2024 年 47 巻 5 号 p. 997-999
Patch tests are often used in safety evaluations to identify the substance causing skin irritation, but the same substance can sometimes give positive or negative results depending on the test conditions. Here, we investigated differences in the skin penetration of two test compounds under different application conditions. We studied the effects of the anionic surfactant sodium dodecyl sulfate (SDS) and the nonionic surfactant polysorbate 80 (PS) on skin penetration of the preservatives methylisothiazolinone (MT) and methylchloroisothiazolinone (MCT), which are used in cosmetics such as shampoos. The skin permeation of MT was enhanced by SDS but was unchanged by PS. Skin impedance decreased in the presence of SDS whereas PS had the same effect as the control aqueous solution, suggesting that SDS reduction of the barrier function of skin affects the permeation of MT, a hydrophilic drug. Application of a mixture of MCT and MT in the presence of SDS did not affect the skin permeation of MCT whereas the permeation of MT was enhanced by SDS, indicating that the skin permeation of MCT is less affected by SDS than is MT. Thus, attention should be paid to the possible effect of co-solutes, especially hydrophilic drugs.
Preservatives used as additives in cosmetics often cause skin irritation, and which preservatives can be used and their maximum concentrations are stated in the Standard of Cosmetics.1) The preservatives methylisothiazolinone (MT) and methylchloroisothiazolinone (MCT) are used mainly in rinse-off products such as shampoos. Patch tests are used to determine possible adverse events,2) but the results can vary depending on the test conditions. Shampoos contain several different surfactants at relatively high concentration (20–40%)3) and thus shampoos that might cause irritation are typically diluted about 100 times, thereby also diluting the MT and MCT in the formulation 100 times. Consequently, their concentration is too low to allow the detection of adverse events, making it difficult to identify the causative agent. Therefore, the patch test is generally performed without surfactants, even though surfactants can alter skin transfer and enhance the skin permeation4,5) of some drugs. Here, we investigated the effect of surfactants on the skin penetration of MT and MCT to provide a reference for setting the basic conditions of the patch test.
MT (MT > 95%) was obtained from Sigma-Aldrich (St. Louis, MO, U.S.A.). A solution of MCT and MT (MCT + MT > 14%, with a ratio of MCT/MT > 2) was obtained from Fluorochem (Hadfield, U.K.). MCT (>98% pure) was obtained from Toronto Research Chemicals (Toronto, ON, Canada) for constructing a standard curve of MCT. Sodium dodecyl sulfate (SDS) was chemical grade from Tokyo Kasei Kogyo (Tokyo, Japan) and polysorbate 80 (PS) was obtained from NOF Corporation (HX2, Tokyo, Japan).
Preparation of Test SolutionMT was dissolved in purified water at a concentration of 0.2%, then diluted using purified water, 2% SDS solution or 2% PS solution to provide 0.1% MT solution (control), 0.1% MT/SDS1% or 0.1% MT/PS1%, respectively. The MCT and MT solution from Fluorochem was diluted 50-times to obtain an approximately 0.2% MCT solution, which was then diluted using purified water or 2% SDS solution. Because of detection sensitivity in HPLC, concentration was set higher than upper limit allowed to add in cosmetics (0.01%).1)
In Vitro Skin Permeation TestFrozen Yucatan micropig skin (Skin set, Jackson Laboratory Japan, Yokohama, Japan) was thawed at room temperature and fat and subcutaneous tissue was removed. The skin was stripped twice using adhesive tape to clean its surface, then mounted in a modified Frantz-type cell (area, 1.1 cm2; receptor fluid, about 17 mL of pH 7 phosphate buffered saline) and saline was applied in donor cell for 30 min to hydrate the stratum corneum. The impedance was measured using an impedance meter (Asahi Techno Lab, AS-TZ1, Yokohama, Japan), then the saline solution was removed, and 0.5 mL of test solution was added to the donor cell. Receptor fluid (0.5 mL) was taken at predetermined times and replaced with the same volume of fresh receptor fluid. After 24 h of sampling, the impedance of the skin was measured again and the skin was removed from the cell. The skin was stripped twice, then the epidermis and dermis were separated by heating. The epidermis and dermis were soaked in 2 mL of acetonitrile to determine the amount of MT and MCT by HPLC.
HPLC MethodThe HPLC system comprised LC-10AD pumps, a SPD-6A UV-visible absorbance spectrometer, an SIL-20A autosampler, (Shimadzu Corporation, Kyoto, Japan), and a Chromato-PRO chromatogram data processing system (Runtime Instruments Co., Ltd., Kanagawa, Japan), together with an L-column2 ODS (5 µm, 4.6 × 150 mm, Kanto Chemical Co., Ltd., Tokyo, Japan) and a mobile phase of acetonitrile:water (15 : 85) at a flow rate of 1.0 mL/min. The detection wavelength was 274 nm and the injection volume was 20 µL. The MT and MCT concentrations were calculated from the peak area using the prepared calibration curves.
Statistical MethodsThe results are shown as mean and standard deviation for at least three independent experiments. The t-test was used to compare two experimental conditions, and Fisher’s t-test was used after ANOVA for three or more experimental conditions. A p-value less than 0.05 was considered to indicate a significant difference.
MT and MCT are often added to shampoos comprising a variety of surfactants at relatively high concentration. We examined the addition of SDS, a typical anionic surfactant, and PS, a nonionic surfactant. Although the surfactant concentration in a typical shampoo is high, a 1% solution of surfactant was used here because critical micelle concentration of SDS is about 0.2%, so SDS formed micelles, 1% SDS is sometimes used in patch tests and preliminary experiments with 2% SDS showed water accumulation under the epidermis, indicative of severe skin irritation.
Figure 1 shows the amounts of MT 24 h after application in the epidermis, dermis and receptor fluid. The amount of MT in the presence of SDS was twice that of the control in epidermis, and 9-times that in dermis and receptor fluid. The increase of MT in dermis and receptor was higher than that in epidermis probably because an increase in the diffusion coefficient of the stratum corneum. The results obtained using PS were not significantly different from the control.
□; Control, ■; 1%SDS, 
; 1%PS. *Significantly different (p < 0.05) from control (without surfactant).
SDS irritates the skin,6) so we measured the impedance of the skin 24 h after application (Fig. 2). The impedance before the start of the experiment was more than 10 kΩ·cm2,7) but after 24 h, a decrease was observed under all the experimental conditions tested. In particular, the impedance was significantly lower after application of the 1% SDS solution compared to aqueous solution, indicating that SDS affects the barrier function of the skin. A hydrophilic drug shows high permeation through skin with low impedance7) and thus may have affected the skin permeation of MT.
□; Control, ■; 1%SDS, 
; 1%PS. *Significantly different (p < 0.05) from control (without surfactant) in impedance and amount permeated.
MCT is supplied as a solution with MT and thus a mixture of MCT and MT solution was used to examine the effect of SDS on the skin permeation of MCT. The amount of MT permeation was also measured. The concentration of the MCT/MT mixture was measured as 11.7% MCT and 3.7% MT, so the applied concentrations were 0.117 and 0.037%, respectively.
Unlike MT, SDS did not cause any increase in the amount of MCT in the skin or permeation (Fig. 3(a)) whereas the effect of SDS on MT using the MCT/MT mixture was the same as that observed for SDS on MT alone (Fig. 3(b)). The impedance was reduced to approximately 2 kΩ·cm2, indicating that SDS had no significant effect on the skin permeation of MCT, despite SDS altering the skin barrier function. Table 1 shows the properties of MT and MCT calculated by the simulation software ADMET Predictor® 10.4. MCT solubility in water is one-tenth, and its octanol/water partition coefficient is about three times higher than that of MT. For lipophilic drugs there is no correlation between barrier function measured by impedance and skin permeation.7) Thus, the effect of SDS on the barrier function of the stratum corneum does not affect the skin permeation of MCT, perhaps due to the nature of MCT.
□; Control, ■; 1%SDS. The concentration of MCT and MT in the applied solution was 0.117% and 0.037%, respectively. *Significantly different (p < 0.05) from control (without surfactant).
| Characteristics | MT | MCT |
|---|---|---|
| Molecular weight | 115 | 150 |
| Solubility in water | 65 mg/mL | 6.75 mg/mL |
| Partition coefficient | 1.32 | 4.47 |
| Skin permeation coefficient | 5.6 × 10−7 cm/s | 17.6 × 10−7 cm/s |
The skin concentration and permeation of a drug is correlated with the drug concentration in the donor phase, therefore, the skin penetration of MCT and MT were converted to 0.1% to compare their drug properties (Table 2). Without SDS, the permeation of MCT was 8 times higher than that of MT in the absence of MCT. SDS enhanced the permeation of MT to a degree similar to that of MCT. Both with and without SDS, MT permeation was significantly higher than that in the presence of MCT. The reason is not clear, but the lack of significant difference in the amount in the skin suggests that the coexistence of MCT may have increased the diffusion coefficient in the skin.
| Epidermis | Dermis | Receptor | |
|---|---|---|---|
| Control | |||
| MCT | 3.3 ± 0.5 | 21.3± 2.4 | 122.8± 4.1 |
| MT with MCT | 4.2 ± 0.6 | 9.2± 1.5 | 40.1± 5.5b) |
| MT without MCT | 3.0 ± 0.6 | 6.2± 2.2 | 14.8± 4.7 |
| SDS1% | |||
| MCT | 5.3 ± 1.6 | 27.4 ± 10.3 | 139.4 ± 13.1 |
| MT with MCT | 6.1 ± 1.8 | 44.5 ± 13.2a) | 189.4 ± 20.7a,b) |
| MT without MCT | 6.1 ± 1.3 | 58.5 ± 11.9a) | 120.5 ± 25.9a) |
a) Significantly different from control. b) Significantly different from without MCT.
A shampoo that might cause irritation is reported as a negative finding whereas the PATCH TEST PANEL® (S) isothiazolinone mix finding is considered positive.8) Shampoo diluted 100 times to prevent severe irritation of surfactant might give a negative finding because the MT and MCT concentrations are below 0.1 µg/mL and thus the amount of MCT and MT transferred to the skin is very low, given that skin penetration is proportional to the concentration in the vehicle. Importantly, MT is affected by the presence of co-solutes, the result of patch test might different with test condition. The results of this study may help to ensure appropriate testing especially with regard to the effect of co-solutes.
We would like to thank Ms. Fumiko Shiotani for her assistance with the experiments in this study. This study was supported by Japan Agency for Medical Research and Development, Research on Regulatory Science of Pharmaceuticals and Medical Devices.
The authors declare no conflict of interest.
