Imaging-based evaluation of lipids in the stratum corneum by label-free stimulated Raman scattering microscopy.

Abstract

Dear Editor, The stratum corneum (SC) plays an important role as a barrier against the external environment. Currently, skin barrier function is usually evaluated by measuring trans-epidermal water loss (TEWL), but TEWL measurement does not directly reveal the internal condition of the SC. Intercellular lipids in the SC play important roles as a barrier preventing moisture evaporation from the body. For example, it has been reported that ceramide, one of the components of intercellular lipids, is reduced in atopic dermatitis, in which the barrier function is disrupted.1, 2 It is also known that organic solvents elute intercellular lipids, resulting in a rough skin condition with reduced barrier function.3 The importance of intercellular lipids in barrier function has become clear through invasive analysis, such as SC sampling. However, if lipids in the SC could be measured noninvasively and simultaneously with TEWL measurement, it would be possible to evaluate the state of the SC barrier in more detail. This could be useful both in basic dermatological research and for clinical research on diseases in which the barrier state is disrupted. In recent years, the importance of noninvasive measurement inside the skin has increased in fields such as dermatological research, aesthetic medicine, and cosmetic research. Raman spectroscopy is one of the most effective methods of noninvasive measurement without the need for a pretreatment process such as fixing or labeling.4 It has been used to measure the water content and absorption of glycerol in human skin.5, 6 For the present study, we focused on stimulated Raman scattering (SRS) microscopy, since it enables the visualization of micro-structures of tissues based on functional-group-specific Raman signals.7, 8 For example, we previously used SRS microscopy to detect changes in the intracellular morphology of keratinocytes in ex vivo skin during differentiation.9 Advantages of SRS microscopy include rapid visualization and component analysis of skin in three dimensions without any necessary processing after skin sample collection, keeping the sample in good condition.10 Ji et al. investigated the diagnostic application of images acquired at SRS microscopy. In mouse brain, SRS microscopy was able to distinguish between tumors and normal tissues by using both 2845 (mainly reflecting lipids) and 2930 cm−1 (mainly reflecting proteins) SRS images. Tumor areas contained many more cells and a lower level of lipid components than normal areas.11 SRS has also been applied to human skin in vivo,12 and handheld SRS microscopes for in vivo skin measurements have been developed.13, 14 We anticipate that noninvasive in vivo observation by SRS with a short measurement time will become widely available in the future, minimizing the burden on the subjects. In this study, we examined the distributions of proteins and lipids in horizontal sections (parallel to the skin surface) of epidermis from subjects of various ages by means of SRS microscopy. We also evaluated the changes of lipid and protein in the SC after acetone treatment based on the signal intensities at 2850 cm−1, which mainly reflects lipids, and 2930 cm−1, which mainly reflects proteins.10, 15-19 For SRS measurements, human skin from the abdomen (n = 7, 31–74 years old, Caucasian females) was purchased from Biopredic International (Rennes, France) via KAC Co., Ltd. (Kyoto, Japan). The skin samples were collected during plastic surgery after informed consent had been obtained according to all applicable French laws including ethics regulations. The samples were refrigerated during transit and until measurement. Details of skin sample preparation and instrument setup are provided in Supporting Information 1. Before SRS measurements, the skin samples were cut into about 1-cm squares, immersed in phosphate-buffered saline (-), mounted on a coverslip, sandwiched with another coverslip, and sealed with enamel. To prepare lipid-extracted skin, a cotton ball soaked in acetone (Wako Chemicals, Japan) was placed on the SC for approximately 1 min, and then the sample was sandwiched between coverslips in the same way as above. Pictures of hematoxylin-eosin-stained skin samples from the 74-year-old with and without acetone treatment are shown in Supporting Information 2. Cellular layers of the SC were slightly disturbed by acetone, but no marked change was seen in living cell layers. The setup of our SRS microscope, which has a Ti:sapphire (Ti:S) laser with a wavelength of 790 nm and a wavelength-tunable Yb fiber laser (1015-1045 nm), is described in Supporting Information 1. SRS images (80 × 80 μm, 500 × 500 pixels, transmission mode) were acquired in the wavenumber range of 2800 to 3100 cm−1 with a step of 3.3 cm−1, corresponding to the CH stretching region.8, 10, 20 Figure 1(A) shows typical spectrum of the SC. The images were acquired parallel to the skin surface. The images were acquired on each focal plane, moving the focus plane from the top surface of the SC toward the dermis layer by 0.5 μm, and analyzed by using ImageJ software and our in-house image analysis software. Intensities from images at 2850 cm−1 representing CH2 symmetric stretching (mainly due to lipids) and 2930 cm−1 representing CH3 symmetric stretching (mainly due to proteins) were obtained (Figure 1(B)). As we reported previously,9 cell morphology in the SC could be visualized by using SRS microscopy, and polygonal cells were observed in the SC (dashed lines in Figure 1(B)). The average values of the signal intensities at 2850 and 2930 cm−1 obtained from all images of the SC layers of 31 and 74-year-old skins were calculated and are shown in Figure 2(A, B). In the SC of the sample from the 74-year-old, the 2850 cm−1 signal, which mainly reflects lipids, was significantly reduced by acetone treatment, but there was no significant change in the 2930 cm−1 signal, which mainly reflects proteins. There was no significant change at either 2850 or 2930 cm−1 in the SC of the sample from the 31-year-old. The average values of the signal intensities at 2850 and 2930 cm−1 from other five donors are shown in Supporting Information 3. The changes of lipid in SC were evaluated in terms of the intensity ratio of the 2850 and 2930 cm−1 signals (2850 cm−1/ 2930 cm−1) calculated for all the samples with and without acetone treatment (Figure 2(C)). The ratio of the skin samples from the 70- and 74-year-old subjects was significantly decreased by acetone. A significant decrease in the 2850 cm−1 signal intensity was also observed in these two skin samples. The intensity ratio of one skin sample from the 31-year-old was decreased by acetone, although no significant change at 2850 cm−1 was observed. The ratio of one sample from the 49-year-old was increased significantly. The intensities at both 2850 and 2930 cm−1 of the skin sample from the 49-year-old were significantly increased (shown in Supporting Information 3). In the skin samples used in this study, most of the sebum component had been removed during the process of disinfecting the skin surface after excision from the donors. We think that changes of signal intensity in 2850 cm−1 primarily reflect intercellular lipid variation. In the SC of the skin samples from subjects over 70 years old, the 2850 cm−1 signal, which mainly reflects lipids, was significantly reduced by acetone treatment, while the 2930 cm−1 signal, which mainly reflects proteins, was hardly affected. The findings for SC of subjects of various ages suggest that skin samples from those over 70 years old may be more easily disturbed by external stimuli. We previously reported that the images at 2850 and 2930 cm−1 obtained by SRS microscopy revealed blurred cell outlines in the SC in skin from subjects over 70 years old, and the morphology of the cornified cells in the SC of old skin was somewhat blurred compared to that of young skin.9 This finding may be related to an age-related increase of the fragility of the SC, resulting in a greater loss of lipids in response to external stimuli. Many studies have shown that intercellular lipids are eluted from human skin by organic solvents such as hexane, ethanol, and acetone, and extraction with such solvents is used to determine the composition and amount of intercellular lipids.3, 21, 22 Studies on barrier function assessment have reported that treating the arm with acetone-soaked cotton balls increases TEWL levels, and subsequent barrier recovery was slower in old skin than in young skin.23 In accordance with this, we found that slight exposure of ex vivo skin to acetone tended to cause a greater decrease of the signal intensity at 2850 cm−1 in skin from subjects over 70 years old, compared to younger skin. These results suggest that lipid elution by acetone may have occurred more easily because of an increase in the age-related fragility of the SC. It is known that intercellular lipids in the SC are disrupted in some skin diseases, and it has been reported that even in healthy skin, intercellular lipids change with age and the barrier function is reduced.24 There is increasing interest in the epidermal function of aging skin due to the worldwide increase of lifespan.25 For example, decreased SC hydration26, 27 and decreased structural proteins of the SC, such as filaggrin and loricrin,28 as well as changes in the amount and composition of intercellular lipids22, 24 have been reported. Our results indicate that SRS microscopy can be a useful tool to evaluate the decrease in the barrier function of the SC in aged skin exposed to external stimuli. Intercellular lipids correspond to the “mortar” portion of the so-called “brick and mortar” structure of the SC,29 and play an important role as components of the skin barrier. Therefore, noninvasive in vivo measurement of the components of intercellular lipids, together with TEWL, is expected to provide important information about the status of the skin's barrier and its function. We believe SRS microscopy could be a valuable tool for evaluation of the SC barrier condition, not only in the cosmetics industry, but also in the clinical diagnosis of skin diseases such as atopic dermatitis and psoriasis. The authors thank Ms. Yuki Koseki and Ms. Midori Nagai for their support with image data analysis. The authors have no conflict of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Supporting Information 1. Methodology. Supporting Information 2. Hematoxylin-eosin-stained sections of 70-year-old skin without acetone treatment (a) and after acetone treatment (b). Scale bars show 50 μm. Supporting Information 3. The average values of signal intensities at 2850 and 2930 cm−1 obtained from all images of SC layers of all donors other than those shown in Figure 2A,B. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.