluated: Suppasansatorn27 investigated their

potential for the topical delivery of an antimelanoma prodrug, while Carvalho26 prepared nonaqueous microemulsions for the efficient skin delivery of progesterone, α-tocopherol, and lycopene. Despite the good skin permeation/retention results obtained in the cited work, the rheological aspects of those formulations, that make them suitable for skin application, were not addressed. The aim of this research was the development of an innovative gel-like TPGS-based formulation for dermal delivery, in order to take advantage of both the permeation enhancing properties of ME and lamellar systems and the favorable rheological properties of a gel. The formulation was developed starting from the construction of a pseudoternary phase diagram, using oleic acid as oil phase, due to the high IMQ solubilization capability,31 TPGS as surfactant, and Transcutol as cosurfactant (Figure 1). Several formulations, with different reological properties, were selected from preparations containing fixed ratio between oil phase and surfactants, along the water dilution line. The formulations were characterized using polarized microscopy and X-rays to collect information on the ME mesostructure. Finally, the role of ME mesostructure on skin uptake and distribution of imiquimod was investigated.



EXPERIMENTAL SECTION Materials. IMQ (MW = 240.3 g/mol; pKa = 7.3) was purchased from Hangzhou Dayangchem (Zhejiang, China). Oleic acid was purchased from Alfa Aesar (Karlsruhe, Germany), Transcutol was a gift from Gattefossè (Lyon, France). 70% perchloric acid solution, triethylamine (TEA) and albumin from bovine serum were purchased from SigmaAldrich (St. Louis, MO, USA). Tocopheryl polyethylene glycol 1000 succinate (Kolliphor TPGS) was a kind gift from BASF (Ludwigshafen, Germany). For HPLC analysis, bidistilled water was used. Acetonitrile and methanol were of HPLC grade; all other reagents were of analytical grade. 3282

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

23 °C. Frequency sweep was at 100% strain for samples ME20 and ME32 and 0.1% strain for ME36, 40, 44, 48, 52, 56. Temperature was set at 23 °C. X-ray scattering experiments. Synchrotron small-angle and wide-angle X-ray scattering techniques (SAXS and WAXS) can be profitably applied to assess the internal structure of formulations for drug delivery.32−35 Experiments were performed at the ID02 high-brilliance beamline (ESRF, Grenoble, France). Measurements were performed on each sample at 23 and 33 °C, i.e. at normal storage temperature and close to the temperature of the skin. Samples were put in plastic capillaries (KI-BEAM, ENKI srl, Concesio, Italy) with 2 mm internal diameter, mounted horizontally onto a thermostated sample holder. The X-ray beam cross section was 200 × 400 μm, with λ = 0.1 nm. The region of investigated momentum transfer, q = (4π/λ) sin(θ), was 0.0116 < q < 40 nm−1, where 2θ is the scattering angle. In order to prevent any radiation damage, several frames with very short exposure time (0.1 s) were acquired, and then checked and averaged. After angular regrouping and background subtraction, the intensity SAXS and WAXS profiles reported the sample scattered radiation intensity as a function of the momentum transfer q. The analysis of intensity peaks gave information on the structure of different formulations on length-scales from tens of nanometers down to the tenths of nanometers. Skin accumulation and permeation experiments. For permeation experiments, porcine skin was used. The skin was excised from the outer part of pig ears within 3 h from animal death, separated from the underlying cartilage with a scalpel and frozen at −20 °C until use. All tissues were used within 3 months from freezing. The skin, once thawed, was mounted on vertical diffusion cells (DISA, Milano, Italy; 0.6 cm2 surface area) with the stratum corneum facing the donor compartment. The receptor compartment was filled with PBS pH 7.4 containing 1% w/v albumin in order to increase IMQ solubility and mantain sink conditions (IMQ solubility in 1% w/v albumin: 143 ± 3 μg/mL). Different formulations (Table in Figure 1, with the exception of ME 36) were evaluated, and each condition was replicated 4 to 6 times. The formulations were applied for 6 h in the donor compartment at infinite dose (200 mg/cm2, occluded). At the end of the experiments, the receptor solution was sampled, the donor formulation was removed, the tissue was rinsed with distilled water, blotted dry with filter paper and tape-stripped twice to remove possible traces of the formulation on the skin surface. Then, the tissue was extracted using 1 mL of a mixture of oleic acid: methanol (1:3) for epidermis and 1 mL of a mixture of PEG400: methanol: 1 M HCl (1:2:2) for dermis. The extraction method was previously validated31 and the recovery was higher than 97%. To measure IMQ permeation, 1 mL of the receptor solution was transferred in an Eppendorf tube, added of 50 μL of 70% v/v perchloric acid to precipitate albumin and centrifuged (12,000 rpm, 15 min). Extraction and permeation samples were analyzed by HPLC with fluorescence detection. Statistical analysis. The significance of the results was assessed using Student’s t test. Differences were considered statistically significant when p < 0.05. All data are reported as mean value ± SD, with a single exception: for the sake of clarity, IMQ accumulation data into the skin are reported as mean value ± standard error of the mean (sem), as indicated in the legend.

Imiquimod quantitative analysis. Imiquimod quantification was performed by HPLC using a Flexar instrument (PerkinElmer, Waltham, MA, USA) and a C18 column (Kinetex C18 2.6 μm, 100 Å, 75 × 4.6 mm, Phenomenex, Torrance, CA, USA), equipped with a guard column (SecurityGuard Widepore C18, 4 × 3 mm, Phenomenex, Torrance, CA, USA). The mobile phase was a mixture CH3OH/CH3CN/H2O/TEA (180/270/530/20) isocratically eluted at 0.5 mL/min. In the case of samples from tissue extraction and permeation experiments, fluorescence detection (λexc 260 nm, λem 340 nm) was used (injection volume: 1 μL), while samples used for imiquimod solubility assessment were analyzed by UV absorbance (λ 242 nm; injection volume: 10 μL). The HPLC methods were previously validated for sensitivity, precision and accuracy.31 Pseudoternary phase diagram construction. Pseudoternary phase diagrams allow to identify the microemulsion region in multiphasic systems. They were built by using oleic acid as oil phase, water and a 1/1 (w/v; g/mL) mixture of surfactant (TPGS) and cosurfactant (Transcutol) (smix). The construction of the diagram is based on the aqueous tritation method: for fixed ratios oil/smix (1/9, 1.25/8.75, 2/8, 3/7, 4/6, 5/5, 6/4, 7/3, 8/2, 9/1) increasing amounts of water, between 5 and 95% of total emulsion content, were added. After each addition, the mixture was vortexed and left 1 min to rest, then by visual observation the viscosity and clearness of the system were evaluated. The system is clear and exhibit low viscosity in the microemulsion region, while it is clear and viscous in the microgel region where the formulation does not slide along the vial walls. The diagram was built using OriginPro 2016 (Originlab, Northampton, MA). Blank microemulsion preparation. Microemulsions have been prepared by admixing the different components into glass vials, under magnetic stirring, in the following order: oil phase, cosurfactant, surfactant and water. The composition of the microemulsions prepared is reported in Figure 1. Imiquimod loaded formulations. In order to load IMQ into ME20 (composition in Figure 1) an excess of drug was added, the suspension was left under magnetic stirring overnight, then the obtained samples were centrifuged for 10 min at 13,000 rpm to remove the excess of IMQ. In the case of microgels, the high viscosity prevented the efficacy of centrifugation. Hence, ME32−56 were prepared from IMQsaturated ME20 by adding known volumes of water, to the final water %, then vortexed to achieve homogeneity. No precipitation occurred upon water addition. Finally, to measure the IMQ concentration in ME20, aliquots of about 10 mg of formulation were taken, accurately weighted, diluted 1:100 with methanol and analyzed by HPLC. Polarized optical microscopy. In order to asses microemulsions optical properties, MEs were spread between glass slide plates, to prevent water loss. Samples were analyzed at 4×, 10×, and 20× magnification using a polarized optical microscope (Nikon, Shinjuku, Japan) and images were taken with a 13 megapixels camera (Samsung Galaxy S4, Seoul, South Korea). Rheological behavior. Rheology measurements were performed in oscillation mode with an Ares Rheometer (TA Instruments Waters, New Castle, DE, USA) equipped with a plastic cone and plate fixture. Cone diameter was 50 mm and cone angle was 0.04 rad. Sample’s linear viscoelastic region (LVE) was determined by strain sweep (10−2−10+4 strain %) at 3283

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics



than the storage modulus, G″ ≫ G′, and the viscosity being nicely constant as a function of frequency (see Supporting Information). ME36 and ME40 belong to a second regime, characterized by the crossover at G′ = G″ for some frequency value, as shown in Figure 3a for ME40. This means that a residual transient stress affects the system after removal of the external solicitation, decaying with some characteristic time, indicating internal structural rearrangement and relaxation. This time is slightly longer for ME40 as compared to ME36. This agrees with the polarized microscopy pattern of these two formulations showing a dependence on the mechanical treatment during sample deposition. Meanwhile, the viscosity becomes typical of viscoelastic systems: lower at higher frequency, with a nonzero plateau at the limit of low frequency. For ME44, ME48, ME52, and ME56 the storage modulus is higher than the loss modulus, G′ > G″, indicating a gel-like behavior, as shown in Figure 3b for ME44. The evolution in different regimes can be appreciated in Figure 3c, where the values of the ratio G″/G′ (tan δ) are reported. X-ray scattering. The structure of the MEs was determined by X-ray scattering experiments, in the range of water content 20−56%, with an oil/smix ratio 1.25/8.75, at both room and skin temperatures (23 and 33 °C). The scattering intensity spectra collected at T = 23 °C are reported in Figure 4. The presence of peaks reveals that all the samples display a defined structure on the supramolecular length-scale. The ME20 sample (panel a, black line) shows the typical intensity spectrum of microemulsions, either w/o or o/ w, with a single broad peak at qpeak = 1.1 nm−1, related to a characteristic length on the mesoscale within the system, which is constituted by disconnected droplets. For higher water content, all over the investigated range, samples exhibit spectra with more than one peak (panel a), a feature that is not usual for ordinary transparent microemulsions.37 The left shift of the peaks upon dilution corresponds to an increase in the characteristic mesoscale length. Figure 4 (panel c) reports the first SAXS peak of the MEs in the range 32−56% of water content and the corresponding qpeak position. The characteristic mesoscale length dmeso was calculated from each first peak position dmeso = 2π/qpeak. These results allow us to build the swelling behavior of the system. The swelling behavior can be defined as the exponent of the power law connecting the volume fraction of the apolar phase to the characteristic mesoscale length and gives important information on the structural properties of the oil/smix 1.25/ 8.75 formulation. In fact, given the general swelling dependence φapol ÷ d−s, in which φapol is the apolar volume fraction, the value of the exponent s is connected to the phase of the system, for example, s = 1 for the lamellar phase (monodimensional swelling), s = 2 for the hexagonal phase (bidimensional swelling), and s = 3 for the micellar phase (tridimensional swelling). Figure 4 (panel d) reports the apolar volume fraction φapol (calculated as [1 − (φH2O + φPEG)]) as a function of the characteristic distance dmeso calculated from the first peak position dmeso = 2π/qpeak for the formulations in the range 32− 56% of water content. The linear fit of the experimental points gives a slope of −1, characteristic for the swelling behavior of lamellar structures. Moreover, the decay of spectral intensity in a lower q region (data not shown) exhibits a q−2 dependence, confirming that those structures are locally lamellar, within 90 nm. Notably the formulations were transparent in all the

RESULTS AND DISCUSSION Pseudoternary diagrams phase. Using the excipients whose structures are reported in Figure 1, a pseudoternary diagram was built: it represents compositions containing water, oleic acid as oil phase, and TPGS/Transcutol 1/1 (w/v) as surfactant/cosurfactant mixture (smix). In the diagram, two different regions are highlighted: in black that of viscous formulations, in white that of clear transparent formulations. Two overlapping domains (gray), corresponding to clear gels (microgels), are present: (a) a small region, found for oil/smix 3/7, in a restricted range of water content (25−30%); (b) a wide region, found for oil/smix ratios 1/9, 1.25/8.75, and 2/8, with a water content from 40% to 53%. At higher water content (e.g. 56%), the formulation was still viscous but slightly opalescent, suggesting the presence of dispersed droplets bigger than 150 nm. It is interesting to notice that for oil/smix ratio of 0.5/9.5, no visually appreciable increase of viscosity occurs, indicating that the formation of a gel-like structure is not only due to the presence of TPGS,24,36 but a specific percentage of oil phase is required. Microemulsion characterization. Appearance and rheological behavior. The oil/smix ratio 1.25/8.75 was further investigated by preparing and characterizing micro-/nanoemulsions in the range 20−56% water content (dilution line, Figure 1); the w/w % composition is reported in the table (table in Figure 1). The formulations were transparent up to 52% water content, whereas the 56% formulation was opalescent (Figure 2).

Figure 2. Appearance of the TPGS-based MEs with increasing water content (from 20% to 56%). Oil/smix ratio 1.25/8.75 (oil phase: oleic acid). The zero-shear viscosity values are also shown for the various formulations (red circles).

Analysis by polarized light microscopy reveals that the ME20, ME32, and ME56 formulations are isotropic. On the contrary, ME36, ME40, ME44, and ME48 showed peculiar patterns, typical of birifrangent ordered structures (data not shown). These patterns were highly variable and depended on the thickness of the gel layer, the temperature, the presence of the covering slide, the sample manipulation, and the delay between spreading and microscopy observation. This suggests that the structures formed are sensitive to temperature and applied forces. The formulation viscosity, measured at 23 °C, macroscopically increases from ME20 to ME48 and then decreases for ME52 and ME56, as shown in Figure 2. According to the rheological properties at 23 °C, the formulations can be regrouped in different classes, as shown in Figure 3 and in the Supporting Information. ME20 and ME32, well outside the (b) region of Figure 1, behave as nearly Newtonian fluids, with the loss modulus being much higher 3284

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

Figure 3. Rheological properties of ME40 (panel a) and ME44 (panel b). Panel c illustrates the G″/G′ ratio for all the formulations. The horizontal line at G″ = G′ (G″/G′ = 1) marks the transition between viscous and elastic behavior.

Figure 4. Plots of X-ray scattering intensity profiles versus momentum transfer (q) as a function of water content for ME20,ME32, ME36, ME40, ME44, ME48, ME52, and ME56 (T = 23 °C) SAXS (panel a) and WAXS (panel b). In panel c the first peak position (italic) and characteristic distance dmeso (bold) calculated from the first peak position dmeso = 2π/qpeak as a function of water content are shown. Panel d: Apolar volume fraction φapol as a function of the characteristic distance dmeso. The line is the best fit obtained from the equation φapol ÷ d−s with s = 1. 3285

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

Figure 5. (Panel a) IMQ skin retention (μg/cm2) in epidermis (light blue) and dermis (blue) from microemulsions with increasing water content (mean ± sem). Diamonds represent IMQ concentration in the formulations (right side axis). (Panel b) Skin transfer efficency, calculated as % of IMQ accumulated in whole skin with respect to the overall IMQ content in ME. Symbols in panels a and b indicate statistical difference (p < 0.05) from ME20 (*); ME32 (#); ME40 (§); ME44 (@); ME48 (^); ME52 (+) or ME56 (●). (Panel c) Percentage of the total delivered IMQ accumulated in the dermis, as a function of the water content of the applied formulation. The symbol ¶ indicates values statistically different (p < 0.01) from ME20, ME32, ME40, ME44. In panels b and c the vertical dashed line marks the onset of the ordered lamellar phase.

concentration region up to 56%, where the system became opalescent (see Figure 2). Locally lamellar structures were described for rigid microemulsions, whereas disordered bicontinuous structures are typical for more flexible ones.38 Interestingly, as noticeable in Figure 4 (panel c), narrower and more defined SAXS peaks are obtained upon increasing the water content in the formulations, meaning that progressively more organized structures are formed. Samples ME48, ME52, and ME56 show the rising of an ordered lamellar phase, potentially coexistent with a less ordered phase. Lamellar gels have been reported to occur in PEG−lipid based systems.39 In the investigated transparent microemulsions the presence of polyethylene glycol groups (MW = 1000 Da) of TPGS molecules can locally induce curved spots on the flat lamellae, as defects, allowing for the formation of stable connections between adjacent bilayers, and giving rise to a transparent locally lamellar gel phase. This system is a bicontinuous mesh, in which both hydrophobic and hydrophilic regions are individually connected all over. Bicontinuous and lamellar structures have been found to exist in adjacent regions of the phase diagram of multicomponent colloidal systems. In fact, at higher water content the connections between adjacent bilayers are less possible and the system enters a lamellar phase.

The results here obtained indicate that in the investigated system a slight change in water content or temperature can induce the transition between the two structures, likely favored by the water−PEG interplay.39 At skin temperature (approx. 33 °C), the general features of the systems are retained, but with a lower degree of order (broader peaks), as can be seen in Figure 3S of the Supporting Information. The highly ordered lamellar phase is less pronounced for ME52 and ME56 and is absent in the ME48 sample. Wide angle X-ray scattering measurements were performed on the same samples. In the WAXS region, 8 nm−1 < q < 24 nm−1, structural information can be obtained regarding the very local length scale, such as the distance between lipid chains in the apolar region, dlocal. Figure 4 (panel b) reports the WAXS intensity spectra, obtained after careful empty cell and solvent signal subtraction, showing peaks centered in the region around 14 nm−1, corresponding to dlocal = 0.45 nm. At high oil content (ME20), the maximum is centered at q = 14.4 nm−1 and the calculated mean characteristic distance in the oil region is dlocal = 0.44 nm. As the water content increases, peak splitting occurs. The main peak, around 14 nm−1, shifts progressively to lower q. 3286

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

The information obtained by SAXS helps interpret this result, suggesting that a correlation exists between microemulsion mesostructure and IMQ skin delivery. In fact, enhanced permeation is observed for connected locally lamellar structures, allowing for both drug segregation into the lipophilic phase and easy diffusion within the delivery system. Both features are likely to be reduced in the presence of disconnected structures, either drops (w/o or o/w, ME20) or disconnected lamellae (ME52 and ME56). It is then evident that it is the diffusion process across the mesostructure that rules drug delivery efficiency and its partition in different districts, rather than the macroscopic viscosity of the formulation. In fact, enhanced delivery occurs for very viscous formulations, but interconnected in structure. The result is particularly relevant since it clearly shows that other factors known to play an important role in drug permeation and uptake, such as formulation viscosity,42,43 drug concentration, water content,8 and surfactant concentration,24 have a far smaller effect (if any) compared to the role of vehicle mesostructure. As also suggested by other authors,8 the mesostructure can impact on drug localization and mobility,44 also modifying its tendency to partition into the stratum corneum. Finally, the ME structure can promote the uptake of ME excipients in the stratum corneum,45 indirectly enhancing IMQ uptake. In fact, literature data indicate that oleic acid and Transcutol show a deeper and faster penetration in the SC when included in a microemulsion compared to pure solvents.46 Additionally, we have previously demonstrated that the transport of IMQ into the skin is coupled to the diffusion of the solvent into the stratum corneum.31 It is interesting to notice that formulations ME40 and ME44 showed a skin accumulation not significantly different from that obtained with Immunocare despite the drug concentration in the commercial product being more than 10 times higher than those present in the microemulsions (50 mg/g vs 3.48−3.35 mg/g). The finding could have a relevant clinical effect, in terms of efficacy and especially of formulation safety, considering the uncontrolled systemic absorption that could occur in the case of the presence of skin wounds. Together with the total amount of drug in the skin, another relevant effect of the formulation structure is on the drug distribution between the two layers, epidermis and dermis. Figure 5c reports the percentage of IMQ accumulated in the dermis, with respect of the total amount recovered in the skin (epidermis + dermis). For water contents up to ME44, only 20% of the drug is located in the dermis, whereas for ME48, ME52, and ME56 more that 50% of the delivered IMQ is recovered in the dermis. In the same figure a vertical dashed line marks the onset of the disconnected lamellar phase in the ME formulations, in correspondence of ME48, where an evident increase in IMQ dermis accumulation takes place. In other words, IMQ accumulation in the dermis displays a sharp increase while crossing the transition from interconnected to ordered lamellar mesostructure, where a lower “trapping” of water can occur. This means that with ME48, ME52, and ME56 an increased hydration of stratum coneum and epidermis occurs; as a consequence, the tissues become “more hydrophilic”, thus favoring the partition of the lipophilic IMQ (calculated log DpH 7.4 = 2.6547) into the dermis.48 Formulation ME48, being at the crossover between the regions characterized by high uptake into the skin and high transfer to the dermis, determines both a relevant IMQ

This result reveals that the average characteristic distance increases from dlocal = 0.44 nm to dlocal = 0.46 nm. The second peak or shoulder is centered around 16.5 nm−1. These results indicate a “phase separation” within the single bilayer between regions with a different packing of the lipid chains. The new phase displays a closer packing of the chains with a shorter characteristic distance d2local = 0.38 nm. Interestingly, as visible in Figure 4 (panel b), the relative importance of the new phase increases with water content. As the oil/smix ratio is constant, this result suggests that the interplay with hydration water can induce a phase separation within the single bilayer and the expansion of regions with higher short-range order. Again an important role is likely to be played by the extended, highly hydrated polyethylene glycol headgroups of TPGS molecules (PEG MW = 1000 Da, Rg ≅ 2.5 nm), that may modulate their arrangement at different water contents.39 The measured closer distance may well correspond to the tight-packing of acyl chains of vitamin E (or tocopherol), that is known to partially segregate within lipid bilayers.40 On the other hand, the loose-packing regions are likely to be depleted of TPGS molecules. Anyway, the observed segregation on the local scale confirms that the presence of defects, regions with different chain packing and thus with different curvature, increases with water content, inducing the observed gel behavior on the mesoscale. Moreover, also the features of the polar/apolar interface are affected by segregation. Regions of the interface depleted of TPGS molecules do not benefit from the shielding effect of hydrated polyethylene glycol headgroups. In these portions of the lamellae, the surface is likely to be more accessible from both the oil and the water sides of the mesostructure. Imiquimod skin delivery. In order to evaluate the potentialities of these systems, MEs were loaded with imiquimod. ME20 was saturated with the drug (concentration 4.35 ± 0.06 mg/g) and the other MEs were obtained by dilution with water. No precipitation occurred upon dilution, showing that IMQ was totally and stably solubilized within the ME structure. The composition of the formulations is reported in Figure 1, and the corresponding IMQ concentrations in the preparations are reported in Figure 5a. The IMQ-loaded MEs were then applied to the skin. After 6 h, IMQ was recovered in the epidermis and dermis (Figure 5), while no transdermal permeation was observed. The presence of albumin in the receiving phase assured sink conditions; thus, the lack of permeation was attributed to the slow permeation rate and the limited application time (6 h), selected as a reasonable time of formulation persistence on the skin. To evaluate the therapeutic relevance of IMQ skin retention data obtained with MEs, the values previously obtained with a commercial product (Immunocare) were taken as reference (1.89 ± 0.77 μg/cm2).31 The commercial formulation is a coarse emulsion, with high IMQ loading (50 mg/g), requiring a complex application regimen to minimize its toxicity41 and thus characterized by a low patient compliance. As can be seen in Figure 5a, the total amount of IMQ retained in the skin (epidermis + dermis) depends on water content: it increases up to 40−44% of water and then decreases. The relative distribution between epidermis and dermis seems to be influenced by the water content as well. Panel b reports the percentage of IMQ accumulated in skin with respect to the overall IMQ content in ME, for the various formulations. It can be seen that also the transfer efficacy increases with water content until 44%, and then it decreases. 3287

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics

Surgery: Review of the Literature. Dermatol. Surg. 2012, 38 (6), 937−946. (2) Salazar, L. G.; Lu, H.; Reichow, J. L.; Childs, J. S.; Coveler, A. L.; Higgins, D. M.; Waisman, J.; Allison, K. H.; Dang, Y.; Disis, M. L. Topical Imiquimod Plus Nab-Paclitaxel for Breast Cancer Cutaneous Metastases: a Phase 2 Clinical Trial. JAMA Oncol 2017, 3, 969. (3) Berman, B.; Maderal, A.; Raphael, B. Keloids and Hypertrophic Scars: Pathophysiology, Classification, and Treatment. Dermatol. Surg. 2017, 43 (Suppl 1), S3−S18. (4) Johnston, D.; Bystryn, J. C. Topical Imiquimod Is a Potent Adjuvant to a Weakly-Immunogenic Protein Prototype Vaccine. Vaccine 2006, 24 (11), 1958−1965. (5) Stein, P.; Gogoll, K.; Tenzer, S.; Schild, H.; Stevanovic, S.; Langguth, P.; Radsak, M. P. Efficacy of Imiquimod-Based Transcutaneous Immunization Using a Nano-Dispersed Emulsion Gel Formulation. PLoS One 2014, 9 (7), e102664. (6) Harrison, L. I.; Stoesz, J. D.; Battiste, J. L.; Nelson, R. J.; Zarraga, I. E. A Pharmaceutical Comparison of Different Commercially Available Imiquimod 5% Cream Products. J. Dermatol. Treat. 2009, 20 (3), 160−164. (7) Santos, P.; Watkinson, A. C.; Hadgraft, J.; Lane, M. E. Application of Microemulsions in Dermal and Transdermal Drug Delivery. Skin Pharmacol Physiol 2008, 21 (5), 246−259. (8) Lopes, L. B. Overcoming the Cutaneous Barrier with Microemulsions. Pharmaceutics 2014, 6 (1), 52−77. (9) Dragicevic, N.; Maibach, H. I. Percutaneous Penetration Enhancers Drug Penetration Into/Through the Skin: Methodology and General Considerations.; Springer-Verlag: Berlin Heidelberg, 2017. (10) Surber, C.; Smith, E. W. The Mystical Effects of Dermatological Vehicles. Dermatology (Basel, Switz.) 2005, 210 (2), 157−168. (11) Gilbert, L.; Savary, G.; Grisel, M.; Picard, C. Predicting Sensory Texture Properties of Cosmetic Emulsions by Physical Measurements. Chemom. Intell. Lab. Syst. 2013, 124, 21−31. (12) Chen, H.; Chang, X.; Du, D.; Li, J.; Xu, H.; Yang, X. Microemulsion-Based Hydrogel Formulation of Ibuprofen for Topical Delivery. Int. J. Pharm. 2006, 315 (1−2), 52−58. (13) Sabale, V.; Vora, S. Formulation and Evaluation of Microemulsion-Based Hydrogel for Topical Delivery. Int. J. Pharm. Investig 2012, 2 (3), 140−149. (14) Zhao, L.; Wang, Y.; Zhai, Y.; Wang, Z.; Liu, J.; Zhai, G. Ropivacaine Loaded Microemulsion and Microemulsion-Based Gel for Transdermal Delivery: Preparation, Optimization, and Evaluation. Int. J. Pharm. 2014, 477 (1−2), 47−56. (15) Bachhav, Y. G.; Patravale, V. B. Microemulsion Based Vaginal Gel of Fluconazole: Formulation, in Vitro and in Vivo Evaluation. Int. J. Pharm. 2009, 365 (1−2), 175−179. (16) Rozman, B.; Zvonar, A.; Falson, F.; Gasperlin, M. TemperatureSensitive Microemulsion Gel: an Effective Topical Delivery System for Simultaneous Delivery of Vitamins C and E. AAPS PharmSciTech 2009, 10 (1), 54−61. (17) Peltola, S.; Saarinen-Savolainen, P.; Kiesvaara, J.; Suhonen, T. M.; Urtti, A. Microemulsions for Topical Delivery of Estradiol. Int. J. Pharm. 2003, 254 (2), 99−107. (18) Huang, Y. B.; Lin, Y. H.; Lu, T. M.; Wang, R. J.; Tsai, Y.-H.; Wu, P. C. Transdermal Delivery of Capsaicin Derivative-Sodium Nonivamide Acetate Using Microemulsions as Vehicles. Int. J. Pharm. 2008, 349 (1−2), 206−211. (19) Ganem-Quintanar, A.; Quintanar-Guerrero, D.; Buri, P. Monoolein: a Review of the Pharmaceutical Applications. Drug Dev. Ind. Pharm. 2000, 26 (8), 809−820. (20) Hosmer, J. M.; Steiner, A. A.; Lopes, L. B. Lamellar Liquid Crystalline Phases for Cutaneous Delivery of Paclitaxel: Impact of the Monoglyceride. Pharm. Res. 2013, 30 (3), 694−706. (21) Milak, S.; Zimmer, A. Glycerol Monooleate Liquid Crystalline Phases Used in Drug Delivery Systems. Int. J. Pharm. 2015, 478 (2), 569−587. (22) Hosmer, J. M.; Shin, S. H.; Nornoo, A.; Zheng, H.; Lopes, L. B. Influence of Internal Structure and Composition of Liquid Crystalline

accumulation in the skin (Figure 5a) and a high dermal localization (Figure 5c). TPGS and Transcutol based microemulsions were demonstrated to be very interesting vehicles for imiquimod, not only for the higher efficiency of drug transport compared to traditional formulations (the transfer efficiency of Immunocare is 0.02%), but also for the possibility to tune drug distribution into the different skin layers. Indeed, this possibility can be very useful for imiquimod, a drug shown to be effective for the treatment of several skin diseases, that however have different localization within the structures of the skin. In fact, for instance, in the case of actinic keratosis an epidermal targeting is desired, while the treatment of more invasive diseases, such as skin carcinomas, could profitably take advantage of a deeper accumulation. A tailored microemulsion formulation could allow for delivery of the drug in the skin layers affected by the different disease with potential benefits in terms of treatment success and reduction of undesired side effects. For instance, using ME 40 it would be possible to obtain elevated epidermal levels without the risk of a significant systemic absorption.



CONCLUSION The use of TPGS and Transcutol has allowed the formulation of microemulsions effective in enhancing imiquimod delivery to the skin and characterized by a viscosity suitable for topical application. X-ray scattering techniques permitted us to explain the rheological characteristics and the imiquimod uptake and distribution into the skin layers based on microemulsion mesostructure. These formulations represent promising vehicles for imiquimod targeted skin delivery, i.e. a strategy to personalize and optimize the treatment of several cutaneous diseases.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.molpharmaceut.7b00348. Rheological properties of microemulsions; SAXS intensity profiles versus q at 33 °C (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Sara Nicoli: 0000-0001-6955-0957 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors want to thank BASF for providing Kolliphor® TPGS., P. Cavallini and Macello Annoni S.p.A. (Busseto, Parma, Italy) for kindly providing porcine tissues, and Massimiliano Rinaldi for support in rheological measurement. The authors are also grateful to ID02 beamline staff and to T. Narayanan at the European Synchrotron Radiation Facility (Grenoble, France) for technical assistance.



REFERENCES

(1) Ellis, L. Z.; Cohen, J. L.; High, W.; Stewart, L. Melanoma in Situ Treated Successfully Using Imiquimod After Nonclearance with 3288

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289

Article

Molecular Pharmaceutics Phases on Topical Delivery of Paclitaxel. J. Pharm. Sci. 2011, 100 (4), 1444−1455. (23) Pham, C. V.; Cho, C.-W. Application of D-A-Tocopheryl Polyethylene Glycol 1000 Succinate (TPGS) in Transdermal and Topical Drug Delivery Systems (TDDS). J. Pharm. Invest. 2017, 47 (2), 111−121. (24) Guo, Y.; Luo, J.; Tan, S.; Otieno, B. O.; Zhang, Z. The Applications of Vitamin E TPGS in Drug Delivery. Eur. J. Pharm. Sci. 2013, 49 (2), 175−186. (25) Ke, W.-T.; Lin, S.-Y.; Ho, H.-O.; Sheu, M.-T. Physical Characterizations of Microemulsion Systems Using Tocopheryl Polyethylene Glycol 1000 Succinate (TPGS) as a Surfactant for the Oral Delivery of Protein Drugs. J. Controlled Release 2005, 102 (2), 489−507. (26) Carvalho, V. F.; de Lemos, D. P.; Vieira, C. S.; Migotto, A.; Lopes, L. B. Potential of Non-Aqueous Microemulsions to Improve the Delivery of Lipophilic Drugs to the Skin. AAPS PharmSciTech 2017, 18, 1739. (27) Suppasansatorn, P.; Nimmannit, U.; Conway, B. R.; Du, L.; Wang, Y. Microemulsions as Topical Delivery Vehicles for the AntiMelanoma Prodrug, Temozolomide Hexyl Ester (TMZA-HE). J. Pharm. Pharmacol. 2007, 59 (6), 787−794. (28) Vicentini, F. T. M. C.; Casagrande, R.; Verri, W. A.; Georgetti, S. R.; Bentley, M. V. L. B.; Fonseca, M. J. V. Quercetin in Lyotropic Liquid Crystalline Formulations: Physical, Chemical and Functional Stability. AAPS PharmSciTech 2008, 9 (2), 591−596. (29) Borgheti-Cardoso, L. N.; Vicentini, F. T. M.; de, C.; Gratieri, T.; Bentley, M. V. L. B. Liquid Crystalline Systems Containing Vitamin E TPGS for the Controlled Transdermal Nicotine Delivery. Braz. J. Pharm. Sci. 2016, 52 (1), 191−200. (30) Goddeeris, C.; Goderis, B.; Van den Mooter, G. Lyotropic, Liquid Crystalline Nanostructures of Aqueous Dilutions of SMEDDS Revealed by Small-Angle X-Ray Scattering: Impact on Solubility and Drug Release. Eur. J. Pharm. Sci. 2010, 40 (2), 110−117. (31) Telò, I.; Pescina, S.; Padula, C.; Santi, P.; Nicoli, S. Mechanisms of Imiquimod Skin Penetration. Int. J. Pharm. 2016, 511 (1), 516−523. (32) Marianecci, C.; Di Marzio, L.; Del Favero, E.; Cantù, L.; Brocca, P.; Rondelli, V.; Rinaldi, F.; Dini, L.; Serra, A.; Decuzzi, P.; Celia, C.; Paolino, D.; Fresta, M.; Carafa, M. Niosomes as Drug Nanovectors: Multiscale pH-Dependent Structural Response. Langmuir 2016, 32 (5), 1241−1249. (33) Sandri, G.; Motta, S.; Bonferoni, M. C.; Brocca, P.; Rossi, S.; Ferrari, F.; Rondelli, V.; Cantù, L.; Caramella, C.; Del Favero, E. Chitosan-Coupled Solid Lipid Nanoparticles: Tuning Nanostructure and Mucoadhesion. Eur. J. Pharm. Biopharm. 2017, 110, 13−18. (34) Clementino, A.; Batger, M.; Garrastazu, G.; Pozzoli, M.; Del Favero, E.; Rondelli, V.; Gutfilen, B.; Barboza, T.; Sukkar, M. B.; Souza, S. A. L.; Cantù, L.; Sonvico, F. The Nasal Delivery of Nanoencapsulated Statins - an Approach for Brain Delivery. Int. J. Nanomed. 2016, 11, 6575−6590. (35) Gerelli, Y.; Di Bari, M. T.; Deriu, A.; Cantù, L.; Colombo, P.; Como, C.; Motta, S.; Sonvico, F.; May, R. Structure and Organization of Phospholipid/Polysaccharide Nanoparticles. J. Phys.: Condens. Matter 2008, 20 (10), 104211. (36) Reintjes, T. Solubility Enhancement with BASF Pharma Polymers: Solubilizer Compendium; BASF, 2011. (37) Kogan, A.; Shalev, D. E.; Raviv, U.; Aserin, A.; Garti, N. Formation and Characterization of Ordered Bicontinuous Microemulsions. J. Phys. Chem. B 2009, 113 (31), 10669−10678. (38) Duvail, M.; Dufrêche, J.-F.; Arleth, L.; Zemb, T. Mesoscopic Modelling of Frustration in Microemulsions. Phys. Chem. Chem. Phys. 2013, 15 (19), 7133−7141. (39) Warriner, H. E.; Davidson, P.; Slack, N. L. Lamellar Biogels Comprising Fluid Membranes with a Newly Synthesized Class of Polyethylene Glycol-Surfactants. J. Chem. Phys. 1997, 107, 3707. (40) Wang, X.; Quinn, P. J. Phase Separations of Alpha-Tocopherol in Aqueous Dispersions of Distearoylphosphatidylethanolamine. Chem. Phys. Lipids 2002, 114 (1), 1−9.

(41) Donnelly, R. F.; McCarron, P. A.; Zawislak, A. A.; Woolfson, A. D. Design and Physicochemical Characterisation of a Bioadhesive Patch for Dose-Controlled Topical Delivery of Imiquimod. Int. J. Pharm. 2006, 307 (2), 318−325. (42) Karakatsani, M.; Dedhiya, M.; Plakogiannis, F. M. The Effect of Permeation Enhancers on the Viscosity and the Release Profile of Transdermal Hydroxypropyl Methylcellulose Gel Formulations Containing Diltiazem HCl. Drug Dev. Ind. Pharm. 2010, 36 (10), 1195−1206. (43) Dragicevic-Curic, N.; Winter, S.; Stupar, M.; Milic, J.; Krajisnik, D.; Gitter, B.; Fahr, A. Temoporfin-Loaded Liposomal Gels: Viscoelastic Properties and in Vitro Skin Penetration. Int. J. Pharm. 2009, 373 (1−2), 77−84. (44) Mahrhauser, D. S.; Kahlig, H.; Partyka-Jankowska, E.; Peterlik, H.; Binder, L.; Kwizda, K.; Valenta, C. Investigation of Microemulsion Microstructure and Its Impact on Skin Delivery of Flufenamic Acid. Int. J. Pharm. 2015, 490 (1−2), 292−297. (45) Mahrhauser, D.; Hoppel, M.; Scholl, J.; Binder, L.; Kahlig, H.; Valenta, C. Simultaneous Analysis of Skin Penetration of Surfactant and Active Drug From Fluorosurfactant-Based Microemulsions. Eur. J. Pharm. Biopharm. 2014, 88 (1), 34−39. (46) Hathout, R. M.; Mansour, S.; Mortada, N. D.; Geneidi, A. S.; Guy, R. H. Uptake of Microemulsion Components Into the Stratum Corneum and Their Molecular Effects on Skin Barrier Function. Mol. Pharmaceutics 2010, 7 (4), 1266−1273. (47) Advanced Chemistry Development Inc. ACD/Percepta, Version 2015, Pack 2; Toronto, On, Canada, http://www.acdlabs.com/ products/percepta/, 2015. (48) Bucks, D. A.; McMaster, J. R.; Maibach, H. I.; Guy, R. H. Bioavailability of Topically Administered Steroids: a “Mass Balance” Technique. J. Invest. Dermatol. 1988, 91 (1), 29−33.

3289

DOI: 10.1021/acs.molpharmaceut.7b00348 Mol. Pharmaceutics 2017, 14, 3281−3289