Confocal Raman microscopy and multivariate statistical analysis for determination of different penetration abilities of caffeine and propylene glycol applied simultaneously in a mixture on porcine skin ex vivo

European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb

Research paper

Confocal Raman microscopy and multivariate statistical analysis for determination of different penetration abilities of caffeine and propylene glycol applied simultaneously in a mixture on porcine skin ex vivo Saul Mujica Ascencio a,b, ChunSik Choe a,c, Martina C. Meinke a, Rainer H. Müller d, George V. Maksimov e, Walter Wigger-Alberti f, Juergen Lademann a, Maxim E. Darvin a,⇑ a Center of Experimental and Applied Cutaneous Physiology, Department of Dermatology, Venerology and Allergology, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany b Centro de Investigación e Innovación Tecnológica (CIITEC) del Instituto Politécnico Nacional (IPN), Cerrada de Cecati S/N, Col. Santa Catarina Azcapotzalco, México D.F. CP: 02250, Mexico c Kim Il Sung University, Ryongnam-Dong, Taesong District, Pyongyang, Democratic People’s Republic of Korea d Institute of Pharmacy, Department of Pharmaceutics, Biopharmaceutics & NutriCosmetics, Freie Universität Berlin, Kelchstraße 31, 12169 Berlin, Germany e M.V. Lomonosov Moscow State University, Department of Biophysics, Faculty of Biology, Leninskie Gory, 1-12, 119991 Moscow, Russia f Bioskin GmbH, Burchardstrasse 17, 20095 Hamburg, Germany

a r t i c l e

i n f o

Article history: Received 24 September 2015 Revised 5 January 2016 Accepted in revised form 20 April 2016 Available online 21 April 2016 Keywords: Stratum corneum Stratum spinosum Epidermis Toxicity Penetration enhancer Multispeed penetration Propanediol Vibrational spectroscopy

a b s t r a c t Propylene glycol is one of the known substances added in cosmetic formulations as a penetration enhancer. Recently, nanocrystals have been employed also to increase the skin penetration of active components. Caffeine is a component with many applications and its penetration into the epidermis is controversially discussed in the literature. In the present study, the penetration ability of two components – caffeine nanocrystals and propylene glycol, applied topically on porcine ear skin in the form of a gel, was investigated ex vivo using two confocal Raman microscopes operated at different excitation wavelengths (785 nm and 633 nm). Several depth profiles were acquired in the fingerprint region and different spectral ranges, i.e., 526–600 cm 1 and 810–880 cm 1 were chosen for independent analysis of caffeine and propylene glycol penetration into the skin, respectively. Multivariate statistical methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) combined with Student’s t-test were employed to calculate the maximum penetration depths of each substance (caffeine and propylene glycol). The results show that propylene glycol penetrates significantly deeper than caffeine (20.7–22.0 lm versus 12.3–13.0 lm) without any penetration enhancement effect on caffeine. The results confirm that different substances, even if applied onto the skin as a mixture, can penetrate differently. The penetration depths of caffeine and propylene glycol obtained using two different confocal Raman microscopes are comparable showing that both types of microscopes are well suited for such investigations and that multivariate statistical PCA–LDA methods combined with Student’s t-test are very useful for analyzing the penetration of different substances into the skin. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction The skin is the largest organ of the body and its principal function is to provide a protection against the external environment. The superficial layer of the skin – the stratum corneum – is mostly responsible for this providing the efficient barrier function [1–4]. Cosmetic formulations are important in daily life for skin ⇑ Corresponding author. E-mail address: [email protected] (M.E. Darvin). http://dx.doi.org/10.1016/j.ejpb.2016.04.018 0939-6411/Ó 2016 Elsevier B.V. All rights reserved.

moisturizing as well as in dermatology for treatment of skin diseases. Some formulations applied should saturate only the superficial layer of the stratum corneum, for example sunscreens [5–7], whereas some formulations should permeate the barrier of the stratum corneum to contact the viable cells of the stratum spinosum, for example medical ointments [7–10]. Different penetration pathways, such as intracellular, intercellular and follicular could be mentioned in this regard [11–13]. Penetration enhancers can be used in dermal, cosmetic and pharmaceutical formulations to increase skin penetration [14]. They are able to destroy or influence

52

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

the lipids, which provide the barrier function of stratum corneum, thus enabling substances to penetrate through the cutaneous barrier [1,12,15]. By acting like this, penetration enhancers distort or impair the lipid skin barrier. Thus less invasive delivery approaches should be preferred, e.g. transport systems like the liposomes, or the nanocrystals used in this study. Some penetration enhancers used in scientific investigations are effective, but are still lacking regulatory approval and therefore cannot be used in products for the patients. A relatively well tolerated compound that is approved for dermal use is propylene glycol. Propylene glycol is an organic alcohol, often used as a penetration enhancer [16,17] and also employed as an optical clearing agent [18], humectant, viscosity decreasing agent, solvent and fragrance ingredient [19–21]. Nanocrystals are a novel concept employed in cosmetic and pharmaceutical formulations to enhance the penetration of poorly soluble actives [22]. Topical application of nanocrystals (particles of the pure active component) saturates the skin surface and the nanocrystals being adhered to the skin maintain a constant release of the active component, providing the penetration through the skin [23]. At the beginning, nanocrystals were produced only from poor soluble actives [24–26] but recently they were also made from medium soluble actives such as caffeine [27,28]. Caffeine is a molecule with many applications and it has been added into cosmetic formulations to stimulate the hair growth [29], to inhibit the UVB sunlight [30], to protect fibroblasts [31,32], to promote wound healing [33] and to treat cellulite [34,35]. These actions can be derived from the capability of caffeine to penetrate through the skin barrier [36] and its antioxidant and protective effects [37– 39]. The follicular penetration pathway of caffeine through the skin barrier was established to be dominant [40,41]. However, the intracellular and intercellular penetration ability of caffeine through the stratum corneum is discussed controversially [14,42]. For investigating the penetration of components of a formulation into the skin, the minimally invasive tape stripping procedure is widely used [43–45] as an alternative to biopsy analysis. The non-invasive methods are very promising in this regard. Among different non-invasive methods [46], more and more investigations are performed using confocal Raman microscopy to investigate penetration of different substances into the skin [47–54] including caffeine [55,56]. Penetration of oils, sunscreens and their components into the skin was shown to be very poor. None of these substances exceeded the depth of the first corneocytes of stratum corneum [57–60]. Therefore, the caffeine and propylene glycol were chosen for investigations as they are known to be able to penetrate much deeper into the skin. Only a few studies on caffeine and propylene glycol permeation employing Raman spectroscopy have been published, so far. Bonnist et al. [61] employed confocal Raman spectroscopy to investigate the penetration of chemicals into porcine skin using the propylene glycol as vehicle substance. Their results show that propylene glycol facilitates slow penetration of the sensitizer under study. Most of the studies about penetration employing confocal Raman microscopy are related to the quantification of caffeine at different depths without determination of the maximum penetration depth [62–64]. In an extensive literature review, it was not possible to find studies that employ statistical methods to follow up the penetration depth of caffeine and propylene glycol into the skin. Most of the works that combined confocal Raman microscopy and statistical methods were focused on the detection of small differences between healthy and cancer cells [65–68]. The main idea of this study was to acquire Raman depth profiles of a gel formulation which contains caffeine nanocrystals and propylene glycol and to employ statistical methods determining their maximum penetration depths into porcine skin ex vivo and to detect whether these methods can differentiate the penetration

depths of both gel components. The Raman depth profiles were obtained using two confocal Raman microscopes operated at different excitation wavelengths (633 nm and 785 nm) for ex vivo and in vivo measurements, and the spectra were analyzed with multivariate statistical methods such as principle components analysis (PCA) and linear discriminant analysis (LDA) combined with Student’s t-test. 2. Materials and methods 2.1. Skin samples Twelve fresh porcine ears obtained from a local abattoir were gently cleaned with cold flowing water and dried with soft tissue. The hairs were carefully removed without damaging the stratum corneum. The skin samples without any scars were cut manually with a scalpel and prepared for gel application. Skin samples from eight porcine ears were treated with the gel containing two main components – caffeine and propylene glycol. Skin samples from four porcine ears were left untreated. The porcine skin was chosen for the measurements due to its morphological similarity with human skin and appropriateness for ex vivo measurements [69,70]. Moreover, the Raman spectra of porcine and human skin stratum corneum are highly similar [54,71] making possible the application of the below-described algorithms for human skin measurements. 2.2. Applied substances A gel containing 30 g of caffeine nanocrystals with a particle size of 650 nm (determined by photon correlation spectroscopy), 105 g of 1,2-propylene glycol, 3 g of carbopol 981 and 45 g of ethanol was applied onto the skin surface at an amount of 2 mg/cm2. After one hour of passive penetration of the gel at normal conditions (room temperature 21 °C, relative humidity 50% in average), the remaining gel was taken away by filter paper. Subsequently, the treated skin sample was subjected to confocal Raman microscopy (CRM). 2.3. Production of caffeine nanocrystals Caffeine nanosuspension was produced by bead milling as described previously [28]. A bead mill Bühler PML 2 (Bühler, Switzerland) equipped with a small milling chamber was used. A suspension of caffeine in water, ethanol, and 1,2-propylene glycol mixture containing carbopol 981 as stabilizer for the crystals was prepared. This suspension was milled in the chamber, using power beads type YSZ with a size of 0.4–0.6 mm, operating condition 2000 rpm of the agitator at 5 °C production temperature. The size decrease was monitored as a function of time, and the production terminated when the desired target size was reached. Size measurements of the nanocrystals were performed by photon correlation spectroscopy (PCS) using a Zetasizer ZS Nano (Malvern Instruments, UK). Short term stability conducted at a storage temperature of 4 °C showed stability over 30 days, no significant change in PCS size, and the absence of aggregates. 2.4. Instrumentation Confocal Raman microscopic measurements were made in the fingerprint (400–2000 cm 1) and high wavenumber regions (2000–4000 cm 1) employing two differently designed confocal Raman microscopes (CRM). The first system was a CRM ‘‘Horiba” (model Labram HR800 Evo, France). The following settings were used: excitation wavelength of 633 nm, grating of 600 g/mm, an objective of 100, spot of 0.85 lm, laser power of 12.6 mW on

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

the sample and exposure time of 1 s. The second one was a CRM ‘‘River Diagnostics” (model 3510 SCA, Rotterdam, Netherlands). The following settings were used: excitation wavelength of 785 nm for fingerprint region and 671 nm for high wavenumber region, oil objective of 50 with spot of 5 lm, laser power of 20 mW on the sample and an exposure time of 5 s. The utilized doses of reference light are not sufficient to influence the skin components [72]. The utilized CRM system was previously described in the literature [73,74]. Every Raman depth profile was measured between 10 lm and 30 lm at increments of 2 lm. The first measuring spectrum was taken at 10 lm (outside the skin) and at the last measuring spectrum at 30 lm (inside the skin). The real skin surface position (point 0 lm) has been evaluated for each measured depth profile, as described below. The Raman data were pre-processed using MATLAB software to reduce the spectral noise and subtract the fluorescence background. The Raman spectra were smoothed using low order Savitzky-Golay filter (polynomial order of 2 and frame size of 3), as it is known algorithm for reduction in spectral noise without altering the intensity and position of Raman peaks. A baseline correction function was employed to subtract the fluorescence background. The skin areas without furrows and hairs were chosen to exclude their influence on the penetration measurements. 2.5. Study design Six gel-treated and three untreated porcine ear samples were measured using CRM ‘‘Horiba”; two gel-treated and two untreated porcine ear samples were measured using CRM ‘‘River D”. Every gel-treated skin sample was measured at 10 different points, while every untreated skin sample was measured at 4 different points. To determine the maximum penetration depth of each gel component (caffeine and propylene glycol) at each measuring point, three general steps were considered. The first step consisted of a comparison between the gel formulation and porcine skin spectra in order to determine the spectral ranges to be analyzed. In the second step the skin surface position in every Raman depth profile was to be determined. Finally, the last step compared the untreated skin with the gel-treated skin using multivariate statistical methods to determine the maximum penetration depth value of caffeine and propylene glycol. Principal component analysis and linear discriminate analysis combined with Student’s t-test (paired-sample t-test in MATLAB) were performed to analyze each porcine ear skin sample individually. The stratum corneum thickness was determined by the water gradient value of 0.5 analyzing the high wavenumber range 3350–3550 cm 1 according to Egawa et al. [75].

53

[82,83]. In addition, LDA reduces the dimension of the data to one dimension (linear discriminant 1 or LD1). The differences between groups of data can be seen by plotting the individual scores. The success of the method is based on the overlapping of the scores [83,84]. In our measurements, the groups of data are the untreated and gel-treated porcine skin samples, the variable to classify the groups is the intensity of caffeine and propylene glycol Raman peaks after PCA procedure and the scores are the total number of measuring points considered in a porcine ear. The overlapping of the measuring points will be delimited using a further statistic method called Student’s t-test. Student’s t-test is a method, which compares the characteristics of two groups of data. Those characteristics are related to the similitudes or differences between groups. In this study, the mean value of each group is the characteristics to be compared. It is expected that the spectra of untreated and gel-treated porcine skin have the same mean value at certain depths when the gel is not detected. In such case, the previous depth will be the maximal penetration depth of caffeine and propylene glycol. 3. Results and discussion 3.1. Measurement spectral range evaluation For successful detection of the exogenous substances in the skin, their own Raman peaks should not be overlapped with Raman peaks of the skin and should have the high-enough intensity. Fig. 1 represents the fingerprint Raman spectra of porcine ear skin, caffeine and propylene glycol gel components. The peaks at 555 cm 1 (CON deformation mode d C@OAN [85]) and 840 cm 1 ((CAO)H stretching mode [86]) serve as a marker peaks for caffeine and propylene glycol respectively because they are highly intense and do not overlap with porcine skin Raman peaks. Other prominent peaks at 1329 cm 1 and 1461 cm 1 of caffeine and propylene glycol are not recommended to be employed due to their overlapping with the Raman spectrum of porcine skin. For the analysis, the spectral ranges of 526–600 cm 1 and 810–880 cm 1 for caffeine and propylene glycol are selected because they cover the maximum width of each peak. 3.2. Skin surface determination The skin surface position should be precisely determined for all measured depth profiles. The skin surface position could be

2.6. Multivariate statistical methods for Raman spectra analysis Principal component analysis (PCA) is employed to find some tendencies between two groups of spectra [76] and to reduce the noise [77,78]. To reduce the noise of the spectra, the first two principle components (PC), i.e., PC1 and PC2, are selected to reconstruct the Raman spectra and highlight the peaks of caffeine and propylene glycol. However, in depth-located skin areas the Raman spectra of untreated and gel-treated porcine skin are comparable and PCA cannot determine the differences between two groups of data [79]. To avoid this limitation and determine the differences at high sensitivity, PCA was complemented together with a method called linear discriminant analysis (LDA) [80,81]. LDA is considered as a group classification as it maximizes the differences between group variances in respect of a variable

Fig. 1. Fingerprint Raman spectra of porcine skin at depth of 2 lm (fat solid line), caffeine (solid line) and propylene glycol (dotted line) gel components measured using CRM ‘‘Horiba” (excitation 633 nm).

54

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

selected considering different Raman peaks of its spectrum. Sometimes, the selection of a peak is difficult because of overlapping of the skin and the Raman peaks of the applied substance. Specialized software can be employed to determine the skin surface position automatically but not all CRMs are provided with such a software. Only the CRM ‘‘River Diagnostics” is equipped with ‘‘SkinTools” software, which considers the keratin peak in the fingerprint range (1600–1700 cm 1) for determination of the skin surface position. Another keratin peak (2900–3000 cm 1) is also used to determine the skin surface position in the high wavenumber region [56,87]. In this study it is not suitable to apply this procedure since the caffeine exhibits own Raman peaks at 1600, 1655 and 1697 cm 1, which are overlapped with the keratin band between 1600 and 1700 cm 1 (see Fig. 1). Other authors consider the phenylalanine/urea peak at 1003 cm 1 in the maximum intensity [88]. In this study, the phenylalanine/urea peak that is typical in skin, serves as a good option to determine the skin surface position at every Raman depth profile acquired with both CRMs because it is not overlapped with the Raman peaks of caffeine and propylene glycol (see Fig. 1). The procedure follows the main idea of ‘‘SkinTools” but an adjustment is applied in accordance with the information acquired. The maximum width of phenylalanine/urea peak was selected from the complete Raman depth profile. The range between 998 and 1008 cm 1 was evaluated to be optimal because it contains the full shape of the peak at every depth (see Fig. 2a). Then, the AUC (area under the curve) was calculated at every depth and plotted as shown in Fig. 2b. The skin surface position was determined as the half value between the minimum and maximum AUC intensities. 3.3. Multivariate statistical analysis Finally, the maximum penetration depth of caffeine and propylene glycol was determined. The maximum penetration depth is defined as the minimum peak intensity that the methods can differentiate when the treated and untreated porcine skin spectra are compared in the same depth and spectral range. For this reason, PCA–LDA multivariate statistical analysis was performed. For representation of the application of multivariate statistical analyses based on the PCA–LDA method, the Raman peak of caffeine centered at 555 cm 1 is considered. The procedure is similar for propylene glycol, except for the range of investigation, which is 810–880 cm 1. Fig. 3 shows the group of Raman spectra in the range between 526 and 600 cm 1 of the untreated (solid line) and gel-treated (fat solid line) porcine skin after reconstruction with PCA (PC1 and PC2) for three exemplary depths. The intensity of the caffeine peak is reduced by increasing the depth into the skin. The difference between untreated and gel-treated skin spectra is visually distinguishable in the depths of 2 lm (Fig. 3a) and 10 lm (Fig. 3b). In the depth of 16 lm (Fig. 3c) the spectra are comparable making impossible to detect caffeine in the skin. Then, the groups of untreated and gel-treated porcine skin samples are generated by PCA–LDA. Their respective individual intensities are plotted in order to visualize the differences between the mean intensity of both groups. Fig. 4 represents the box plots of the three exemplary depths showing the differences between untreated porcine skin (squares) and gel-treated porcine skin (triangles) spectra after the PCA–LDA analyses. The dotted line helps to visualize the difference between groups. The depth of 2 lm (Fig. 4a) is a good example that the gel component (caffeine) is present in the skin. By increasing the depth, the differences between untreated and gel-treated skin samples are reduced (Fig. 4b) and disappear at some depth (Figs. 4c) showing the absence of penetration.

Fig. 2. Phenylalanine/urea Raman peak centered at 1003 cm 1 measured using CRM ‘‘Horiba” (excitation 633 nm) in porcine skin at depths from 7.5 lm to 32.5 lm at 2 lm increments (a) and the subsequent depth-dependent AUC determined in the range between 998 and 1008 cm 1 (b). Skin surface position (0 lm) is determined at the half value between the minimum and maximum intensities. The depth-dependent AUC of phenylalanine/urea Raman peak is not normalized to keratin Raman peak intensity.

To distinguish between untreated and gel-treated groups, mathematical criteria should be employed. The Student’s t-test method works efficiently selecting the appropriate null hypothesis [89]. The standard value of the null hypothesis is considered as p < 0.05 [90,91]. Transferring the data of the PCA–LDA method to the Student’s t-test method, the mean intensity and variance of untreated and gel-treated groups are shown in Fig. 4. The same procedure for the Raman peak of propylene glycol centered at 840 cm 1 was applied to find the maximal penetration depth (data not shown). Table 1 shows the maximum penetration depth (mean ± SD) of caffeine and propylene glycol obtained after the application of PCA–LDA statistical methods combined with Student’s t-test with a significant criteria of p < 0.05 using two CRMs. Now, comparing the penetration depth values for both substances and considering that samples were analyzed under the same conditions, it is clearly visible that propylene glycol penetrates significantly deeper into the porcine skin than caffeine. This means that both CRMs combined with statistical methods such as PCA, LDA and Student’s ttest can be employed to determine the penetration depth values of different substances in a gel formulation independently.

55

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

Fig. 3. Untreated (fat solid lines) and gel-treated (solid lines) porcine skin Raman spectra in the range between 526 cm 1 and 600 cm 1 after spectra reconstruction with PC1 and PC2 in the depths of 2 lm (a), 10 lm (b) and 16 lm (c).

3.4. Detection limit The detection limit of caffeine and propylene glycol is obtained by calculating their concentrations at the corresponding maximum penetration depths. Knowing the exact composition of the gel and its initial concentration applied on the skin, the concentration of caffeine and propylene glycol at the different depths can be estimated. The procedure consists in obtaining the depth profiles for caffeine and propylene glycol in the spectral range of 526– 600 cm 1 and 810–880 cm 1, respectively, by calculating the AUCs at every depth (exponential depth-dependent attenuation). Postulating that the maximum AUC value is considered to be an initial concentration (2 mg/cm2), the detection limit was calculated as minimum detected AUC measured at the maximum penetration depth. The results are summarized in Table 2.

Fig. 4. Differences between untreated (squares) and gel-treated (triangles) porcine skin Raman spectra after the PCA–LDA analyses employing Student’s t-test at the depths of 2 lm (a), 10 lm (b) and 16 lm (c).

Table 1 Maximum penetration depth of caffeine and propylene glycol with a significance value p < 0.05 employing both CRMs. Substance Caffeine Propylene glycol

Range, cm 1

Penetration depth, lm CRM ‘‘Horiba”

lm CRM ‘‘River D”

526–600 810–880

12.3 ± 3.0 20.7 ± 3.0

13.0 ± 1.4 22.0 ± 0.2

Penetration depth,

4. Conclusions In this study, two CRMs operated at different excitation wavelengths of 633 nm and 785 nm were employed to measure the penetration ability of gel containing two main components – caffeine and propylene glycol into the porcine ear skin ex vivo. For analyses of the penetration profiles of caffeine and propylene glycol into the

56

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

Table 2 Detection limit of caffeine and propylene glycol measured using two different CRMs. Substance

Range, cm 1

Concentration, mg/cm2 CRM ‘‘Horiba”

Concentration, mg/cm2 CRM ‘‘River D”

Caffeine Propylene glycol

526–600 810–880

0.09 0.15

0.07 0.17

skin, PCA–LDA multivariate statistical methods combined with Student’s t-test were applied. The individual penetration profiles of each component applied simultaneously in the form of gel formulation were determined. It was found that propylene glycol penetrates into the skin up to a depth of 20.7–22.0 lm, while caffeine penetrates up to a depth of 12.3–13.0 lm in average obtained with two CRMs. The obtained results are in accordance with the previously published data for caffeine [55] and propylene glycol [92]. The obtained difference between caffeine and propylene glycol penetration depths was found to be statistically significant. Taking into consideration the thickness of the stratum corneum, which was determined to be 18.1 ± 1.0 lm, it can be concluded that caffeine does not penetrate efficiently through the stratum corneum saturating the upper 70–80%, while propylene glycol easily penetrates through the stratum corneum reaching the stratum spinosum layer, i.e., the living cells of the epidermis. Independent on the penetration enhancement properties of propylene glycol this effect was not observed for caffeine nanocrystals, probably due to its size. It should be taken into consideration, however, that the follicular penetration pathway that is dominant for caffeine [40] was not investigated in the present study. The results obtained ex vivo using two CRMs are comparable showing that both types of microscopes are well suited for such investigations and that the application of the PCA–LDA multivariate statistical methods could be very useful for analyzing the penetration of different substances into the skin. Moreover, taking into consideration the higher permeability of caffeine into the skin in vivo in comparison with ex vivo [40,93], the highest penetration values in case of in vivo application could be expected. The optical clearing effect [94] induced by the applied gel could be expected, as propylene glycol (also known as 1,2-propanediol) is able to improve tissue optical transmittance [95]. This can potentially increase the measurement depth [96]. This effect was not investigated in the present study and is postulated not to influence the obtained results. Acknowledgment The study was supported by the Foundation for Skin Physiology of the Donor Association for German Science and Humanities. References [1] A.C. Williams, B.W. Barry, Penetration enhancers, Adv. Drug Deliver Rev. 64 (2012) 128–137. [2] L. Norlen, Current understanding of skin barrier morphology, Skin Pharmacol. Physiol. 26 (2013) 213–216. [3] M. Windbergs, S. Hansen, A. Schroeter, U.F. Schaefer, C.M. Lehr, J. Bouwstra, From the structure of the skin barrier and dermal formulations to in vitro transport models for skin absorption: skin research in the Netherlands and in Germany, Skin Pharmacol. Physiol. 26 (2013) 317–330. [4] M.E. Darvin, J.W. Fluhr, P. Caspers, A. van der Pool, H. Richter, A. Patzelt, W. Sterry, J. Lademann, In vivo distribution of carotenoids in different anatomical locations of human skin: comparative assessment with two different Raman spectroscopy methods, Exp. Dermatol. 18 (2009) 1060–1063. [5] T. Vergou, A. Patzelt, S. Schanzer, M.C. Meinke, H.J. Weigmann, G. Thiede, W. Sterry, J. Lademann, M.E. Darvin, Methods for the evaluation of the protective efficacy of sunscreen products, Skin Pharmacol. Physiol. 26 (2013) 30–35.

[6] M.E. Darvin, K. Konig, M. Kellner-Hoefer, H.G. Breunig, W. Werncke, M.C. Meinke, A. Patzelt, W. Sterry, J. Lademann, Safety assessment by multiphoton fluorescence/second harmonic generation/hyper-Rayleigh scattering tomography of ZnO nanoparticles used in cosmetic products, Skin Pharmacol. Physiol. 25 (2012) 219–226. [7] K. Heinrich, U. Heinrich, H. Tronnier, Influence of different cosmetic formulations on the human skin barrier, Skin Pharmacol. Physiol. 27 (2014) 141–147. [8] J. Wohlrab, R.H. Neubert, M.L. Heskamp, J. Michael, Cutaneous drug delivery of capsaicin after in vitro administration of the 8% capsaicin dermal patch system, Skin Pharmacol. Physiol. 28 (2015) 65–74. [9] N.A. Belsey, N.L. Garrett, L.R. Contreras-Rojas, A.J. Pickup-Gerlaugh, G.J. Price, J. Moger, R.H. Guy, Evaluation of drug delivery to intact and porated skin by coherent Raman scattering and fluorescence microscopies, J. Control. Release 174 (2014) 37–42. [10] Ajazuddin, A. Alexander, A. Khichariya, S. Gupta, R.J. Patel, T.K. Giri, D.K. Tripathi, Recent expansions in an emergent novel drug delivery technology: Emulgel, J. Control. Release 171 (2013) 122–132. [11] M.A. Bolzinger, S. Briancon, J. Pelletier, Y. Chevalier, Penetration of drugs through skin, a complex rate-controlling membrane, Curr. Opin. Colloid Interf. 17 (2012) 156–165. [12] M.E. Lane, Skin penetration enhancers, Int. J. Pharm. 447 (2013) 12–21. [13] J.R. Weiser, W.M. Saltzman, Controlled release for local delivery of drugs: barriers and models, J. Control. Release 190 (2014) 664–673. [14] E. Abd, M.S. Roberts, J.E. Grice, A comparison of the penetration and permeation of caffeine into and through human epidermis after application in various vesicle formulations, Skin Pharmacol. Physiol. 29 (2016) 24–30. [15] C. Puglia, A. Offerta, G.G. Tirendi, M.S. Tarico, S. Curreri, F. Bonina, R.E. Perrotta, Design of solid lipid nanoparticles for caffeine topical administration, Drug Del. (2014). [16] G.B. Kasting, Lipid solubility and molecular weight: whose idea was that, Skin Pharmacol. Physiol. 26 (2013) 295–301. [17] D. Mohammed, P.J. Matts, J. Hadgraft, M.E. Lane, In vitro–in vivo correlation in skin permeation, Pharm. Res. 31 (2014) 394–400. [18] T. Yu, X. Wen, V.V. Tuchin, Q. Luo, D. Zhu, Quantitative analysis of dehydration in porcine skin for assessing mechanism of optical clearing, J. Biomed. Opt. 16 (2011) 095002. [19] M.M. Fiume, W.F. Bergfeld, D.V. Belsito, R.A. Hill, C.D. Klaassen, D. Liebler, J.G. Marks Jr., R.C. Shank, T.J. Slaga, P.W. Snyder, F.A. Andersen, Safety assessment of propylene glycol, tripropylene glycol, and PPGs as used in cosmetics, Int. J. Toxicol. 31 (2012) 245S–260S. [20] S.H. Moghadam, E. Saliaj, S.D. Wettig, C. Dong, M.V. Ivanova, J.T. Huzil, M. Foldvari, Effect of chemical permeation enhancers on stratum corneum barrier lipid organizational structure and interferon alpha permeability, Mol. Pharm. 10 (2013) 2248–2260. [21] D. Mohammed, K. Hirata, J. Hadgraft, M.E. Lane, Influence of skin penetration enhancers on skin barrier function and skin protease activity, Eur. J. Pharm. Sci. 51 (2014) 118–122. [22] R. Petersen, Nanocrystals for use in topical cosmetic formulation and method of production threof, 2006. [23] R.H. Muller, S. Gohla, C.M. Keck, State of the art of nanocrystals–special features, production, nanotoxicology aspects and intracellular delivery, Eur. J. Pharm. Biopharm. 78 (2011) 1–9. [24] P.R. Mishra, L. Al Shaal, R.H. Muller, C.M. Keck, Production and characterization of Hesperetin nanosuspensions for dermal delivery, Int. J. Pharm. 371 (2009) 182–189. [25] S. Kobierski, K. Ofori-Kwakye, R.H. Muller, C.M. Keck, Resveratrol nanosuspensions for dermal application–production, characterization, and physical stability, Pharmazie 64 (2009) 741–747. [26] K. Mitri, R. Shegokar, S. Gohla, C. Anselmi, R.H. Muller, Lutein nanocrystals as antioxidant formulation for oral and dermal delivery, Int. J. Pharm. 420 (2011) 141–146. [27] X. Zhai, J. Lademann, C.M. Keck, R.H. Muller, Dermal nanocrystals from medium soluble actives – physical stability and stability affecting parameters, Eur. J. Pharm. Biopharm. 88 (2014) 85–91. [28] X. Zhai, J. Lademann, C.M. Keck, R.H. Muller, Nanocrystals of medium soluble actives–novel concept for improved dermal delivery and production strategy, Int. J. Pharm. 470 (2014) 141–150. [29] J. Lademann, H. Richter, S. Schanzer, A. Klenk, W. Sterry, A. Patzelt, Analysis of the penetration of a caffeine containing shampoo into the hair follicles by in vivo laser scanning microscopy, Laser Phys. 20 (2010) 551–556. [30] A.H. Conney, Y.P. Lu, Y.R. Lou, M. Kawasumi, P. Nghiem, Mechanisms of caffeine-induced inhibition of UVB carcinogenesis, Front. Oncol. 3 (2013) 144. [31] J.I. Silverberg, M. Patel, N. Brody, J. Jagdeo, Caffeine protects human skin fibroblasts from acute reactive oxygen species-induced necrosis, J. Drugs Dermatol.: JDD 11 (2012) 1342–1346. [32] M. Donejko, A. Przylipiak, E. Rysiak, K. Gluszuk, A. Surazynski, Influence of caffeine and hyaluronic acid on collagen biosynthesis in human skin fibroblasts, Drug Des., Dev. Ther. 8 (2014) 1923–1928. [33] N. Ojeh, O. Stojadinovic, I. Pastar, A. Sawaya, N. Yin, M. Tomic-Canic, The effects of caffeine on wound healing, Int. Wound J. (2014). [34] N.H.C.S. Silva, I. Drumond, I.F. Almeida, P. Costa, C.F. Rosado, C.P. Neto, C.S.R. Freire, A.J.D. Silvestre, Topical caffeine delivery using biocellulose membranes: a potential innovative system for cellulite treatment, Cellulose 21 (2014) 665– 674.

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58 [35] H. Hamishehkar, J. Shokri, S. Fallahi, A. Jahangiri, S. Ghanbarzadeh, M. Kouhsoltani, Histopathological evaluation of caffeine-loaded solid lipid nanoparticles in efficient treatment of cellulite, Drug Dev. Ind. Pharm. (2014) 1–7. [36] A. Herman, A.P. Herman, Caffeine’s mechanisms of action and its cosmetic use, Skin Pharmacol. Physiol. 26 (2013) 8–14. [37] M.M. Camouse, K.K. Hanneman, E.P. Conrad, E.D. Baron, Protective effects of tea polyphenols and caffeine, Expert Rev. Anticancer Ther. 5 (2005) 1061– 1068. [38] T. Arakawa, H. Yamasaki, K. Ikeda, D. Ejima, T. Naito, A.H. Koyama, Antiviral and virucidal activities of natural products, Curr. Med. Chem. 16 (2009) 2485– 2497. [39] J. Jagdeo, N. Brody, Complementary antioxidant function of caffeine and green tea polyphenols in normal human skin fibroblasts, J. Drugs Dermatol. 10 (2011) 753–761. [40] S. Trauer, A. Patzelt, N. Otberg, F. Knorr, C. Rozycki, G. Balizs, R. Buttemeyer, M. Linscheid, M. Liebsch, J. Lademann, Permeation of topically applied caffeine through human skin – a comparison of in vivo and in vitro data, Br. J. Clin. Pharmacol. 68 (2009) 181–186. [41] A. Teichmann, H. Richter, F. Knorr, C. Antoniou, W. Sterry, J. Lademann, Investigation of the penetration and storage of a shampoo formulation containing caffeine into the hair follicles by in vivo laser scanning microscopy, Laser Phys. Lett. 4 (2007) 464–468. [42] L. Rubio, C. Alonso, O. Lopez, G. Rodriguez, L. Coderch, J. Notario, A. de la Maza, J.L. Parra, Barrier function of intact and impaired skin: percutaneous penetration of caffeine and salicylic acid, Int. J. Dermatol. 50 (2011) 881– 889. [43] C.M. Gee, J.A. Nicolazzo, A.C. Watkinson, B.C. Finnin, Assessment of the lateral diffusion and penetration of topically applied drugs in humans using a novel concentric tape stripping design, Pharm. Res. 29 (2012) 2035–2046. [44] K. Kikuchi, H. Todo, K. Sugibayashi, Usefulness of pressure-sensitive adhesives as a pretreatment material before application of topical drug formulations and a peeling tape for excess stratum corneum layers, Chem. Pharm. Bull. 62 (2014) 559–567. [45] M. Hoppel, E. Holper, D. Baurecht, C. Valenta, Monitoring the distribution of surfactants in the stratum corneum by combined ATR-FTIR and tape-stripping experiments, Skin Pharmacol. Physiol. 28 (2015) 167–175. [46] J. Lademann, M.C. Meinke, S. Schanzer, H. Richter, M.E. Darvin, S.F. Haag, J.W. Fluhr, H.J. Weigmann, W. Sterry, A. Patzelt, In vivo methods for the analysis of the penetration of topically applied substances in and through the skin barrier, Int. J. Cosmet. Sci. 34 (2012) 551–559. [47] B. Kann, M. Windbergs, Chemical imaging of drug delivery systems with structured surfaces – a combined analytical approach of confocal Raman microscopy and optical profilometry, AAPS J. 15 (2013) 505–510. [48] M. Ashtikar, C. Matthaus, M. Schmitt, C. Krafft, A. Fahr, J. Popp, Non-invasive depth profile imaging of the stratum corneum using confocal Raman microscopy: first insights into the method, Eur. J. Pharm. Sci. 50 (2013) 601– 608. [49] D. Lunter, R. Daniels, Confocal Raman microscopic investigation of the effectiveness of penetration enhancers for procaine delivery to the skin, J. Biomed. Opt. 19 (2014) 126015. [50] W. Werncke, I. Latka, S. Sassning, B. Dietzek, M.E. Darvin, M.C. Meinke, J. Popp, K. Konig, J.W. Fluhr, J. Lademann, Two-color Raman spectroscopy for the simultaneous detection of chemotherapeutics and antioxidative status of human skin, Laser Phys. Lett. 8 (2011) 895–900. [51] A. Vogt, F. Rancan, S. Ahlberg, B. Nazemi, C.S. Choe, M.E. Darvin, S. Hadam, U. Blume-Peytavi, K. Loza, J. Diendorf, M. Epple, C. Graf, E. Ruhl, M.C. Meinke, J. Lademann, Interaction of dermatologically relevant nanoparticles with skin cells and skin, Beilstein J. Nanotechnol. 5 (2014) 2363–2373. [52] J.W. Fluhr, P. Caspers, J.A. van der Pol, H. Richter, W. Sterry, J. Lademann, M.E. Darvin, Kinetics of carotenoid distribution in human skin in vivo after exogenous stress: disinfectant and wIRA-induced carotenoid depletion recovers from outside to inside, J. Biomed. Opt. 16 (2011) 035002. [53] J. Lademann, P.J. Caspers, A. van der Pol, H. Richter, A. Patzelt, L. Zastrow, M. Darvin, W. Sterry, J.W. Fluhr, In vivo Raman spectroscopy detects increased epidermal antioxidative potential with topically applied carotenoids, Laser Phys. Lett. 6 (2009) 76–79. [54] Y. Zhu, C.S. Choe, S. Ahlberg, M.C. Meinke, U. Alexiev, J. Lademann, M.E. Darvin, Penetration of silver nanoparticles into porcine skin ex vivo using fluorescence lifetime imaging microscopy, Raman microscopy, and surface-enhanced Raman scattering microscopy, J. Biomed. Opt. 20 (2015) 51006. [55] S. Tfaili, C. Gobinet, G. Josse, J.F. Angiboust, A. Baillet, M. Manfait, O. Piot, Vibrational spectroscopies for the analysis of cutaneous permeation: experimental limiting factors identified in the case of caffeine penetration, Anal. Bioanal. Chem. 405 (2013) 1325–1332. [56] S. Tfaili, G. Josse, J.F. Angiboust, M. Manfait, O. Piot, Monitoring caffeine and resveratrol cutaneous permeation by confocal Raman microspectroscopy, J. Biophoton. 7 (2014) 676–681. [57] C.S. Choe, J. Lademann, M.E. Darvin, Gaussian-function-based deconvolution method to determine the penetration ability of petrolatum oil into in vivo human skin using confocal Raman microscopy, Laser Phys. 24 (2014) 105601. [58] A. Patzelt, J. Lademann, H. Richter, M.E. Darvin, S. Schanzer, G. Thiede, W. Sterry, T. Vergou, M. Hauser, In vivo investigations on the penetration of various oils and their influence on the skin barrier, Skin Res. Technol.: Off. J. Int. Soc. Bioeng. Skin 18 (2012) 364–369.

57

[59] G.N. Stamatas, J. de Sterke, M. Hauser, O. von Stetten, A. van der Pol, Lipid uptake and skin occlusion following topical application of oils on adult and infant skin, J. Dermatol. Sci. 50 (2008) 135–142. [60] C. Choe, J. Lademann, M.E. Darvin, Confocal Raman microscopy for investigating the penetration of various oils into the human skin in vivo, J. Dermatol. Sci. (2015). [61] E.Y. Bonnist, J.P. Gorce, C. Mackay, R.U. Pendlington, P.D. Pudney, Measuring the penetration of a skin sensitizer and its delivery vehicles simultaneously with confocal Raman spectroscopy, Skin Pharmacol. Physiol. 24 (2011) 274– 283. [62] L. Franzen, J. Anderski, M. Windbergs, Quantitative detection of caffeine in human skin by confocal Raman spectroscopy – a systematic in vitro validation study, Eur. J. Pharm. Biopharm. 95 (2015) 110–116. [63] L. Franzen, J. Anderski, V. Planz, K.H. Kostka, M. Windbergs, Combining confocal Raman microscopy and freeze-drying for quantification of substance penetration into human skin, Exp. Dermatol. 23 (2014) 942–944. [64] L. Franzen, D. Selzer, J.W. Fluhr, U.F. Schaefer, M. Windbergs, Towards drug quantification in human skin with confocal Raman microscopy, Eur. J. Pharm. Biopharm. 84 (2013) 437–444. [65] S.K. Teh, W. Zheng, K.Y. Ho, M. Teh, K.G. Yeoh, Z. Huang, Diagnostic potential of near-infrared Raman spectroscopy in the stomach: differentiating dysplasia from normal tissue, Br. J. Cancer 98 (2008) 457–465. [66] S. Feng, R. Chen, J. Lin, J. Pan, G. Chen, Y. Li, M. Cheng, Z. Huang, J. Chen, H. Zeng, Nasopharyngeal cancer detection based on blood plasma surface-enhanced Raman spectroscopy and multivariate analysis, Biosens. Bioelectron. 25 (2010) 2414–2419. [67] S. Duraipandian, W. Zheng, J. Ng, J.J. Low, A. Ilancheran, Z. Huang, Nearinfrared-excited confocal Raman spectroscopy advances in vivo diagnosis of cervical precancer, J. Biomed. Opt. 18 (2013) 067007. [68] J. Schleusener, P. Gluszczynska, C. Reble, I. Gersonde, J. Helfmann, H.J. Cappius, J.W. Fluhr, M.C. Meinke, Perturbation factors in the clinical handling of a fibercoupled Raman probe for cutaneous in vivo diagnostic Raman spectroscopy, Appl. Spectrosc. 69 (2015) 243–256. [69] U. Jacobi, M. Kaiser, R. Toll, S. Mangelsdorf, H. Audring, N. Otberg, W. Sterry, J. Lademann, Porcine ear skin: an in vitro model for human skin, Skin Res. Technol.: Off. J. Int. Soc. Bioeng. Skin 13 (2007) 19–24. [70] M.E. Darvin, H. Richter, Y.J. Zhu, M.C. Meinke, F. Knorr, S.A. Gonchukov, K. Koenig, J. Lademann, Comparison of in vivo and ex vivo laser scanning microscopy and multiphoton tomography application for human and porcine skin imaging, Quantum Electron. 44 (2014) 646–651. [71] C. Choe, J. Lademann, M.E. Darvin, Analysis of human and porcine skin in vivo/ ex vivo for penetration of selected oils by confocal Raman microscopy, Skin Pharmacol. Physiol. 28 (2015) 318–330. [72] M.Y. Akhalaya, G.V. Maksimov, A.B. Rubin, J. Lademann, M.E. Darvin, Molecular action mechanisms of solar infrared radiation and heat on human skin, Ageing Res. Rev. 16 (2014) 1–11. [73] P.J. Caspers, G.W. Lucassen, E.A. Carter, H.A. Bruining, G.J. Puppels, In vivo confocal Raman microspectroscopy of the skin: noninvasive determination of molecular concentration profiles, J. Invest. Dermatol. 116 (2001) 434–442. [74] M.E. Darvin, M.C. Meinke, W. Sterry, J. Lademann, Optical methods for noninvasive determination of carotenoids in human and animal skin, J. Biomed. Opt. 18 (2013) 61230. [75] M. Egawa, T. Hirao, M. Takahashi, In vivo estimation of stratum corneum thickness from water concentration profiles obtained with Raman spectroscopy, Acta Derm.-Venereol. 87 (2007) 4–8. [76] T. Rajalahti, O.M. Kvalheim, Multivariate data analysis in pharmaceutics: a tutorial review, Int. J. Pharm. 417 (2011) 280–290. [77] S. Dochow, C. Krafft, U. Neugebauer, T. Bocklitz, T. Henkel, G. Mayer, J. Albert, J. Popp, Tumour cell identification by means of Raman spectroscopy in combination with optical traps and microfluidic environments, Lab Chip 11 (2011) 1484–1490. [78] L. Zhang, W.S. Dong, D. Zhang, G.M. Shi, Two-stage image denoising by principal component analysis with local pixel grouping, Pattern Recogn. 43 (2010) 1531–1549. [79] C. Sarbu, R.D. Nascu-Briciu, A. Kot-Wasik, S. Gorinstein, A. Wasik, J. Namiesnik, Classification and fingerprinting of kiwi and pomelo fruits by multivariate analysis of chromatographic and spectroscopic data, Food Chem. 130 (2012) 994–1002. [80] K. Gajjar, L.D. Heppenstall, W.Y. Pang, K.M. Ashton, J. Trevisan, I.I. Patel, V. Llabjani, H.F. Stringfellow, P.L. Martin-Hirsch, T. Dawson, F.L. Martin, Diagnostic segregation of human brain tumours using Fourier-transform infrared and/or Raman spectroscopy coupled with discriminant analysis, Anal. Meth. – Uk 5 (2013) 89–102. [81] W.W. Chen, J. Lin, R. Chen, S.Y. Feng, Y. Yu, D. Lin, M.Z. Huang, H. Shi, H. Huang, Detection and identification of Huo-Xue-Hua-Yu decoction (HXHYD) using surface-enhanced Raman scattering (SERS) spectroscopy and multivariate analysis, Laser Phys. Lett. 12 (2015) 045602. [82] G. Vitkova, K. Novotny, L. Prokes, A. Hrdlicka, J. Kaiser, J. Novotny, R. Malina, D. Prochazka, Fast identification of biominerals by means of stand-off laserinduced breakdown spectroscopy using linear discriminant analysis and artificial neural networks, Spectrochim. Acta B 73 (2012) 1–6. [83] A. Subasi, M.I. Gursoy, EEG signal classification using PCA, ICA, LDA and support vector machines, Expert Syst. Appl. 37 (2010) 8659–8666. [84] W. Chen, Z. Li, Y. Yu, D. Lin, H. Huang, H. Shi, Houttuynia cordata Thunb extract induces cytotoxicity in human nasopharyngeal carcinoma cells: Raman spectroscopic studies, Laser Phys. Lett. 13 (2016) 015601.

58

S. Mujica Ascencio et al. / European Journal of Pharmaceutics and Biopharmaceutics 104 (2016) 51–58

[85] M. Baranska, L.M. Proniewicz, Raman mapping of caffeine alkaloid, Vib. Spectrosc. 48 (2008) 153–157. [86] H.Y. Wei, J. Lin, W.H. Huang, Z.B. Feng, D.W. Li, Preparation of TeO2 based thin films by nonhydrolytic sol–gel process, Mater. Sci. Eng. B – Adv. 164 (2009) 51–59. [87] P.J. Caspers, G.W. Lucassen, H.A. Bruining, G.J. Puppels, Automated depthscanning confocal Raman microspectrometer for rapid in vivo determination of water concentration profiles in human skin, J. Raman Spectrosc. 31 (2000) 813–818. [88] A. Tfayli, O. Piot, F. Pitre, M. Manfait, Follow-up of drug permeation through excised human skin with confocal Raman microspectroscopy, Eur. Biophys. J.: EBJ 36 (2007) 1049–1058. [89] D.J. Biau, B.M. Jolles, R. Porcher, P value and the theory of hypothesis testing an explanation for new researchers, Clin. Orthop. Relat. Res. 468 (2010) 885–892. [90] S.K. Teh, W. Zheng, K.Y. Ho, M. Teh, K.G. Yeoh, Z. Huang, Near-infrared Raman spectroscopy for optical diagnosis in the stomach: identification of Helicobacter-pylori infection and intestinal metaplasia, Int. J. Cancer 126 (2010) 1920–1927.

[91] L. Chen, Y. Wang, N.R. Liu, D. Lin, C.C. Weng, J.X. Zhang, L.H. Zhu, W.S. Chen, R. Chen, S.Y. Feng, Near-infrared confocal micro-Raman spectroscopy combined with PCA–LDA multivariate analysis for detection of esophageal cancer, Laser Phys. 23 (2013) 065601. [92] B.G. Saar, L.R. Contreras-Rojas, X.S. Xie, R.H. Guy, Imaging drug delivery to skin with stimulated Raman scattering microscopy, Mol. Pharm. 8 (2011) 969–975. [93] M. Gajewska, A. Paini, J.V. Sala Benito, J. Burton, A. Worth, C. Urani, H. Briesen, K.W. Schramm, In vitro-to-in vivo correlation of the skin penetration, liver clearance and hepatotoxicity of caffeine, Food Chem. Toxicol.: Int. J. Published Brit. Ind. Biol. Res. Assoc. 75 (2015) 39–49. [94] D. Zhu, K.V. Larin, Q. Luo, V.V. Tuchin, Recent progress in tissue optical clearing, Laser Photon. Rev. 7 (5) (2013) 732–757. [95] T. Yu, X. Wen, V.V. Tuchin, Q. Luo, Dan Zhu, Quantitative analysis of dehydration in porcine skin for assessing mechanism of optical clearing, J. Biomed. Opt. 16 (9) (2011) 095002. [96] D. Huang, W. Zhang, H. Zhong, H. Xiong, X. Guo, Z. Guo, Optical clearing of porcine skin tissue in vitro studied by Raman microspectroscopy, J. Biomed. Opt. 17 (1) (2012) 015004.