Skin Research and Technology 2000; 6: 128-134

Characterization of the physical properties of the stratum corneum by a new tactile sensor

Shingo Sakai¹ , Shu Sasai², Yoko Endo¹, Kohji Matue³ , Hachiro Tagami²and Shintaro Inoue¹

¹ Basic Research Laboratory, Kanebo Ltd., Kanagawa, Japan
² Department of Dermatology, Tohoku University School of Medicine, Sendai, Japan
³ Cosmetics Laboratory, Kanebo Ltd., Kanagawa, Japan

Background/aims:The physical properties of the stratum corneum (SC) change with its water content, which is regulated by the presence of water solutes (natural moisturizing factors) and lipids in the SC, and are considered to be responsible for the induction of desquamation, skin surface roughness, and fine wrinkles. Recently a new type of tactile sensor developed for evaluating the physical properties of objects has been introduced as a simple, non-invasive method to evaluate them; because the data obtained with this sensor have not yet been characterized in detail, we compared them with other physical parameters of the skin.
Methods:A change in resonant frequency (D f) was measured under various levels of pressure applied by the tactile sensor placed on the cheeks of 29 women. We also measured high-frequency conductance that reflects the hydration state of the skin surface, water holding capacity of the SC, transepidermal water loss (TEWL), skin flexibility, skin elasticity, amino acid composition of the SC, and desquamation index of the SC at the same time, and evaluated the correlation between Dfand other physical parameters.
Results:The correlation betweenD fand high-frequency conductance of the SC, and that betweenD fand the water holding capacity of the SC were substantially high. Dfunder high pressure was more closely correlated with the acidic amino acid ratio of the SC. Dfalso showed a correlation with desquamation index for thickness of the SC as well as with skin elasticity, independent of the applied pressure.
Conclusions:BecauseD fdetermined with the tactile sensor correlated with other physical parameters specific to the properties of the SC, such as high frequency conductance of the SC, its water holding capacity, ratio of amino acids, and the desquamation index for thickness, the physical properties of the SC may be adequately evaluated with the measurements of Df.

Key words:tactile sensor - stratum corneum - high-frequency - conductance - elasticity - acidic amino acids

The stiffness and firmness of the stratum corneum (SC) is mainly influenced by its water content, and a number of studies have shown that the amount of water contained in the SC is dependent on natural moisturizing factors (NMF) (1, 2) and its intercellular lipids (3-5). Moreover, it was recently reported that the flexibility of the SC is controlled by the amount of amino acids, constituents of the NMF in the SC (6). NMF (7-9) and intercellular lipids of the SC (10, 11) vary quantitatively and qualitatively under various skin conditions. Therefore, the elasticity and flexibility of the SC should change with such physical conditions of the skin as skin roughness and formation of wrinkles. However, a simple, non-invasive method to evaluate the physical properties of the SC has not been established yet. The tactile sensor used in the present study is a device that has been developed for evaluating the stiffness of objects (12, 13). This device determines changes in the resonant frequency ( Df) that occur when a vibrating probe comes into contact with an object. The developers of the tactile sensor reported that skin stiffness can be evaluated by measuring Df, becauseD fis correlated with the spring constant k (12). They also noted with this device that skin became softer immediately after topical application of water or a moisturizer and that the softness of skin persisted for 30 min after treatment with water. These findings seem to suggest that the physical properties of the SC can be assessed with this device, rather than those of the entire skin. On the other hand, Lindahl et al. reported that Dfindicated decreased elasticity of the skin with aging (12), suggesting that Dfreflects the physical properties of the entire skin, including the dermis. Sasai et al. (13) recently reported that changes in Dfwere independent of the conductance of the SC after topical use of some moisturizers and in hypertrophic scars. Therefore, a rigorous and much more analytical investigation of the functional characteristics of the sensor in evaluating physical properties of skin is needed. In the present study, we used the tactile sensor and other devices to evaluate various skin parameters at the same skin site.

Subjects and Methods

Subjects
We evaluated the properties of the cheeks of 29 female subjects by various techniques. The subjects consisted of one woman in her 20s, 15 women in their 30s, 6 women in their 40s, and 7 women in their 50s. The temperature and relative humidity of the instrument measurements were kept at 23 ^oC and 60%, respectively.

Tactile sensor
We used a Venustron tactile sensor (Axiom Inc., Fukushima, Japan) an improved version of the tactile sensor developed by Omata (13). With this improved tactile sensor, the depth and pressure exerted by the sensor as well as Dfof the skin can be determined at certain intervals while the sensor is pressed and released from the skin at a constant rate. The resonant frequency of the sensor was 50 Hz, and the maximum depth was 3 mm. Measurements of the skin were performed three times at the same site, and Dfvalues (mean) at pressures of 2, 5, 7, and 10 g were determined. Differences in Dfvalues obtained at a pressure of 5 g during the process of pressing and releasing the sensor ( DDf, absolute value) were also determined (Fig.1).

Fig. 1. Hysteresis curve obtained by tactile sensor measurement. The pressure exerted by the probe and Dfwere determined at intervals. Dfwas determined at applied pressures of 2, 5, 7, and 10 g from the beginning of measurements (arrow). DDfwas defined as the difference in Dfbetween the value of the pressing and releasing process of the probe measured at a pressure of 5 g.

Hydration state of SC
The hydration state of the SC was determined by measuring high-frequency conductance of the skin. It was measured with an impedance meter (Skincon-200, IBS Co., Ltd., Hamamatsu, Japan) (14) and determined as an average of three measurements. The water holding capacity of the SC was calculated from the changes in skin conductance after artificial hydration. Distilled water (20 ml) was placed on the skin for 10 s, and removed by absorption with cotton. After removal, skin conductance was estimated at an interval of 30 s. The water holding capacity was calculated as follows:

Water holding capacity (m S_*min)
= (C₀+ 2C₃₀+ 2C₆₀+ 2C₉₀ + C₁₂₀) / 4

where C_xis the skin high-frequency conductance atx seconds after artificial hydration.

TEWL
Transepidermal water loss (TEWL) (15) was measured with a Tewameter (TM210, Courage+Khazaka, Cologne, Germany). Measurements were performed for 2 min and the mean value determined for the last stable 30-s period was used as the TEWL.

Elasticity of the skin
The elasticity of the skin was measured with a Cutometer (SEM474, Courage+Khazaka, Cologne, Germany) (16). The skin was stretched under the constant vacuum pressure (3 X10⁴Pa) of the probe for 5 s. The total elongation was related to extensibility. The immediate retraction to total elongation ratio at 0.1 s after the final point of vacuum application was utilized as elasticity. Each value was an average of three measurements.

Analyses of artificially removed SC samples for exfoliating SC pattern
The SC was collected by applying a piece of adhesive tape (Kanebo Co. Ltd., Tokyo, Japan) under pressure of 200 g/cm ²for 3 s. Transmission images of the SC removed were taken by scanning the adhesive tape using a film scanner (LS3500, Nikon Co. Ltd., Tokyo, Japan). The light transmission images of the stripped SC were measured by an image analyzer (Nexus 6800, Nexus Inc., Tokyo, Japan). Using the intensity and distribution of the image obtained on a gray-scale level, the following four parameters were determined (17): desquamation index for amount (DIA) and desquamation index for thickness (DIT) were evaluated from the amount and thickness of exfoliated SC, respectively; desquamation index for size (DIS) and desquamation index for localization (DIL) were evaluated from the unevenness f the area and localization of the stripped SC, respectively.

Quantification of amino acids in the SC
From pieces of the SC collected with a 2 cmX 2 cm adhesive tape (Plus Inc., Tokyo, Japan), water-soluble fractions were extracted with 10 mM HCl by shaking, and the amino acid content per unit area (pmol/cm ²) was quantified. Amino acid analysis was performed using a Hitachi l-6200 (Hitachi, Ltd., Tokyo, Japan) post-column reaction chromatography system with o-phthalaldehyde, essentially according to the method of Spackman et al. (18). The amino acids detected this was were aspartic acid, threonine, serine, asparagine, glutamic acid, glutamine, glycine, alanine, citrulline, valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, ornitine, lysine, histidine, and arginine. The ratio of acidic amino acids to total amino acids was determined by dividing the peak area of all amino acids. The ratios of basic amino acids (lysine, histidine and arginine), neutral amino acids (threonine, serine, asparagine, glutamine, glycine and tyrosine) and hydrophobic amino acids (alanine, valine, leucine, isoleucine, methionine and phenylalanine) to total amino acids were similarly determined.

Statistics
Probability levels ofP<0.05 andP<0.01 were considered significant when results were analyzed by Student's t-test.

Results

The tactile sensor used in this study can simultaneously determine Df, the depth of the probe, and pressure exerted by the probe and a hysteresis curve with a single round of measurement (Fig.1). Dfwas determined on the cheek applied at pressures of 2, 5, 7, and 10 g, exerted by the probe and we evaluated the correlation between Dfand various parameters of the skin obtained with other instrument measurements. DDf, the difference in theD fvalue obtained between a pressing and releasing process of the sensor under a pressure of 5 g was determined, and we evaluated the correlation between DDfand various parameters of the skin (Fig.1).

Dfand the physical properties of the skin
Table 1 shows the values obtained for the correlation coefficient between Dfand various physical parameters of the skin and those between DDfand the latter (n= 29).

There was a strong correlation betweenD fand high-frequency conductance (r= -0.733 at 2 g pressure), which became more prominent when a low pressure was applied on the probe (Fig. 2). A correlation was also evident between Dfand the water holding capacity of the SC, which was also marked at low applied pressure ( r= -0.571 at a pressure of 2 g), as seen with the correlation between Dfand high-frequency conductance (Fig. 3).D fcorrelated with the elasticity of the skin (r = 0.493 at 2 g pressure), regardless of the pressure levels applied by the probe (Fig. 4). Dfdid not show any significant correlation with extensibility of the skin or with TEWL (Table 1).

On the other hand,DDf did not show any significant correlation with any of the parameters measured in this study (Table 1).

Fig. 2. Correlation betweenD fand high-frequency conductance of the SC (m S). There was a correlation betweenD fand the conductance of the SC at pressures of 2 g (A), 5 g (B) and 10 g (C). **: P<0.01.

Fig. 3. Correlation betweenD fand water holding capacity of the SC (m S*min). The relationship betweenD fand the water holding capacity of the SC was examined at pressures of 2 g (A), 5 g (B) and 10 g (C). *, **: P<0.05 andP<0.01, respectively; N.S.: not significant.

Fig. 4. Correlation betweenD fand the elasticity of the skin (%). There was a significant correlation between Dfand the elasticity of the skin at pressures of 2 g (A), 5 g (B) and 10 g (C). **: P<0.01.

Dfand amino acid content of the SC
There was a significant correlation betweenD fand the ratio of acidic amino acids (Asp+Glu / total amino acids) in the SC, which became more conspicuous at higher applied pressure ( r= 0.596 at 10 g pressure) (Fig. 5).D falso correlated with Asp / (Ala+Asp) ratio (9), which is thought to correlate with the superficial morphology of rough skin (Fig. 6). Dfdid not correlate with the other amino acid ratios (Table 1).

Fig. 5. Correlation betweenD fand the acidic amino acid ratio. The relationship between Dfand the acidic amino acid ratio in the SC was examined at pressures of 2 g (A), 5 g (B) and 10 g (C). The acidic amino acid ratio was determined by dividing the amount f aspartic acid and glutamic acid by the total amino acid content. *, **: P<0.05 andP<0.01, respectively. N.S.: not significant.

Fig. 6. Correlation betweenD fand the Ala/(Ala+Asp) ratio. There was a significant correlation between Dfand the Ala/(Ala+Asp) ratio at a pressure 10 g. **: P<0.01.

Dfand stripped SC
Dfweakly correlated with the heterogeneity of the area of stripped SC (DIS) ( r= 0.411 at a pressure of 10 g, Fig. 7A) and the thickness of exfoliated SC (DIT) ( r= 0.474 at 10 g pressure, Fig. 7B).D fdid not correlate with DIA or DIL (Table 1).

Fig. 7. Correlation betweenD fand the desquamation index of SC. The relationship between Dfand the area of homogeneity of the stratified stripped part (DIS) (A) and that between Dfand the thickness of exfoliated SC (DIT) (B) at a pressure of 10 g are shown. *: P<0.05.

Discussion

In the present study, we found that a change in resonant frequency (detected with the tactile sensor) and Dfcorrelated with various parameters of the SC obtained with instrument measurements. Because stiffer objects show larger values of Df, at these values ofD f, high-frequency conductance, hydration state of the superficial portion of the SC, and water holding capacity of the SC became lower. The correlation between Dfand high-frequency conductance values, and that between Dfand water holding capacity of the SC, became more apparent when a lower pressure was exerted by the probe. These findings indicate that the firmness of the SC is caused by a decrease in its water content and could be evaluated by measuring Df.

Interestingly, we found thatD fcorrelated with the ratio of acidic amino acids in the SC. Jokura et al. (6) reported that elasticity of the SC could be improved by the extrinsic supply of amino acids, excluding acidic amino acids. From these findings, we presume that the acidic amino in the SC increase the firmness of the SC.

Desquamation of the SC will also be affected by its stiffness, which could be surmised from the data showing the correlation between heterogeneity of area of stripped part (DIS) and Dfand that between the thickness of exfoliated SC (DIT) and Df.

Koyama et al. (9) reported that the Ala/(Ala+Asp) ratio became lower with increased morphological roughness of the skin surface. In the present study, we found that the Ala/(Ala+Asp) ratio correlated with Df. Because we found a very high interrelationship between the acidic amino acids to total amino acids ratio and the Ala/(Ala+Asp) ratio ( r= -0.929, data not shown), we think that these two ratios represent similar characteristics of the SC. Both the correlation between the acidic amino acids to total amino acids ratio and Dfand that between the Ala/(Ala+Asp) ratio and Dfdepended on the pressure exerted by the probe (data not shown). Although the reasons for the higher correlation between the ratio of the acidic amino acids and Dfunder the higher pressure applied by the probe are unknown, Dfseems to be affected by the superficial morphology of the SC, because the contact area between the probe and the SC can be increase by applying pressure.

The depth in the skin attained by the tactile sensor is also considered to be an important factor in measurements. Sasai et al. (13) reported that there were cases in which changes in conductance value did not correlate with changes in the Dfvalue, particularly in the experiments with topical applications of various moisturizers. This may be induced from the fact that certain topical agents, such as emollients, improve the condition of the skin surface without affecting the hydration state of the SC.

Generally, the elasticity f the skin is reduced with aging (16, 19) and photoaging (20). These studies (16, 19, 20) mainly evaluated the elasticity of the dermis. In the present study, the elasticity of the skin was significantly correlated with the age of the subjects (data not shown, r= -0.653,P<0.01). As documented by Lindahl et al. (12), our study showed that Dfwas weakly correlated with aging.

Interestingly,Dfwas found to correlate significantly with the elasticity of the skin. However, Dfmay reflect more directly the physical properties of the SC than those of the entire skin when the applied pressure exerted by the probe is low, because correlation between Dfand the elasticity of the skin was independent of the pressure exerted by the probe. The correlation coefficient was low. In contrast, when Dfmeasured at a constant depth of the sensor probe (0.2, 0.5, 1.0 mm) correlated significantly with high-frequency conductance, the observed correlation between Dfand the elasticity of the skin was no longer apparent (data not shown).

The sensor used in the present study can give otherD fvalues, such asDfandDD fat constant depth. In the future, the relationship between these other Dfvalues and the elasticity of the SC will be evaluated.

Dfmeasured by the tactile sensor in the present study was significantly correlated with parameters specific to the SC, such as high-frequency conductance and water holding capacity of the SC, as well as with the acidic amino acids to total amino acids ratio. The tactile sensor is a useful tool for evaluating the physical properties of the SC, above the multi-layered system that consists of the epidermis and dermis.

References