The stratum corneum, as the outer skin layer, plays an important role in biological defense mechanisms, such as protection from bacteria and foreign matter intrusion and maintenance of moisture. The rigid arrangement of intercellular lipids, composed of ceramides (CERs), cholesterol (CHOL), free fatty acids (FFAs), and their derivatives, plays a major role in the barrier function of the tissue.1–3) Recent reports have shown that CER [EOS] is associated with the barrier function of the skin, while CER [NS] is related to water retention in the intercellular lipid structure of the stratum corneum.4,5) CERs, CHOL, and FFAs, components of intercellular lipids, form the lipid bilayer, in which the hydrophobic groups face each other.6)
Small-angle X-ray diffraction has shown that there are two types of lamellar structures: a short lamellar structure with a repeat distance of about 6 nm, and a long lamellar structure with a repeat distance of about 13 nm.7–9) The observation of hydrocarbon chain packing by wide-angle X-ray diffraction has reveal hexagonal packing with a lattice distance of about 0.42 nm and orthorhombic packing with lattice distances of about 0.42 and 0.37 nm.9–11) Although the majority of water is held in the stratum corneum of corneocytes, a portion of the water is taken into the intercellular lipid structure of the stratum corneum. The water incorporated into the corneocytes affects the short period lamellar structure, but not the long period lamellar structure.12) Thus, the short period lamellar structure is thought to be play an important role in controlling water content as the moisture content of the skin changes. Furthermore, the formation of characteristics structures of intercellular lipids can indicate phase transitions as a function of temperature.10) Moreover, although the relevance of the barrier function and structure of the stratum corneum has been revealed, the components and microstructure of intercellular lipids have still not been completely clarified. Additionally, it is difficult to analyze information on intercellular lipids alone because the stratum corneum consists of many components, and intercellular lipids account for only about 10% of these components.10,13)
Analysis of intracellular lipids is generally carried out using IR and X-ray diffraction. Additionally, to consider the interactions between intercellular lipid microstructures and permeation enhancers, synchrotron X-ray diffraction and Fourier transform IR (FTIR) measurements can be performed. In many studies, l-menthol is used as a model permeation enhancer because this compound is widely used as an additive in patches and is convenient and safe.14) The mechanisms through which l-menthol enhances drug permeation involves increasing the distribution of the drug on the skin surface, promoting diffusion of the drug in the skin, and increasing intercellular lipid fluidity.15–17) In addition, drug permeation can be altered when the intercellular lipid structure changes,15) thereby affecting the molecular interactions between intercellular lipids and permeation enhancers. Clarification of the effects of permeation enhancers on the intercellular lipid structure of the stratum corneum is expected to contribute to the development of new transdermal drug delivery systems. In previous studies, the effects of l-menthol on intercellular lipids in the stratum corneum have been evaluated using synchrotron X-ray diffraction, differential scanning calorimetry (DSC), attenuated total reflection (ATR)-FTIR. However, l-menthol is a solid at room temperature and is dissolved in a water–ethanol solution, which is then applied to the stratum corneum.18) Therefore, it is difficult to analyze the effects of l-menthol itself without the presence of water and ethanol.
Accordingly, in this study, we used vaporized l-menthol and examined the direct effects of l-menthol on intercellular lipids in the stratum corneum in molecular level.
Experimental
MaterialsCER [EOS] (N-(30-octadecanoyloxy-heptacosanoyl) sphingosine) was purchased from Evonik Goldschmidt (Essen, Germany). CER [NS] (N-octadecanoyl-D-erythro-sphingosine) was purchased from MATREYA LLC (PA, U.S.A.). CHOL, palmitic acid (PA), and trypsin were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). l-Menthol was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All other chemicals used were of reagent grade.
Preparation of Hairless Mouse Stratum Corneum SheetsHairless mice skin (HOS-HR-1, 8 weeks old; Sankyo Lab Service, Tokyo, Japan) were soaked in 0.1% (w/w) trypsin in phosphate buffer solution (pH 7.4) and stored under refrigeration for 12 h. The stratum corneum was separated from the abdominal region at 37°C for 6 h. The separated stratum corneum was washed in purified water and dried in vacuo for 24 h. All procedures involving animals and their care complied with the regulations of the Committee on Ethics in the Care and Use of Laboratory Animals of Hoshi University.
Amount of l-Menthol in Lipids Isolated from the Stratum Corneum of Hairless MiceSamples of the stratum corneum (2.5 mg) were placed in 1-mm glass capillaries, and vaporized l-menthol was applied for 3 h. The amount of l-menthol in the stratum corneum lipids was analyzed using a GC-17 instrument (Shimadzu Co., Kyoto, Japan). A USB120937H capillary column (30 m×0.2 µm; Agilent Technologies, CA, U.S.A.) was used with nitrogen as the carrier gas. The column oven, injector and detector temperature were maintained at 100, 250, and 300°C, respectively. The flow rate was 1.9 mL/min and split ratio was 1 : 17.
Preparation of the Lipid ModelLipid models using CER [EOS], CER [NS], CHOL, and PA were prepared as reported previously.19,20) The formulation of the lipid model is shown in Table 1. Briefly, the designated amounts of lipids dissolved in chloroform–methanol (2 : 1) were transferred into a flask, and the solvent was removed by evaporation at room temperature under a stream of nitrogen. This procedure results in the formation of a thin lipid film on the inside wall of the flask. Acetate buffer (10 mL, pH 5.0) was added to the flask, and the lipids were hydrated for 30 min. The suspension was sonicated for 5 min at above the phase transition temperature.
Table 1. Formulation of the Lipid Model Composed of CER [EOS], CER [NS], CHOL, and PA
| Molar ratio | CER [EOS] | CER [NS] | CHOL | PA |
|---|
| CER[EOS]/CHOL/PA | 1.0 | — | 1.0 | 1.0 |
| CER[NS]/CHOL/PA | — | 1.0 | 1.0 | 1.0 |
| CERs/CHOL/PA | 0.5 | 0.5 | 1.0 | 1.0 |
| CERs/CHOL0.5/PA | 0.5 | 0.5 | 0.5 | 1.0 |
| CERs/CHOL1.5/PA | 0.5 | 0.5 | 1.5 | 1.0 |
The DSC measurements were made with a Thermo Plus DSC-8230 (Rigaku Co., Tokyo, Japan), with heating scans at a rate of 10°C/min. The lipid models were placed in an aluminum pan (Rigaku Co.). The transition temperatures were determined as the peak top of the endothermic transition peaks.
Synchrotron X-Ray Diffraction MeasurementThe measurement of small- and wide-angle X-ray scattering profiles of lipid models was carried out at the Photon Factory BL6A in the High Energy Accelerator Research Organization (Ibaraki, Japan) and BL40B2 at SPring-8 (Hyogo, Japan). The wavelength (λ) of the X-ray beam was about 0.151 and 0.1 nm, and the sample-to-detector distances were about 1030 mm, and 540 nm for Photon Factory BL6A, and BL40B2 at SPring-8, respectively.
The reciprocal spacing S=1/d=2λ sin(2θ/2), where 2θ is the scattering angle and d is the lattice distance, was calibrated using the lattice spacing of a silver behenate and a plumbum stearate crystal at room temperature. Capillary tubes with a diameter of 1 mm containing the prepared stratum corneum or lipid models were sealed using a gas burner and placed in the sample holder of the X-ray diffractometer. The sample temperature was controlled using DSC (FP-99; Mettler-Toled, Tokyo, Japan).
ATR-FTIR MeasurementATR-FTIR measurements were carried out using an FTIR-4200 spectrometer fitted with ATR PRO (JASCO International, Tokyo, Japan). All spectra were obtained as an average of 210 scans recorded from 4000 to 1000 cm−1 at 4 cm−1 resolution. Spectral analysis was performed using ORIGIN Pro 9 software (OriginLab Corporation, MA, U.S.A.). Synchrotron FTIR measurement was performed to obtain IR spectrum of lipid models as a preliminary data at the BL43IR in SPring-8 (Hyogo, Japan). They were obtained as an average of 500 scans recorded between 7500 and 550 cm−1 at 2 cm−1 resolutions. The lipid model was placed on a Tallium Bromide-lodide (KRS5) plate, and dried completely.
Statistical AnalysisStudent’s t-test was employed for evaluation of the results. Differences at **: p<0.01, *: p<0.05, were considered significant.
Results and Discussion
Effects of l-Menthol on the Lipid Structures of the Stratum Corneum from Hairless MiceSynchrotron X-ray diffraction of the hairless mouse stratum corneum revealed two diffraction peaks near 2.4 and 2.7 nm−1. The diffraction peak at 2.4 nm−1 was derived from orthorhombic and hexagonal structures (lattice space of 0.42 nm), whereas that at 2.7 nm−1 was derived only from the orthorhombic structures.9–11) Figure 1 shows the time-dependent changes in the wide-angle X-ray diffraction profiles of the stratum corneum treated with vaporized l-menthol. The intensities at 2.4 and 2.7 nm−1 gradually decreased over time. Therefore, l-menthol affected the lateral packing structures of the lipids.
To analyze these changes in detail, Gaussian fitting was performed, and the integrated intensity, lamellar distance, and half-width at half maximum of the stratum corneum were calculated (Fig. 2). Depending on the application time of vaporized l-menthol, the X-ray profiles near 2.40 nm−1 were gradually changed. The integrated intensity (Fig. 2a) was decreased, whereas the lamellar distance (Fig. 2b) and half-width at half maximum (Fig. 2c) were increased. Additionally, the X-ray profiles near 2.7 nm−1 indicated similar behaviors for the 2.4 nm−1 diffraction, except for that of the half-width at half maximum. The half-width at half maximum near 2.7 nm−1 showed minor changes compared with the diffractions of 2.4 nm−1. Because the diffraction of 2.7 nm−1 is very small intensity compared with the diffraction of 2.4 nm−1, it was not observed major changes in the half-width at half maximum near 2.7 nm−1. Therefore, l-menthol affected the intercellular lipid structures and altered the orthorhombic and hexagonal structures in the liquid crystal.
Effects of l-Menthol on the Lamellar StructureSmall angle X-ray diffraction studies on the lamellar structures of the stratum corneum have shown that the lamellar structure can have long (13 nm) or short (6 nm) repeat distances.7–9) Moreover, previous studies have reported that the long lamellar structure plays an important role in mediating the barrier function of intercellular lipids in the stratum corneum.21–23) In order to extrapolate the effects of l-menthol on barrier function, we focused on changes in the long lamellar structure. We analyzed the long lamellar structures and determined the integrated intensity, lamellar distance, and half-width at half maximum (Fig. 3). Immediately after administration of vaporized l-menthol, the integrated intensity (Fig. 3a), lamellae distance (Fig. 3b), and half-width at half maximum decreased (Fig. 3c). These results indicated that l-menthol changed the long lamellar structures into liquid crystal and shortened the lamellar distance.18)
Intercellular lipids are composed of CERs, CHOL, and FFAs. Among them, ceramides comprise most components in the stratum corneum of the intercellular lipids.3,24) In the human stratum corneum, 12 CER subclasses have been identified.25) The length of the hydrocarbon chain is different for each of these CER types. Because of the changes in the longer lamellar structures into liquid crystals, we assumed that the lamellar distance was shortened.
Uptake of l-Menthol by Lipids in the Stratum CorneumAfter synchrotron X-ray diffraction measurement, the amount of l-menthol in the stratum corneum was measured by gas chromatography. l-Menthol was taken up at a concentration of 13.37±1.09 µg/mg in the stratum corneum. The amount of l-menthol incorporated into the stratum corneum was approximately 1% of the total weight of the stratum corneum because the stratum corneum was composed of 90% keratin and 10% lipids.26) Therefore, l-menthol was incorporated into approximately 10% of intracellular lipids within the stratum corneum.
Thermal Profiles of the Lipid ModelsRecently, simple lipid model systems have been utilized to evaluate the microstructures and barrier functions of the intercellular lipids in the stratum corneum.18) These model systems have suggested that the microstructures of the intercellular lipids and drug diffusion rates in the stratum corneum are related to the composition of the intercellular lipids in the stratum corneum. Therefore, simple lipid model systems may be a useful substitute for the intercellular lipids in the stratum corneum because the effects of transdermal absorption enhancers on the microstructure and lipid components of the intercellular lipids in the stratum corneum can be detected with a high sensitivity.18)
In order to evaluate the effects of l-menthol on intercellular lipids, we used lipid models composed of CER, CHOL, and PA. CER [EOS] and CER [NS] were selected for the CER component of the lipid models. CER [EOS] is composed of a long hydrocarbon chain with bound ester groups in the chemical structure. CER [NS] is major CER component in stratum corneum lipids. The formulations of the lipid models are shown in Table 1.
Thermograms of the various lipid models, as determined by DSC measurement, are shown in Fig. 4. The phase transitions of the CER [EOS]/CHOL/PA model occurred at 97.2 and 102.1°C, whereas those of the CER [NS]/CHOL/PA model were observed at 32.5, 56.7, and 62.6°C. These phase transition temperatures were lower than each of the melting points of CER, CHOL, and PA. Therefore, the CER [EOS]/CHOL/PA and CER [NS]/CHOL/PA models were considered to represent eutectic systems.27) The phase transition temperature of CER [EOS]/CHOL/PA was higher than that of CER [NS]/CHOL/PA. This could be explained by the observation that CER [EOS] requires higher energy than CER [NS] during the phase transition. In addition, the lipid model of CERs/CHOL/PA including CER [EOS] and CER [NS] showed phase transitions at 62.7 and 90.7°C. Because the lipid models containing CERs, CER [EOS], or CER [NS] had different phase transition temperatures, complexed interactions were expected to occur in the lipids. Because the thermograms of human stratum corneum had several phase transition temperatures, these lipid models may be useful for mimicking the human stratum corneum.28)
Microstructure of the Lipid ModelIn order to identify the complex changes occurring in lipid models during the phase transition, synchrotron X-ray diffraction measurements were performed at 20°C (Fig. 5). The lipid model of CER [EOS]/CHOL/PA showed three diffraction peaks at S=0.17 nm−1 (5.9 nm), 0.29 nm−1 (3.5 nm), and 0.34 nm−1 (2.94 nm). The CER [NS]/CHOL/PA model had a large diffraction peak at S=0.23 nm−1 (4.35 nm) and a minor diffraction peak at S=0.29 nm−1 (3.45 nm). The CERs/CHOL/PA model exhibited diffraction peaks for both CER [EOS]/CHOL/PA and the CER [NS]/CHOL/PA.
Next, we performed DSC measurements and identified the origin of the diffraction peaks. Figure 6 shows the contour profile of the small-angle region of CERs/CHOL/PA. Diffractions of S=0.17 nm−1 (5.9 nm), 0.23 nm−1 (4.35 nm), 0.29 nm−1 (3.5 nm), and 0.34 nm−1 (2.94 nm) were observed at 20°C. As the temperature increased, a new diffraction was revealed at S=0.25 nm−1 (4.0 nm), and this diffraction peak was shifted to the small angle side with increased intensity. These data suggested that there was strong steric repulsion of the CER hydrocarbon chain as the temperature and length of the lamellar period increased. The increase in intensity as a function of temperature was supported by the observation that lipids excluding CER [NS] were melted, and only CER [NS] was present as a crystal.27) The diffraction peaks at S=0.17 nm−1 (5.9 nm) and 0.34 nm−1 (2.9 nm) were stable structures that were less susceptible to temperature. Furthermore, the diffraction peaks showed similar thermal behaviors. The diffraction of S=0.17 nm−1 (5.9 nm) was considered to be a single-order diffraction peak for S=0.34 nm−1 (2.9 nm). The diffraction peak of S=0.29 nm−1 (3.4 nm) disappeared at about 65°C, consistent with the phase transitions observed by DSC. Thus, these data suggested that the free CHOL molecule, separated from the lipid lamellar structure, was melted.29) Therefore, free CHOL, separated from the lipid bilayer membrane, was present in the lipid model.27) However, structural changes coinciding with the phase transition at 90.7°C by DSC measurement were not observed. In addition, the diffraction peaks that disappear at 90°C in the wide-angle region of CERs/CHOL/PA could not be observed.
Next, we examined the formation of lamellar structures of different lengths. Different lamellar periods of S=0.17 nm−1 (5.9 nm) and S=0.23 nm−1 (4.3 nm) were observed for CER [EOS] and CER [NS], respectively. The differences in lamellar length could be explained by the varying lengths of the hydrocarbon chains of CER [EOS] and CER [NS]. CER [EOS] has a linoleic group (30-carbon chain), and CER [NS] has a stearyl group (18-carbon chain). If the band distance of the CER C–C band lengths in the hydrocarbon chain is 0.127 nm,29) the length of the linoleic group is about 3.81 nm, and a stearyl group can be predicted to have a length of 2.28 nm. When the hydrocarbon chains of CER [EOS] formed a lipid bilayer in the form of a fork body arranged face to face, the lamella period would theoretically be 7.62 nm or more. Similarly, the theoretical value of the lamella period of CER [NS] is greater than 4.58 nm. However, lamellar period of the CERs/CHOL/PA by synchrotron X-ray diffraction was shorter than 5.9, 4.3 nm, and the theoretical value, likely because the hydrocarbon chain forming the bilayer was in the gel phase. Furthermore, because intercellular lipids within the stratum corneum contained a small amount of water, the potential effects of water molecules cannot be ignored.
The phase transition at 90.2°C, as observed by DSC measurement, was thought to be caused by disordering of the lamellar structure at S=0.23 nm−1 (4.3 nm). Because the CER [NS] crystal separated from melted bilayer in the lipid model as a function of temperature, the X-ray diffraction peak did not disappear at 90.2°C. From the results of DSC and X-ray diffraction measurements, the diffraction at S=0.17 nm−1 (5.9 nm) was derived from CER [EOS], whereas that at S=0.23 nm−1 (4.3 nm) was derived from CER [NS]. These lipid models formed two types of lamellar periods in addition to the stratum corneum lipids. Therefore, we concluded that the lipid model could clearly reproduce a part of the intercellular lipid structure of the stratum corneum.
Mode of Action of l-Menthol in the Lipid ModelThe effects of l-menthol on the intercellular lipid structures were examined using synchrotron X-ray diffraction. Figure 7 shows changes in small-angle X-ray diffraction profiles by vaporized l-menthol and the rate of change of the integrated intensity. All diffraction peaks showed decreased integrated intensity values over time with application of vaporized l-menthol. After 90 min of vaporized l-menthol application, the integrated intensity of the lamellar structure with S=0.17 nm-1 (5.9 nm) derived from CER [EOS] was reduced by about 11% compared with the intensity before treatment. In contrast, the diffraction at S=0.23 nm−1 (4.3 nm) derived from CER [NS] was reduced by about 11%. The diffraction peak derived from the CHOL crystal at S=0.29 nm−1 (3.4 nm) was reduced by about 8%. Therefore, l-menthol preferentially modified the longer lamellar structure into liquid crystal, consistent with the results showing that l-menthol induced liquid crystallization from the longest period in the stratum corneum in mice.
The decreases in intensity ratios derived from CHOL and CER [EOS] were similar, suggesting that l-menthol may interact with CHOL crystals. Bouwstra and colleagues reported that the ester groups in the side chains of CER [EOS], having two OH group in sphingosine skeleton, and CER [EOP], having three OH group in sphingosine skeleton, interact with the hydroxyl group of CHOL, stabilizing the lipid bilayer.29) Furthermore, the long-period lamellar structure must have a specific length and an ester group as a side chain.
Effects of l-Menthol on CHOL in the Lipid ModelWhen the CERs/CHOL/PA was treated with vaporized l-menthol, the 5.9 nm period lamellar structure derived from CER [EOS] was liquid crystallized, and the CHOL crystal was melted. We then examined the relationships between CER [EOS] and CHOL and evaluated the effects of l-menthol on the lipid model. The small-angle X-ray diffraction profiles of the lipid model are shown in Fig. 8. The mixing ratios of CHOL for CER and PA were 0.5, 1, and 1.5, respectively. The diffraction peaks of integrated intensity at S=0.17 nm−1 (5.9 nm) and S=0.29 nm−1 (3.4 nm) were increased with CHOL content. However, the diffraction peak of integrated intensity at S=0.23 nm−1 (4.3 nm) was not related to CHOL content. This result suggested that there was an interaction between CHOL and CER [EOS].
Next, we examined the effects of l-menthol on lipid models composed of different concentrations of CHOL. Figure 9 shows changes in the integrated intensity of lamellar distances of 5.9 nm (Fig. 9(a)), 4.3 nm (Fig. 9(b)), and CHOL crystals (Fig. 9(c)). For the lamellar structure of 5.9 nm derived from CER [EOS], the integrated intensities of the CERs/CHOL/PA and CERs/CHOL1.5/PA were decreased by 10%, whereas the integrated intensity of the CERs/CHOL0.5/PA was not changed. Moreover, for the structure with a lamellar distance of 4.3 nm derived from CER [NS], the integrated intensity decreased only by about 5% for CERs/CHOL1.5/PA. In addition, for the structure with a lamellar distance of 3.4 nm derived from CHOL crystals, the integrated intensities of the CERs/CHOL/PA and CERs/CHOL1.5/PA were decreased by approximately 10%, similar to the structure with a lamellar distance of 5.9 nm derived from CER [EOS], whereas the integrated intensity of CERs/CHOL0.5/PA was not changed. Therefore, l-menthol selectively affected the longer lamellar structure composed of CER [EOS] and weakened the structure or caused liquid crystallization. The long lamellar structure in the human stratum corneum contributes to the barrier function of the stratum corneum, and the short lamellar structure contributes to water retention of the intercellular lipids in the stratum corneum.4) Therefore, l-menthol acts on the long lamellar structure, disrupts the regular arrangement of the intercellular lipid structure, and promotes drug absorption by the skin. These results may explain the phenomenon observed in the actual skin. Thus, our current results suggested that l-menthol penetrated into the intercellular lipids of the stratum corneum and modified the long lamellar structure into liquid crystal. The barrier function of the stratum corneum lipids was decreased, and the drug permeation was increased.
IR Spectrum of the Lipid ModelAlthough X-ray diffraction measurement can provide information on lipid structures, such as the lamellar structure and lateral packing structure, it is difficult to observe the liquid crystal lipid state. Thus, in order to visualize the molecular vibration of the intercellular lipids in the stratum corneum at the functional group level, IR spectroscopy was performed. The IR spectra of the stratum corneum showed various absorption peaks derived from intercellular lipids and keratin in corneocytes. The characteristic absorption peaks observed in the IR spectra of the human stratum corneum are CH2 asymmetric stretching vibrations (ca. 2920 cm−1), CH2 symmetric stretching vibrations (ca. 2850 cm−1), amide I (ca. 1650 cm−1: C=O stretching), and amide II (ca. 1550 cm−1: N–H bending stretching). The CH2 asymmetric and symmetric stretching vibrations are mainly derived from the intercellular lipids, whereas the amide I and amide II peaks are derived from keratin.30–32) The IR spectrum of CERs/CHOL/PA is shown in Fig. 10. Characteristic absorption peaks were observed at 2917, 2850, 1565, and 1417 cm−1. The absorption peaks at 2917 and 2850 cm−1 represented CH2 asymmetric and symmetric stretching vibrations, whereas those at 1565 and 1417 cm−1 represented amides I and II derived from the sphingosine base of CER because CER was the only component of the model that contained amide bonding. The peak positions of the lipid model derived from CH2 asymmetric and symmetric stretching and amides I and II were determined. In addition, those of the lipid model treated with vaporized l-menthol were determined and compared with the normal lipid model (Table 2).
Table 2. Peak Top Positions of CH
2 Asymmetric Stretching, Symmetric Stretching, Amide I, and Amide II before and after Treatment with Vaporized
l-Menthol
Each data represents the mean±S.D. of 3 determinations. **: p<0.01, *: p<0.05.
The peak positions of CH2 asymmetric and symmetric stretching vibrations did not differ as the amount of CHOL was altered in the lipid model. However, the lipid models treated with vaporized l-menthol were all shifted to a higher wavenumber by approximately 1 cm−1. Because the peak top shift toward a higher wavenumber indicated the activity of functional group vibrations, the change represented an increase in lipid fluidity. In this study, the measurement was performed at a resolution at 4 cm−1. If a peak top shift of more than 2 cm−1 was observed, this change could generally be considered significant. Although the peak shift was a slight change of about 1 cm−1, these peak shifts in lipid models treated with vaporized l-menthol were caused systematically. Therefore, these findings suggested that l-menthol caused a peak shift to a higher wavenumber. A previous report showed that the lipid model treated with l-menthol was dramatically altered in response to heat application.18) Although the changes were minor at 20°C, the lipid acyl chains were affected by l-menthol, resulting in increased lipid fluidity.
For amides I and II, the peak top position of the CERs/CHOL/PA and CERs/CHOL1.5/PA shifted to a lower wavenumber compared with the CERs/CHOL0.5/PA. Depending on the CHOL content of the lipid model, there were strong interactions between the hydrophilic groups of CHOL and CER or formation of hydrogen bonds between CER and CER. In amide I, the lipid model of CERs/CHOL0.5/PA treated with vaporized l-menthol shifted to a higher wavenumber by approximately 2 cm−1. In addition, the lipid models of CERs/CHOL/PA and CERs/CHOL1.5/PA shifted to higher wavenumbers by approximately 8 cm−1. In contrast, for amide II, the lipid model of CERs/CHOL0.5/PA treated with vaporized l-menthol significantly shifted to a higher wavenumber by approximately 2 cm−1, and the lipid models of CERs/CHOL/PA and CERs/CHOL1.5/PA significantly shifted to higher wavenumbers by approximately 4 cm−1. Therefore, these data suggested that l-menthol dissociated the interaction between the hydrophilic groups of CHOL and CER. Terpenes affect the CER of sphingosine bases within the lipid bilayer, attenuate the hydrogen bonding of CER, and cause increasing of the lipid fluidity. Thus, there was an increase in the aqueous layer in the vertical direction and a decrease in barrier function in the lateral direction.33) According to these findings, l-menthol acts directly on lipid carbon chains and CER of sphingosine bases, increasing lipid fluidity similar to the results observed in the previous study.18,30)
Inferred Effects of l-Menthol on the Lipid ModelIn the lipid model of CERs/CHOL/PA, CER [EOS] formed a long lamellar structure, and an ester group of the lipid acyl chain of CER [EOS] formed a hydrogen bond with an OH group of CHOL. In addition, CHOL and PA were present in the lipid bilayer containing CER [EOS]. In contrast, CER [NS] formed a short lamellar structure, and CHOL and PA were present in the lipid bilayer containing CER [NS]. Because there was no ester group in CER [NS], CER [NS] could not form a hydrogen bond with CHOL. In addition, the structure facing the sphingosine base was formed for both lamellar structures, and hydrogen bonds were formed in both the vertical and lateral directions.
From ATR-FTIR measurements, we found that l-menthol affected the dissociation of the hydrogen bonding involving the sphingosine base and acted directly on the CER carbon chain. In addition, the results of X-ray diffraction suggested that l-menthol preferentially caused liquid crystallization of the longer period lamellar structure. Therefore, our data supported that l-menthol acted on the both long and short lamellar structures and dissociated the hydrogen bonds within these structures. Because CER [EOS] having a long lamellar structure formed hydrogen bonds with CHOL, CER [EOS] formed more hydrogen bonds than CER [NS]. In addition, fluidity was increased owing to the loss of interaction with CHOL because of the long acyl chain of CER [EOS]. Thus, l-menthol had different effects on CER [EOS] and CER [NS].
