Skin Research and Technology 2016; 22: 55–62 Printed in Singapore All rights reserved doi: 10.1111/srt.12228
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Skin Research and Technology
Human skin penetration of hyaluronic acid of different molecular weights as probed by Raman spectroscopy M. Essendoubi1, C. Gobinet1, R. Reynaud2, J. F. Angiboust1, M. Manfait1 and O. Piot1 1
MEDyC Unit, MeDIAN Biophotonique et Technologies pour la Sante, SFR CAP SANTE, CNRS UMR 7369, Faculty of Pharmacy, University of Reims Champagne – Ardenne (URCA), Reims, France and 2Soliance, Route de Bazancourt, Pomacle, France
Background: Topical delivery of molecules into the human skin is one of the main issues in dermatology and cosmetology. Several techniques were developed to study molecules penetration into the human skin. Although widely accepted, the conventional methods such as Franz diffusion cells are unable to provide the accurate localization of actives in the skin layers. A different approach based on Raman spectroscopy has been proposed to follow-up the permeation of actives. It presents a high molecular specificity to distinguish exogenous molecules from skin constituents. Methods: Raman micro-imaging was applied to monitor the skin penetration of hyaluronic acids (HA) of different molecular weights. The first step, was the spectral characterization of these HA. After, we have determined spectral features of HA by which they can be detected in the skin. In the second part, transverse skin sections were realized and spectral images were recorded.
Results: Our results show a difference of skin permeation of the three HA. Indeed, HA with low molecular weight (20– 300 kDa) passes through the stratum corneum in contrast of the impermeability of high molecular weight HA (1000– 1400 kDa). Conclusion: Raman spectroscopy represents an analytical, non-destructive, and dynamic method to evaluate the permeation of actives in the skin layers.
YALURONIC ACID (HA), or hyaluronan, is a natural carbohydrate linear polysaccharide; that is found in almost all living organisms. Its chemical structure is consisting of multiple disaccharides which are N-acetylglucosamine and D-glucuronic acid, linked via alternating b-1,4 and b-1,3 glycosidic bonds (1, 2). HA, is a highly hydrophilic molecule, plays an important role in tissue hydrodynamics and contributes to the transport of water, it helps to maintain the hydration and elastoviscosity of tissues (2, 3). The remarkable viscoelastic and water holding property of HA, besides its biocompatibility, biodegradability, and non-immunogenicity, has increased its appeal in numerous medical and cosmetic applications (4–9). Human body contains approximately 15 g of HA. It is reported that in human skin, the quantity of HA represents a third of the all amount of HA that exists in human body (4). Also
believed that one-third of the whole human body quantity of HA is broken down and synthesized on a daily basis (2). However, human cells do not always produce HA efficiently lifelong. When HA production declines, it results in myofascial rigidity, skin aging, dryness, and wrinkles (4).In human skin, HA is found in epidermis and dermis, where it is acting as protective, structure stabilizing, and shock-absorbing agent. By age, human skin loses its hydration, smoothness, softening, and elasticity which are associated with the decrease of HA content, thus the development of wrinkles. Nowadays, HA plays an important role in cosmetics by its highly effective moisturizing property. It is the most natural hydrophilic molecule, since it can attract and retain till 1000 times its weight in water (3, 6). Recent in-vivo studies have shown that topically applied HA on the skin reduces signs of skin aging. HA, especially of high molecular
H
Key words: Raman spectroscopy – hyaluronic acid of different molecular weight – skin – penetration/permeation – localization
Ó 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Accepted for publication 15 February 2015
55
Essendoubi et al.
weight, hydrate the skin by forming a film on the skin surface and preventing water loss. Whereas, low molecular weight of HA can penetrate skin to protect and support the epidermal hydration, to moisturize the stratum corneum continuously to assure high quality of the epidermal texture (3, 7, 9). Hyaluronic acids based products are also used in cosmetic surgery. Either in a stabilized form or in combination with other polymers, HA is used as a component of commercial dermal fillers. Scientific data has shown that injection of such products into the dermis can be effective in augmentation therapy of soft tissues of the face by reducing facial lines and wrinkles in the longterm with fewer side-effects and better tolerability compared with the use of collagen (10). Recent investigations about HA size and its effect on skin aging have shown that low molecular weight HA of approximately 50 kDa, has revealed significantly higher skin penetration rates than larger sized HA. Indeed, low molecular weight HA can influence the expression of many genes including those contributing to keratinocyte differentiation and formation of intercellular tight junction complexes which are reported to be reduced in aged and photodamaged skin, thereby improving the skin quality (9). However, this plentitude of beneficial effects of HA is limited by its molecular size. Which can range from 5000 to 5,000,000 Da, thus interfering with efficient skin penetration for better effect (11). Furthermore, the penetration of low molecular weight is more controversial, indeed according to the bibliography most of the chemical compounds with a molecular weight higher than 500 Da cannot penetrate the skin. Topical penetration of molecules through the skin structures is one of the main issues in dermatology and cosmetology. Several techniques are commonly used for evaluating the rate, the speed and the depth of penetration of these molecules. Diffusion cells (e.g. Franz Cell Chamber) and tape-stripping (12, 13), although widely accepted, these methods lack of spatial accuracy. Diffusion cells cannot be used for molecules whose penetration is limited to the outermost layers (stratum corneum) of the skin. In addition tape-stripping is considered as a destructive method and the complete stratum corneum can be removed by using 50 tape strips (14).
56
Novel approaches based on radiolabeling are finding their way into the field of drugs permeation through the skin (15). These techniques, present some technical limits due to protocol complexities, reagent costs, and are not up to now applied in routine. Biophotonic techniques are presented as candidate methods since they can provide valuable information on the kinetics, speed and depth of penetration of exogenous molecules into the skin. Hence, Raman spectroscopy is a powerful laser spectroscopic technique that detects the characteristic vibrational energy levels of a molecule. Thereby Raman spectra constitute highly specific spectroscopic fingerprints of molecules by which they can be identified and tracked into the skin (16–18). This approach represents an analytical, non-destructive, and dynamic method to investigate drugs or actives permeation in skin layers with a micrometer spatial resolution. It permits also a detailed structural analysis to identify certain molecular alterations in the skin due to a physiological change or to interactions with exogenous molecules (19). Furthermore, combining chemometric methods with spectral imaging performed on skin transverse sections allows a precise localization of molecules into the skin layers (20, 21). In this study, we have applied Raman imaging to investigate the penetration of HA of different molecular weights on human skin sections. Three HA derivatives were used: Cristalhyal (1000–1400 kDa), Bashyal (100– 300 kDa), and Renovhyal (20–50 kDa) (SOLIANCE, Bazancourt, France). In a first part of this study, we have determined the spectral signature of the three HA in the solid and solubilized forms. By applying multivariate statistical processing, the reference HA signatures were then compared to the reference spectra of untreated skin sample (negative control), to identify the spectral features which will be used to detect the HA solution in skin samples. In the second part, each HA solution was deposited on the skin surface and 10 lm-thick transverse skin sections were realized to perform Raman micro-imaging and to investigate the HA cutaneous penetration levels.
Materials and Methods Human skin sample Human dermatomed skins were provided by Bioiopredic International (Rennes, France). The
Human skin penetration of hyaluronic acid
skin samples were obtained from plastic surgical intervention, isolated from the abdomen of 44 years old Caucasian woman and have a mean thickness of 323 lm (CV = 8%). For cutaneous permeation tests, full thickness human dermatomed skin was used and surface area of 1.5 cm² of skin were cut manually using a scalpel. Each square of the skin was stretched and sewed between two circular slices with a central hole of 10 mm in diameter. After this the skin was deposed on a sterile Petri dish (small size; 35 mm 9 10 mm) filled with 5 mL of a phosphate buffered saline solution (PBS, Invitrogen, Paisley, UK).
Preparation of HA solutions Three HA with different molecular weights Cristalhyal (1000–1400 kDa), Bashyal (100– 300 kDa) and Renovhyal (20–50 kDa) were provided by SOLIANCE (Bazancourt, France). According to the small concentration of HA in cosmetic products and the sensitivity of Raman spectroscopy, HA solutions were prepared at 1% by dissolving 1 g of each HA in 100 mL of sterile distilled water. For control test, we analyzed 1% glycerin aqueous solution (Merck Schuchardt, Hohenbrunn, Germany) and water treated skin samples, respectively, corresponding to the positive and negative controls. The glycerin is known to have a high diffusivity through the human skin. Reference Raman spectra of pure HA and glycerin were recorded to identify the characteristic vibrational bands of the molecules in the solid and solubilized form.
Preparation of transverse skin sections For permeation tests of the HA solutions, 300 lL of solution were deposited on the skin surface during 8 h at 32°C. After the diffusion period, the skin surface was gently cleaned with a dry cotton swab. The samples were then frozen at 20°C and 10 lm-thick transverse skin sections were realized by cryo-microtome.
Raman spectroscopy Raman spectral acquisitions were recorded from skin sections deposited on Calcium Fluoride slides by using a Labram microspectrometer (Horiba Jobin Yvon, Villeneuve d’Ascq, France).
The set-up included a microscope (BX40; Olympus, Rungis, France) coupled to a dispersive Raman spectrograph. Raman measurements were recorded using a 1009 long working distance objective operating in air with a numerical aperture of 0.75. The Raman scattering was analyzed by the spectrometer equipped with a charge-coupled device (CCD) detector of 1024 9 256 elements cooled by Peltier effect at 65°C. The spectrometer comprises also a grating of 950 grooves/mm permitting to collect data with a spectral resolution of 4/cm and axial resolution of 3 lm. The excitation source was a 660 nm laser diode (Ignis Laser Quantum GmbH, Konstanz, Germany), this for an optimal compromise between generation of parasitic fluorescence and sensitivity of the CCD detector. The spectral acquisitions were recorded on the 400 to 4000/cm spectral range and each spectrum was acquired with two accumulations of 15 s. Before using, the instrument was calibrated to the 520.7/cm Raman line of silicon. The skin cryo-sections were placed on the microscope and Raman maps were recorded in a zone of Y: 10 lm/X: 100 lm perpendicular on the surface of the skin. Raman maps were measured by defining a scanning xy-step size of 5 lm and data acquisition was performed using LabSpec 5 software (Horiba Jobin Yvon).
Data pre-processing and image analysis The pre-processing of Raman spectral data was performed using LabSpec 5 software. The procedure was as follows: correction of instrument response in the spectra by subtracting the black current, detector response and optical system signals, noise reduction using a 5 points average Savitsky–Golay smoothing and finally baseline correction using a polynomial function of degree 5 for removing the fluorescence background. This pre-processing was done on the total spectral region (400–4000/cm) and spectra with low signal to noise ratio (SNR < 10) were discarded. The mean signal is calculated in – CH region (2800–3000/cm) and the noise in 2000–2200/cm. After these pre-processing steps, mean Raman spectra were calculated for each test from three to six repetitive measurements of three different skin sections. The processing of corrected data maps was performed by using homemade software based
57
Essendoubi et al.
on least squares fitting method that operates in the Matlab environment (The Math Works Inc., Natick, MA, USA). A thorough description of this statistical analysis has been described elsewhere (22). For spectral image treatment, the set of spectra were vector normalized on the whole spectral range. For each HA solution, corrected Raman spectrum can thus be described by the following linear model: S ¼ aHA sHA þ acontrol scontrol þ b0 þ b1 K þ b2 K2 þ b3 K3 þ b4 K4 þ b5 K5 þ e where S is the considered corrected Raman spectrum, sHA and scontrol, respectively, corresponding to the reference Raman spectra of HA and untreated skin sample (negative control). For positive control, reference Raman spectrum of glycerin was used. The quantities aHA and acontrol corresponding to the abundance fraction of each reference spectrum into recorded spectrum S. The term b0 + b1Λ + b2Λ2 + b3Λ3 + b4Λ4 + b5Λ5 is a five order polynomial function used for baseline correction, where Λ is the vector of recorded wavenumbers and b0, b1, b2, b3, b4, and b5 are the coefficients of the polynomial function. e represents the modelization error. Spectra with a modelization error higher than 0.05 were discarded. For each spectral image of a transverse skin sections, the estimation of the fitting coefficients aHA and acontrol for each spectrum informs about the HA concentration in each pixel and leads to the formation of images reflecting the HA distribution into the skin sections.
glycerin. The corresponding Raman spectra are shown in Fig. 1. The major bands of HA are found in two spectral windows: 800–1660/cm and 2700–3000/cm. Table 1 summarizes the Raman bands of various HA and glycerin (Table 1). Although the signal of HA and glycerin is very weak compared to skin signal, it is possible to highlight the HA or glycerin vibrations by using Mann–Whitney statistical test. Indeed, the comparison between mean spectra of HA and skin by using the Mann–Whitney statistical test (P = 0.05) allow to determine the most pertinent spectral features which can be used as markers for detecting the HA in the skin. The same procedure was followed for the determination of glycerin markers. Figure 2 and Table 2 summarizes the vibration bands which were used to identify HA and glycerin in the skin (Fig. 2, Table 2).
Spectral Raman micro-imaging and spatial distribution of HAs in skin sections Raman maps were recorded on a zone of 100 9 10 lm2, the longest size being perpendicular to the surface of the skin, this zone corresponds mainly to the epidermal layers of the skin. In Fig. 3, the white light image represents a skin section together with the spectral acquisition zone (Fig. 3). To determine the levels of HAs permeation, a supervised processing approach was used by taking into account the reference spectra of the various HA, and of glycerin and water treated skin samples, respectively, corresponding to the positive and negative controls. Raman images were reconstructed using a fitting procedure
Results and Discussion Raman spectroscopy is sensitive enough to reflect small variations due to skin origin, storage mode, and sample preparation. The standardization of the experimental conditions and spectral acquisition parameters are important for the reproducibility of spectra. Consequently, only spectra recorded for the same batch of skin and under the same precise conditions of sample preparation were compared.
Spectral signature and spectral features of HA The first step of this work was to determine the reference Raman spectra of each HA and
58
Fig. 1. Raman spectra of various hyaluronic acid and glycerin.
Human skin penetration of hyaluronic acid TABLE 1. Assignment of Raman bands for hyaluronic acid (HA) and glycerin (26, 27) HA Raman shift (per cm)
Assignment
446 899 949 1047 1091 1125
– – – C-C and C-O stretching C-OH bend, acetyl group C4-OH bend and C4-H bend
1205 1328 1372 1406 1660 2904 2933
CH2 twist C-H bend, Amide III C-H bend C-N stretching and C-H deformation C=C and Amid I CH stretching N-H stretching
(a)
Glycerin Raman shift (per cm)
Assignment
414 484 672 818 849 923 975 1055 1465 2886 2939
CCO rock CCO rock – CC stretch CC stretch CH2 rock CH2 rock CO stretch from C-1 (=C-3 CH2 deformation Symmetric CH stretch from CH2 Antisymmetric CH stretch from CH2
(b)
Fig. 2. Comparison between skin reference spectra (gray) and (a) hyaluronic acid spectra (black), (b) and glycerin spectra (black).
TABLE 2. Assignment of vibrational peaks were selected to be used as hyaluronic acid (HA) and glycerin markers in the skin HA Raman shift (per cm) 446 949 1047 1372 1406 2904
Glycerin
Assignment – – C-C and C-O stretching C-H bend C-N stretching and C-H deformation CH stretching
Raman shift (per cm) 484 672 818 923 1055 1465
Assignment CCO rock – CC stretch CH2 rock CO stretch from C-1 (=C-3 CH2 deformation
that takes into account all reference spectra, and each pixel spectrum is represented by a linear combination of the reference spectra, respectively, weighted by a fitting coefficient. For each treatment condition (HAs or glycerin), the fitting coefficients were averaged over three measurements performed on different but adjacent skin sections.
Fig. 3. Light microscopy image of skin section with mapping zone.
In Fig. 4, the reconstructed spectral images permit to see that the signals of HAs with low molecular weight: Renovhyal and Bashyal are present in the skin section, at around 100 lm (full epidermal depth) for Renovhyal and 50 lm of epidermal depth for Bashyal (Fig. 4d and e). Whereas for the HA with high molecular weight:
59
Essendoubi et al. (a)
(b)
(c)
(d)
(e)
Fig. 4. Representation of the permeation and localization of various hyaluronic acid (HA) after fitting by their reference spectra and by comparison to the (a) positive (glycerin treated skin) and (b) negative (water treated skin) controls, (c) Cristalhyal (1000–1400 kDa HA), (d) Bashyal (100–300 kDa HA), and (e) Renovhyal (20–50 kDa HA).
Cristalhyal the permeation does not exceed 25 lm (Fig. 4c). By inspecting the light microscopy images and Raman reconstructed images, we can observe that for all HAs, the major quantity was found in the stratum corneum, around 25 lm from surface. Renovhyal HA is highly present in the deepest epidermis layers of epidermis, whereas Bashyal HA is localized in the superficial layer of the epidermis under stratum corneum. The HA with high molecular weight Cristalhyal is found only in the stratum corneum. These results show that HAs permeation increases with decreasing molecular weight of HA. This is in agreement with other studies demonstrating that low molecular weight HA (50 kDa) was associated with significant improvement in skin hydration and wrinkle depth, which is due to better penetration
60
abilities of low molecular weight of HA (3, 7, 9). Whereas, HA with a high molecular weight (more than 1000 kDa) needs to stay at the surface of the skin where it makes a film permitting the diminution of skin evaporation and limiting interaction with the environmental factors such as temperature, humidity, and UV radiation. In this case, penetration is not required (4, 6, 23). We can see also in (Fig. 4a and b), for control tests, there is no HA signal in negative control, while for positive control the glycerin signal is recovered in the skin section at full epidermal depth (100 lm) reflecting the high diffusion of glycerin through the human skin. Several studies demonstrate a significant improvement in skin hydration, wrinkle depth, and elasticity by using HA-based creams while they do not show any argument of the HA
Human skin penetration of hyaluronic acid
penetration. Furthermore, the penetration of the chemical compounds with a molecular weight higher than 500 Da is more controversial according to the bibliography (3). Compared to other reported data and according to our knowledge, this is the first report Raman micro-imaging to prove permeation and spatial distribution of HA with different molecular weight in human skin. Raman spectroscopy is promising approach for probing exogenous molecules in the human skin. Recent studies using this technique have demonstrated the possibility of tracking cosmetic and pharmaceutical formulations (24). Recently, the spatial distribution of Sodium Dodecyl Sulfate and monitoring of caffeine and Metronidazole in skin by Raman spectroscopy have been described (18, 21, 25).
References 1. Fraser JR, Laurent TC, Laurent UB. Hyaluronan: its nature, distribution, functions and turnover. J Intern Med 1997; 242: 27–33. 2. Nusgens BV. Hyaluronic acid and extracellular matrix: a primitive molecule? Ann Dermatol Venereol 2010; 137(Suppl.): S3–S8. 3. Pavicic T, Gauglitz GG, Lersch P, Schwach-Abdellaoui K, Malle B, Korting HC, Farwick M. Efficacy of cream-based novel formulations of hyaluronic acid of different molecular weights in anti-wrinkle treatment. J Drugs Dermatol 2011; 10: 990–1000. 4. Necas JLB, Brauner P, Kolar J. Hyaluronic acid (hyaluronan). Vet Med 2008; 53: 397–411. 5. Barbucci R, Lamponi S, Borzacchiello A, Ambrosio L, Fini M, Torricelli P, Giardino R. Hyaluronic acid hydrogel in the treatment of osteoarthritis. Biomaterials 2002; 23: 4503–4513. 6. Weindl G, Schaller M, Schafer-Korting M, Korting HC. Hyaluronic acid in the treatment and prevention of skin diseases: molecular biological, pharmaceutical and clinical aspects. Skin Pharmacol Physiol 2004; 17: 207–213. 7. Farwick MPL, Strutz G. Low molecular weight hyaluronic acid: its effects on epidermal gene expression & skin ageing. SOFW Journal 2008; 134: 17–22. 8. Chen WY, Abatangelo G. Functions of hyaluronan in wound repair. Wound Repair Regen 1999; 7: 79– 89.
As a conclusion, the present work demonstrates that Raman imaging combined with chemometric methods is an interesting alternative tool for cutaneous permeation tests. In particular, in the present study, we demonstrated the skin permeability of low molecular weight HA (20–300 kDa) and the impermeability of high molecular weight HA (1000–1400 kDa). An essential aspect of this work was to determine the most pertinent spectral features to detect various HA in the skin section, by using a statistical test (Mann–Whitney P = 0.05). Raman spectroscopy exhibits several advantages. It enables the detection and the localization of exogenous molecules at a micrometric resolution, in a label-free, non-destructive manner and with a limited sample preparation.
9. Farwick M, Gauglitz G, Pavicic T, Kohler T, Wegmann M, SchwachAbdellaoui K, Malle B, Tarabin V, Schmitz G, Korting HC. Fifty-kDa hyaluronic acid upregulates some epidermal genes without changing TNF-alpha expression in reconstituted epidermis. Skin Pharmacol Physiol 2011; 24: 210–217. 10. Duranti F, Salti G, Bovani B, Calandra M, Rosati ML. Injectable hyaluronic acid gel for soft tissue augmentation. A clinical and histological study. Dermatol Surg 1998; 24: 1317–1325. 11. Saari H, Konttinen YT, Friman C, Sorsa T. Differential effects of reactive oxygen species on native synovial fluid and purified human umbilical cord hyaluronate. Inflammation 1993; 17: 403– 415. 12. Moran JW, Bell S. Skin permeation of organic gunshot residue: implications for sampling and analysis. Anal Chem 2014; 86: 6071–6079. 13. 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: 34–39. 14. Jacobi U, Weigmann HJ, Ulrich J, Sterry W, Lademann J. Estimation of the relative stratum corneum amount removed by tape stripping. Skin Res Technol 2005; 11: 91–96. 15. Brown TJ, Alcorn D, Fraser JR. Absorption of hyaluronan applied to the surface of intact skin. J
16.
17.
18.
19.
20.
21.
22.
Invest Dermatol 1999; 113: 740– 746. Tfayli A, Piot O, Pitre F, Manfait M. Follow-up of drug permeation through excised human skin with confocal Raman microspectroscopy. Eur Biophys J 2007; 36: 1049– 1058. Xiao C, Moore DJ, Rerek ME, Flach CR, Mendelsohn R. Feasibility of tracking phospholipid permeation into skin using infrared and Raman microscopic imaging. J Invest Dermatol 2005; 124: 622–632. Tfaili S, Josse G, Angiboust JF, Manfait M, Piot O. Monitoring caffeine and resveratrol cutaneous permeation by confocal Raman microspectroscopy. J Biophotonics 2013; 7: 676–6811. Caspers PJ, Lucassen GW, Wolthuis R, Bruining HA, Puppels GJ. In vitro and in vivo Raman spectroscopy of human skin. Biospectroscopy 1998; 4(Suppl.): S31–S39. Caspers PJ, Lucassen GW, Carter EA, Bruining HA, Puppels GJ. In vivo confocal Raman microspectroscopy of the skin: noninvasive determination of molecular concentration profiles. J Invest Dermatol 2001; 116: 434–442. Pudney PD, Melot M, Caspers PJ, Van Der Pol A, Puppels GJ. An in vivo confocal Raman study of the delivery of trans retinol to the skin. Appl Spectrosc 2007; 61: 804– 811. Draux F, Gobinet C, Sule-Suso J, Manfait M, Jeannesson P, Sockalingum GD. Raman imaging of single living cells: probing effects of noncytotoxic doses of an anti-cancer
61
Essendoubi et al. drug. Analyst 2011; 136: 2718– 2725. 23. Vyumvuhore R, Tfayli A, Duplan H, Delalleau A, Manfait M, BailletGuffroy A. Effects of atmospheric relative humidity on Stratum Corneum structure at the molecular level: ex vivo Raman spectroscopy analysis. Analyst 2013; 138: 4103–4111. 24. Patel A, Bell M, O’Connor C, Inchley A, Wibawa J, Lane ME. Delivery of ibuprofen to the skin. Int J Pharm 2013; 457: 9–13. 25. Mao G, Flach CR, Mendelsohn R, Walters RM. Imaging the distribu-
62
tion of sodium dodecyl sulfate in skin by confocal Raman and infrared microspectroscopy. Pharm Res 2012; 29: 2189–2201. 26. Alkrad JA, Mrestani Y, Stroehl D, Wartewig S, Neubert R. Characterization of enzymatically digested hyaluronic acid using NMR, Raman, IR, and UV-Vis spectroscopies. J Pharm Biomed Anal 2003; 31: 545–550. 27. Mendelovici E, Frost RL, Kloprogge T. Cryogenic Raman spectroscopy of glycerol. J Raman Spectrosc 2000; 31: 1121–1126.
Address: Professor Olivier Piot Faculty of Pharmacy University of Reims Champagne – Ardenne (URCA) 51 rue Cognacq Jay 51096 Reims Cedex France Tel.: +33 3 26 91 81 28 Fax: +33 3 26 91 35 50 e-mail: [email protected]
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Skin Research and Technology
Human skin penetration of hyaluronic acid of different molecular weights as probed by Raman spectroscopy M. Essendoubi1, C. Gobinet1, R. Reynaud2, J. F. Angiboust1, M. Manfait1 and O. Piot1 1
MEDyC Unit, MeDIAN Biophotonique et Technologies pour la Sante, SFR CAP SANTE, CNRS UMR 7369, Faculty of Pharmacy, University of Reims Champagne – Ardenne (URCA), Reims, France and 2Soliance, Route de Bazancourt, Pomacle, France
Background: Topical delivery of molecules into the human skin is one of the main issues in dermatology and cosmetology. Several techniques were developed to study molecules penetration into the human skin. Although widely accepted, the conventional methods such as Franz diffusion cells are unable to provide the accurate localization of actives in the skin layers. A different approach based on Raman spectroscopy has been proposed to follow-up the permeation of actives. It presents a high molecular specificity to distinguish exogenous molecules from skin constituents. Methods: Raman micro-imaging was applied to monitor the skin penetration of hyaluronic acids (HA) of different molecular weights. The first step, was the spectral characterization of these HA. After, we have determined spectral features of HA by which they can be detected in the skin. In the second part, transverse skin sections were realized and spectral images were recorded.
Results: Our results show a difference of skin permeation of the three HA. Indeed, HA with low molecular weight (20– 300 kDa) passes through the stratum corneum in contrast of the impermeability of high molecular weight HA (1000– 1400 kDa). Conclusion: Raman spectroscopy represents an analytical, non-destructive, and dynamic method to evaluate the permeation of actives in the skin layers.
YALURONIC ACID (HA), or hyaluronan, is a natural carbohydrate linear polysaccharide; that is found in almost all living organisms. Its chemical structure is consisting of multiple disaccharides which are N-acetylglucosamine and D-glucuronic acid, linked via alternating b-1,4 and b-1,3 glycosidic bonds (1, 2). HA, is a highly hydrophilic molecule, plays an important role in tissue hydrodynamics and contributes to the transport of water, it helps to maintain the hydration and elastoviscosity of tissues (2, 3). The remarkable viscoelastic and water holding property of HA, besides its biocompatibility, biodegradability, and non-immunogenicity, has increased its appeal in numerous medical and cosmetic applications (4–9). Human body contains approximately 15 g of HA. It is reported that in human skin, the quantity of HA represents a third of the all amount of HA that exists in human body (4). Also
believed that one-third of the whole human body quantity of HA is broken down and synthesized on a daily basis (2). However, human cells do not always produce HA efficiently lifelong. When HA production declines, it results in myofascial rigidity, skin aging, dryness, and wrinkles (4).In human skin, HA is found in epidermis and dermis, where it is acting as protective, structure stabilizing, and shock-absorbing agent. By age, human skin loses its hydration, smoothness, softening, and elasticity which are associated with the decrease of HA content, thus the development of wrinkles. Nowadays, HA plays an important role in cosmetics by its highly effective moisturizing property. It is the most natural hydrophilic molecule, since it can attract and retain till 1000 times its weight in water (3, 6). Recent in-vivo studies have shown that topically applied HA on the skin reduces signs of skin aging. HA, especially of high molecular
H
Key words: Raman spectroscopy – hyaluronic acid of different molecular weight – skin – penetration/permeation – localization
Ó 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Accepted for publication 15 February 2015
55
Essendoubi et al.
weight, hydrate the skin by forming a film on the skin surface and preventing water loss. Whereas, low molecular weight of HA can penetrate skin to protect and support the epidermal hydration, to moisturize the stratum corneum continuously to assure high quality of the epidermal texture (3, 7, 9). Hyaluronic acids based products are also used in cosmetic surgery. Either in a stabilized form or in combination with other polymers, HA is used as a component of commercial dermal fillers. Scientific data has shown that injection of such products into the dermis can be effective in augmentation therapy of soft tissues of the face by reducing facial lines and wrinkles in the longterm with fewer side-effects and better tolerability compared with the use of collagen (10). Recent investigations about HA size and its effect on skin aging have shown that low molecular weight HA of approximately 50 kDa, has revealed significantly higher skin penetration rates than larger sized HA. Indeed, low molecular weight HA can influence the expression of many genes including those contributing to keratinocyte differentiation and formation of intercellular tight junction complexes which are reported to be reduced in aged and photodamaged skin, thereby improving the skin quality (9). However, this plentitude of beneficial effects of HA is limited by its molecular size. Which can range from 5000 to 5,000,000 Da, thus interfering with efficient skin penetration for better effect (11). Furthermore, the penetration of low molecular weight is more controversial, indeed according to the bibliography most of the chemical compounds with a molecular weight higher than 500 Da cannot penetrate the skin. Topical penetration of molecules through the skin structures is one of the main issues in dermatology and cosmetology. Several techniques are commonly used for evaluating the rate, the speed and the depth of penetration of these molecules. Diffusion cells (e.g. Franz Cell Chamber) and tape-stripping (12, 13), although widely accepted, these methods lack of spatial accuracy. Diffusion cells cannot be used for molecules whose penetration is limited to the outermost layers (stratum corneum) of the skin. In addition tape-stripping is considered as a destructive method and the complete stratum corneum can be removed by using 50 tape strips (14).
56
Novel approaches based on radiolabeling are finding their way into the field of drugs permeation through the skin (15). These techniques, present some technical limits due to protocol complexities, reagent costs, and are not up to now applied in routine. Biophotonic techniques are presented as candidate methods since they can provide valuable information on the kinetics, speed and depth of penetration of exogenous molecules into the skin. Hence, Raman spectroscopy is a powerful laser spectroscopic technique that detects the characteristic vibrational energy levels of a molecule. Thereby Raman spectra constitute highly specific spectroscopic fingerprints of molecules by which they can be identified and tracked into the skin (16–18). This approach represents an analytical, non-destructive, and dynamic method to investigate drugs or actives permeation in skin layers with a micrometer spatial resolution. It permits also a detailed structural analysis to identify certain molecular alterations in the skin due to a physiological change or to interactions with exogenous molecules (19). Furthermore, combining chemometric methods with spectral imaging performed on skin transverse sections allows a precise localization of molecules into the skin layers (20, 21). In this study, we have applied Raman imaging to investigate the penetration of HA of different molecular weights on human skin sections. Three HA derivatives were used: Cristalhyal (1000–1400 kDa), Bashyal (100– 300 kDa), and Renovhyal (20–50 kDa) (SOLIANCE, Bazancourt, France). In a first part of this study, we have determined the spectral signature of the three HA in the solid and solubilized forms. By applying multivariate statistical processing, the reference HA signatures were then compared to the reference spectra of untreated skin sample (negative control), to identify the spectral features which will be used to detect the HA solution in skin samples. In the second part, each HA solution was deposited on the skin surface and 10 lm-thick transverse skin sections were realized to perform Raman micro-imaging and to investigate the HA cutaneous penetration levels.
Materials and Methods Human skin sample Human dermatomed skins were provided by Bioiopredic International (Rennes, France). The
Human skin penetration of hyaluronic acid
skin samples were obtained from plastic surgical intervention, isolated from the abdomen of 44 years old Caucasian woman and have a mean thickness of 323 lm (CV = 8%). For cutaneous permeation tests, full thickness human dermatomed skin was used and surface area of 1.5 cm² of skin were cut manually using a scalpel. Each square of the skin was stretched and sewed between two circular slices with a central hole of 10 mm in diameter. After this the skin was deposed on a sterile Petri dish (small size; 35 mm 9 10 mm) filled with 5 mL of a phosphate buffered saline solution (PBS, Invitrogen, Paisley, UK).
Preparation of HA solutions Three HA with different molecular weights Cristalhyal (1000–1400 kDa), Bashyal (100– 300 kDa) and Renovhyal (20–50 kDa) were provided by SOLIANCE (Bazancourt, France). According to the small concentration of HA in cosmetic products and the sensitivity of Raman spectroscopy, HA solutions were prepared at 1% by dissolving 1 g of each HA in 100 mL of sterile distilled water. For control test, we analyzed 1% glycerin aqueous solution (Merck Schuchardt, Hohenbrunn, Germany) and water treated skin samples, respectively, corresponding to the positive and negative controls. The glycerin is known to have a high diffusivity through the human skin. Reference Raman spectra of pure HA and glycerin were recorded to identify the characteristic vibrational bands of the molecules in the solid and solubilized form.
Preparation of transverse skin sections For permeation tests of the HA solutions, 300 lL of solution were deposited on the skin surface during 8 h at 32°C. After the diffusion period, the skin surface was gently cleaned with a dry cotton swab. The samples were then frozen at 20°C and 10 lm-thick transverse skin sections were realized by cryo-microtome.
Raman spectroscopy Raman spectral acquisitions were recorded from skin sections deposited on Calcium Fluoride slides by using a Labram microspectrometer (Horiba Jobin Yvon, Villeneuve d’Ascq, France).
The set-up included a microscope (BX40; Olympus, Rungis, France) coupled to a dispersive Raman spectrograph. Raman measurements were recorded using a 1009 long working distance objective operating in air with a numerical aperture of 0.75. The Raman scattering was analyzed by the spectrometer equipped with a charge-coupled device (CCD) detector of 1024 9 256 elements cooled by Peltier effect at 65°C. The spectrometer comprises also a grating of 950 grooves/mm permitting to collect data with a spectral resolution of 4/cm and axial resolution of 3 lm. The excitation source was a 660 nm laser diode (Ignis Laser Quantum GmbH, Konstanz, Germany), this for an optimal compromise between generation of parasitic fluorescence and sensitivity of the CCD detector. The spectral acquisitions were recorded on the 400 to 4000/cm spectral range and each spectrum was acquired with two accumulations of 15 s. Before using, the instrument was calibrated to the 520.7/cm Raman line of silicon. The skin cryo-sections were placed on the microscope and Raman maps were recorded in a zone of Y: 10 lm/X: 100 lm perpendicular on the surface of the skin. Raman maps were measured by defining a scanning xy-step size of 5 lm and data acquisition was performed using LabSpec 5 software (Horiba Jobin Yvon).
Data pre-processing and image analysis The pre-processing of Raman spectral data was performed using LabSpec 5 software. The procedure was as follows: correction of instrument response in the spectra by subtracting the black current, detector response and optical system signals, noise reduction using a 5 points average Savitsky–Golay smoothing and finally baseline correction using a polynomial function of degree 5 for removing the fluorescence background. This pre-processing was done on the total spectral region (400–4000/cm) and spectra with low signal to noise ratio (SNR < 10) were discarded. The mean signal is calculated in – CH region (2800–3000/cm) and the noise in 2000–2200/cm. After these pre-processing steps, mean Raman spectra were calculated for each test from three to six repetitive measurements of three different skin sections. The processing of corrected data maps was performed by using homemade software based
57
Essendoubi et al.
on least squares fitting method that operates in the Matlab environment (The Math Works Inc., Natick, MA, USA). A thorough description of this statistical analysis has been described elsewhere (22). For spectral image treatment, the set of spectra were vector normalized on the whole spectral range. For each HA solution, corrected Raman spectrum can thus be described by the following linear model: S ¼ aHA sHA þ acontrol scontrol þ b0 þ b1 K þ b2 K2 þ b3 K3 þ b4 K4 þ b5 K5 þ e where S is the considered corrected Raman spectrum, sHA and scontrol, respectively, corresponding to the reference Raman spectra of HA and untreated skin sample (negative control). For positive control, reference Raman spectrum of glycerin was used. The quantities aHA and acontrol corresponding to the abundance fraction of each reference spectrum into recorded spectrum S. The term b0 + b1Λ + b2Λ2 + b3Λ3 + b4Λ4 + b5Λ5 is a five order polynomial function used for baseline correction, where Λ is the vector of recorded wavenumbers and b0, b1, b2, b3, b4, and b5 are the coefficients of the polynomial function. e represents the modelization error. Spectra with a modelization error higher than 0.05 were discarded. For each spectral image of a transverse skin sections, the estimation of the fitting coefficients aHA and acontrol for each spectrum informs about the HA concentration in each pixel and leads to the formation of images reflecting the HA distribution into the skin sections.
glycerin. The corresponding Raman spectra are shown in Fig. 1. The major bands of HA are found in two spectral windows: 800–1660/cm and 2700–3000/cm. Table 1 summarizes the Raman bands of various HA and glycerin (Table 1). Although the signal of HA and glycerin is very weak compared to skin signal, it is possible to highlight the HA or glycerin vibrations by using Mann–Whitney statistical test. Indeed, the comparison between mean spectra of HA and skin by using the Mann–Whitney statistical test (P = 0.05) allow to determine the most pertinent spectral features which can be used as markers for detecting the HA in the skin. The same procedure was followed for the determination of glycerin markers. Figure 2 and Table 2 summarizes the vibration bands which were used to identify HA and glycerin in the skin (Fig. 2, Table 2).
Spectral Raman micro-imaging and spatial distribution of HAs in skin sections Raman maps were recorded on a zone of 100 9 10 lm2, the longest size being perpendicular to the surface of the skin, this zone corresponds mainly to the epidermal layers of the skin. In Fig. 3, the white light image represents a skin section together with the spectral acquisition zone (Fig. 3). To determine the levels of HAs permeation, a supervised processing approach was used by taking into account the reference spectra of the various HA, and of glycerin and water treated skin samples, respectively, corresponding to the positive and negative controls. Raman images were reconstructed using a fitting procedure
Results and Discussion Raman spectroscopy is sensitive enough to reflect small variations due to skin origin, storage mode, and sample preparation. The standardization of the experimental conditions and spectral acquisition parameters are important for the reproducibility of spectra. Consequently, only spectra recorded for the same batch of skin and under the same precise conditions of sample preparation were compared.
Spectral signature and spectral features of HA The first step of this work was to determine the reference Raman spectra of each HA and
58
Fig. 1. Raman spectra of various hyaluronic acid and glycerin.
Human skin penetration of hyaluronic acid TABLE 1. Assignment of Raman bands for hyaluronic acid (HA) and glycerin (26, 27) HA Raman shift (per cm)
Assignment
446 899 949 1047 1091 1125
– – – C-C and C-O stretching C-OH bend, acetyl group C4-OH bend and C4-H bend
1205 1328 1372 1406 1660 2904 2933
CH2 twist C-H bend, Amide III C-H bend C-N stretching and C-H deformation C=C and Amid I CH stretching N-H stretching
(a)
Glycerin Raman shift (per cm)
Assignment
414 484 672 818 849 923 975 1055 1465 2886 2939
CCO rock CCO rock – CC stretch CC stretch CH2 rock CH2 rock CO stretch from C-1 (=C-3 CH2 deformation Symmetric CH stretch from CH2 Antisymmetric CH stretch from CH2
(b)
Fig. 2. Comparison between skin reference spectra (gray) and (a) hyaluronic acid spectra (black), (b) and glycerin spectra (black).
TABLE 2. Assignment of vibrational peaks were selected to be used as hyaluronic acid (HA) and glycerin markers in the skin HA Raman shift (per cm) 446 949 1047 1372 1406 2904
Glycerin
Assignment – – C-C and C-O stretching C-H bend C-N stretching and C-H deformation CH stretching
Raman shift (per cm) 484 672 818 923 1055 1465
Assignment CCO rock – CC stretch CH2 rock CO stretch from C-1 (=C-3 CH2 deformation
that takes into account all reference spectra, and each pixel spectrum is represented by a linear combination of the reference spectra, respectively, weighted by a fitting coefficient. For each treatment condition (HAs or glycerin), the fitting coefficients were averaged over three measurements performed on different but adjacent skin sections.
Fig. 3. Light microscopy image of skin section with mapping zone.
In Fig. 4, the reconstructed spectral images permit to see that the signals of HAs with low molecular weight: Renovhyal and Bashyal are present in the skin section, at around 100 lm (full epidermal depth) for Renovhyal and 50 lm of epidermal depth for Bashyal (Fig. 4d and e). Whereas for the HA with high molecular weight:
59
Essendoubi et al. (a)
(b)
(c)
(d)
(e)
Fig. 4. Representation of the permeation and localization of various hyaluronic acid (HA) after fitting by their reference spectra and by comparison to the (a) positive (glycerin treated skin) and (b) negative (water treated skin) controls, (c) Cristalhyal (1000–1400 kDa HA), (d) Bashyal (100–300 kDa HA), and (e) Renovhyal (20–50 kDa HA).
Cristalhyal the permeation does not exceed 25 lm (Fig. 4c). By inspecting the light microscopy images and Raman reconstructed images, we can observe that for all HAs, the major quantity was found in the stratum corneum, around 25 lm from surface. Renovhyal HA is highly present in the deepest epidermis layers of epidermis, whereas Bashyal HA is localized in the superficial layer of the epidermis under stratum corneum. The HA with high molecular weight Cristalhyal is found only in the stratum corneum. These results show that HAs permeation increases with decreasing molecular weight of HA. This is in agreement with other studies demonstrating that low molecular weight HA (50 kDa) was associated with significant improvement in skin hydration and wrinkle depth, which is due to better penetration
60
abilities of low molecular weight of HA (3, 7, 9). Whereas, HA with a high molecular weight (more than 1000 kDa) needs to stay at the surface of the skin where it makes a film permitting the diminution of skin evaporation and limiting interaction with the environmental factors such as temperature, humidity, and UV radiation. In this case, penetration is not required (4, 6, 23). We can see also in (Fig. 4a and b), for control tests, there is no HA signal in negative control, while for positive control the glycerin signal is recovered in the skin section at full epidermal depth (100 lm) reflecting the high diffusion of glycerin through the human skin. Several studies demonstrate a significant improvement in skin hydration, wrinkle depth, and elasticity by using HA-based creams while they do not show any argument of the HA
Human skin penetration of hyaluronic acid
penetration. Furthermore, the penetration of the chemical compounds with a molecular weight higher than 500 Da is more controversial according to the bibliography (3). Compared to other reported data and according to our knowledge, this is the first report Raman micro-imaging to prove permeation and spatial distribution of HA with different molecular weight in human skin. Raman spectroscopy is promising approach for probing exogenous molecules in the human skin. Recent studies using this technique have demonstrated the possibility of tracking cosmetic and pharmaceutical formulations (24). Recently, the spatial distribution of Sodium Dodecyl Sulfate and monitoring of caffeine and Metronidazole in skin by Raman spectroscopy have been described (18, 21, 25).
References 1. Fraser JR, Laurent TC, Laurent UB. Hyaluronan: its nature, distribution, functions and turnover. J Intern Med 1997; 242: 27–33. 2. Nusgens BV. Hyaluronic acid and extracellular matrix: a primitive molecule? Ann Dermatol Venereol 2010; 137(Suppl.): S3–S8. 3. Pavicic T, Gauglitz GG, Lersch P, Schwach-Abdellaoui K, Malle B, Korting HC, Farwick M. Efficacy of cream-based novel formulations of hyaluronic acid of different molecular weights in anti-wrinkle treatment. J Drugs Dermatol 2011; 10: 990–1000. 4. Necas JLB, Brauner P, Kolar J. Hyaluronic acid (hyaluronan). Vet Med 2008; 53: 397–411. 5. Barbucci R, Lamponi S, Borzacchiello A, Ambrosio L, Fini M, Torricelli P, Giardino R. Hyaluronic acid hydrogel in the treatment of osteoarthritis. Biomaterials 2002; 23: 4503–4513. 6. Weindl G, Schaller M, Schafer-Korting M, Korting HC. Hyaluronic acid in the treatment and prevention of skin diseases: molecular biological, pharmaceutical and clinical aspects. Skin Pharmacol Physiol 2004; 17: 207–213. 7. Farwick MPL, Strutz G. Low molecular weight hyaluronic acid: its effects on epidermal gene expression & skin ageing. SOFW Journal 2008; 134: 17–22. 8. Chen WY, Abatangelo G. Functions of hyaluronan in wound repair. Wound Repair Regen 1999; 7: 79– 89.
As a conclusion, the present work demonstrates that Raman imaging combined with chemometric methods is an interesting alternative tool for cutaneous permeation tests. In particular, in the present study, we demonstrated the skin permeability of low molecular weight HA (20–300 kDa) and the impermeability of high molecular weight HA (1000–1400 kDa). An essential aspect of this work was to determine the most pertinent spectral features to detect various HA in the skin section, by using a statistical test (Mann–Whitney P = 0.05). Raman spectroscopy exhibits several advantages. It enables the detection and the localization of exogenous molecules at a micrometric resolution, in a label-free, non-destructive manner and with a limited sample preparation.
9. Farwick M, Gauglitz G, Pavicic T, Kohler T, Wegmann M, SchwachAbdellaoui K, Malle B, Tarabin V, Schmitz G, Korting HC. Fifty-kDa hyaluronic acid upregulates some epidermal genes without changing TNF-alpha expression in reconstituted epidermis. Skin Pharmacol Physiol 2011; 24: 210–217. 10. Duranti F, Salti G, Bovani B, Calandra M, Rosati ML. Injectable hyaluronic acid gel for soft tissue augmentation. A clinical and histological study. Dermatol Surg 1998; 24: 1317–1325. 11. Saari H, Konttinen YT, Friman C, Sorsa T. Differential effects of reactive oxygen species on native synovial fluid and purified human umbilical cord hyaluronate. Inflammation 1993; 17: 403– 415. 12. Moran JW, Bell S. Skin permeation of organic gunshot residue: implications for sampling and analysis. Anal Chem 2014; 86: 6071–6079. 13. 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: 34–39. 14. Jacobi U, Weigmann HJ, Ulrich J, Sterry W, Lademann J. Estimation of the relative stratum corneum amount removed by tape stripping. Skin Res Technol 2005; 11: 91–96. 15. Brown TJ, Alcorn D, Fraser JR. Absorption of hyaluronan applied to the surface of intact skin. J
16.
17.
18.
19.
20.
21.
22.
Invest Dermatol 1999; 113: 740– 746. Tfayli A, Piot O, Pitre F, Manfait M. Follow-up of drug permeation through excised human skin with confocal Raman microspectroscopy. Eur Biophys J 2007; 36: 1049– 1058. Xiao C, Moore DJ, Rerek ME, Flach CR, Mendelsohn R. Feasibility of tracking phospholipid permeation into skin using infrared and Raman microscopic imaging. J Invest Dermatol 2005; 124: 622–632. Tfaili S, Josse G, Angiboust JF, Manfait M, Piot O. Monitoring caffeine and resveratrol cutaneous permeation by confocal Raman microspectroscopy. J Biophotonics 2013; 7: 676–6811. Caspers PJ, Lucassen GW, Wolthuis R, Bruining HA, Puppels GJ. In vitro and in vivo Raman spectroscopy of human skin. Biospectroscopy 1998; 4(Suppl.): S31–S39. Caspers PJ, Lucassen GW, Carter EA, Bruining HA, Puppels GJ. In vivo confocal Raman microspectroscopy of the skin: noninvasive determination of molecular concentration profiles. J Invest Dermatol 2001; 116: 434–442. Pudney PD, Melot M, Caspers PJ, Van Der Pol A, Puppels GJ. An in vivo confocal Raman study of the delivery of trans retinol to the skin. Appl Spectrosc 2007; 61: 804– 811. Draux F, Gobinet C, Sule-Suso J, Manfait M, Jeannesson P, Sockalingum GD. Raman imaging of single living cells: probing effects of noncytotoxic doses of an anti-cancer
61
Essendoubi et al. drug. Analyst 2011; 136: 2718– 2725. 23. Vyumvuhore R, Tfayli A, Duplan H, Delalleau A, Manfait M, BailletGuffroy A. Effects of atmospheric relative humidity on Stratum Corneum structure at the molecular level: ex vivo Raman spectroscopy analysis. Analyst 2013; 138: 4103–4111. 24. Patel A, Bell M, O’Connor C, Inchley A, Wibawa J, Lane ME. Delivery of ibuprofen to the skin. Int J Pharm 2013; 457: 9–13. 25. Mao G, Flach CR, Mendelsohn R, Walters RM. Imaging the distribu-
62
tion of sodium dodecyl sulfate in skin by confocal Raman and infrared microspectroscopy. Pharm Res 2012; 29: 2189–2201. 26. Alkrad JA, Mrestani Y, Stroehl D, Wartewig S, Neubert R. Characterization of enzymatically digested hyaluronic acid using NMR, Raman, IR, and UV-Vis spectroscopies. J Pharm Biomed Anal 2003; 31: 545–550. 27. Mendelovici E, Frost RL, Kloprogge T. Cryogenic Raman spectroscopy of glycerol. J Raman Spectrosc 2000; 31: 1121–1126.
Address: Professor Olivier Piot Faculty of Pharmacy University of Reims Champagne – Ardenne (URCA) 51 rue Cognacq Jay 51096 Reims Cedex France Tel.: +33 3 26 91 81 28 Fax: +33 3 26 91 35 50 e-mail: [email protected]
