If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
As the outermost barrier of the body, the stratum corneum (SC) is frequently and directly exposed to a pro-oxidative environment, including ultraviolet solar radiation (UVR). Therefore, we hypothesized that the SC is susceptible to UVR induced depletion of vitamin E, the major lipophilic antioxidant. To test this, we investigated (i) the susceptibility of SC tocopherols to solar simulated UVR in hairless mice, (ii) the baseline levels and distribution patterns of tocopherols in human SC, and (iii) the impact of a suberythemogenic dose of solar simulated UVR on human SC tocopherols. SC tocopherol levels were measured by high performance liquid chromotography analysis of SC extracts from tape strippings. In murine SC, overall tocopherol concentrations were determined, whereas in human SC, 10 consecutive layers were analyzed for each individual. The results on SC tocopherols demonstrated (i) their concentration dependent depletion by solar simulated UVR in hairless mice; (ii) a gradient distribution within untreated human SC, with the lowest levels at the surface (α-tocopherol 6.5 ± 1.4 pmol per mg, and γ-tocopherol 2.2 ± 1.3 pmol per mg) and the highest levels in the deepest layers (α-tocopherol 76 ± 12 pmol per mg, and γ-tocopherol 7.9 ± 3.7 pmol per mg, n = 10; p < 0.0001); and (iii) the depletion of tocopherols in human SC by a single suberythemogenic dose of solar simulated UVR (α-tocopherol by 45%, and γ-tocopherol by 35% as compared with controls; n = 6; both p < 0.01). These results demonstrate that the SC is a remarkably susceptible site for UVR induced depletion of vitamin E.
solar simulated ultraviolet radiation (280–400 nm)
Located at the interface between body and environment, the stratum corneum (SC) is frequently and directly exposed to a pro-oxidative environment, including chemical oxidants, air pollutants, and ultraviolet solar light (
). The SC is comprised of a unique two-compartment system of structural, enucleated cells (corneocytes) embedded in a lipid enriched intercellular matrix, forming stacks of bilayers that are rich in ceramides, cholesterol, and free fatty acids (
); however, the underlying mechanisms of UVR induced changes in SC barrier function remain unclear.
Although the SC is a major optically protective element of the epidermis, reflecting approximately 5% of the incident light and absorbing significant portions of ultraviolet C (UVC; 200–280 nm) and UVB (280–320 nm) (
Previously, we found the SC to be quite sensitive to ozone induced oxidative damage in hairless mice. Although we failed to detect depletion of vitamin E following ozone exposure in whole murine skin (
). Therefore, we hypothesized that SC tocopherols would also be a sensitive target for UVR. To test this hypothesis, we investigated (i) the susceptibility of SC α- and γ-tocopherol to solar simulated UV radiation in an animal model (SKH-1 hairless mouse), (ii) the baseline levels of α- and γ-tocopherol and its distribution gradient in human SC, and (iii) the impact of a suberythemogenic dose of solar simulated UVR on α- and γ-tocopherol in human SC.
Materials and Methods
All chemicals used were of the highest grade available. α- and γ-tocopherol standards were a kind gift from Henkel (La Grange, IL).
Permission was granted from the UC Berkeley Committee for the Protection of Human Subjects to obtain human SC using a tape stripping technique before and after UV irradiation; all subjects gave their informed, written consent. Ten healthy volunteers (skin types II and III; mean age 26 ± 3 y) were questioned to ensure that they did not take any medication and had no history of photosensitivity or other current medical problems.
The animal care, handling, and experimental procedures were carried out as described in the animal use protocol approved by Animal Care and Use Committee of the University of California, Berkeley. Hairless mice (SKH-1, females, 7 wk old, Charles River Laboratories, Wilmington, MA) were kept under standard light and temperature conditions. Food (Harlan Teklad Rodent Diet #1846, WI) and water were provided ad libitum.
Ultraviolet light was generated by an Oriel 1000 W solar UV simulator equipped with a 1000 W ozone free xenon lamp (Oriel, Stratford, CT). The energy output of the solar simulator was measured at the level of the skin surface (beam size of 152 × 152 mm) using an IL 443 radiometer (International Light, Newburyport, MA) for the UVB portion and a UVX-36 radiometer (UVX-Radiometers, UVP, San Gabriel, CA) for the UVA portion of the solar simulated UV spectrum. The output was 1.85 mW per cm2, as measured with the IL 443 radiometer, and 5.16 mW per cm2, as measured with the UVX-36-radiometer. For comparison, the irradiances of sunlight measured on a sunny day (July 15 at noon) on the campus of the University of California at Berkeley (latitude 38° N) were 0.5 mW per cm2 (IL 443 radiometer) and 2.2 mW per cm2 (UVX-36 radiometer), respectively. The solar UV simulator set-up included an in-built UVB/UVA dichroic mirror that allows wavelengths from 280 to 400 nm to pass and greatly reduces the VIS and IR output of the lamp. Furthermore, a UVC blocking filter (Oriel) was used to cut off radiation at wavelengths below 280 nm. The resulting radiation spectrum is called “solar simulated ultraviolet radiation” (SSUV), and is displayed in Figure 1. To obtain a UVA spectrum, the UVC blocking filter was replaced by a UVB/C blocking filter (Oriel), and the 280–400 nm dichroic mirror was kept in place. To obtain the UVB spectrum, the 280–400 nm dichroic mirror was replaced by a 260–320 nm dichroic mirror (Oriel), and the UVC blocking filter was kept in place.
Human UVR exposure
To estimate the UV sensitivity of each subject (n = 6), nontanned forearm skin was exposed to increasing doses of SSUV. After 24 h, the MED was determined individually by reading of erythema responses. The solar simulator generated 1 MED in 4–6 min, depending on the subject. An area of 5 × 5 cm of the left upper arm was irradiated with 0.75 MED, whereas the contralateral site on the right upper arm served as control. The total administered doses for 0.75 MED ranged from 1.26 to 1.89 J per cm2 (SSUV, sum of UVB and UVA irradiances). The total irradiance was measured by the IL 443 and UVX-36 radiometers and may be an overestimation, due to overlapping sensitivities.
Murine UVR exposure
Prior to the UV exposure or sham irradiation (controls), mice were anesthetized by an intraperitoneal injection of sodium pentobarbital (50 mg per kg body weight, Nembutal, Abott Laboratories, North Chicago, IL). The murine MED, as determined on the back skin of three hairless mice, was obtained in a 3 min exposure to SSUV, equivalent to a total irradiance of 1.26 J per cm2. As the mice were all of the same age and skin type, this MED value was used for all mice and was not determined individually.
To evaluate a concentration dependency for SSUV treatment, animals were exposed to 0, 0.75, 1.5, and 9 MED SSUV [each level, n = 3 mice; two groups of mice were used, one group for vitamin E and one group for malondialdehyde (MDA) measurements].
In a separate experiment, mice were exposed to a single dose of 2 MED SSUV (6 min exposure, SSUV, 2.52 J per cm2, n = 4 mice). To differentiate between the UVB and UVA effects, a second and third group of mice, respectively, were irradiated for the same time as for the 2 MED SSUV exposure (6 min); however, different filters were used to divide the SSUV spectrum into its UVB portion and its UVA portion. For UVB the 280–400 nm dichroic mirror was replaced by a 260–320 nm dichroic mirror, and the UVC filter kept in place, irradiation was for the same time (6 min, irradiance 0.67 J per cm2, n = 4); for UVA the UVC filter was replaced by a UVB/C blocking filter, and the 280–400 nm dichroic mirror was kept in place, irradiation was for the same time (6 min, irradiance 1.86 J per cm2, n = 4) (see Figure 3). Control mice (n = 12) were exposed to 0 MED (sham irradiation) but treated identically in terms of anesthesia and housing conditions.
Tape stripping of human SC
The skin was not prepared in any way before the irradiation or the tape stripping; each subject was told not to use any skin preparations (moisturizers, etc.) on the day of the experiments. Tape stripping was performed 10 min after the irradiation treatment. All human tape strippings were performed on equivalent locations of the upper arm. Samples of human SC were obtained by tape stripping the skin with 5 cm × 5 cm pieces of an adhesive tape (Highland 3710, 3M, St. Paul, MN) using a standardized protocol. Tapes strips (5 × 5 cm) were smoothly adhered onto the skin, flattened equally three times, and removed gently using moderate and even traction. The resultant SC layers adherent to the tapes appeared by light microscopy to be of uniform thickness. Due to the possibility of surface contamination and to remove surface lipids, the first (uppermost) tape stripping was discarded; subsequent tape strippings were used for analyses. The tape strippings were numbered (1–20), and each pair of consecutive tapes (first and second, third and fourth, etc.) was pooled for vitamin E analysis. The amount of SC was determined by the difference in weight before and after application and immediate removal from the skin. In accordance with an earlier report (
), the weight of SC removed per tape (33 ± 7 μg per cm2; n = 10) was independent of tape strip number, and the resultant cumulative SC weight for 20 tape strippings was 660 ± 140 μg per cm2 (n = 10). Based on the SC weights and the size of the tape, the thickness of SC removed by 20 tape strips was calculated to be 6.6 ± 0.7 μm (n = 10), assuming a density of 1 g per cm3 and a uniform coverage of SC on the tape strip as described (
). No significant differences were observed concerning the cumulative weights of the removed SC between the UV treated and the control site (672 ± 91 μg per cm2 and 660 ± 80 μg per cm2, respectively; both n = 6).
Tape stripping of murine SC
Tape stripping was performed 10 min after the irradiation treatment. Samples of murine SC were also obtained by tape stripping the skin with 2.5 cm × 5 cm pieces of adhesive tape (Supreme Brand, Sekisui TA, Garden Grove, CA). This tape was found most suitable for tape stripping murine skin because its weaker adhesion allowed more homogenous tape stripping of the considerably thinner murine SC. In addition, as opposed to most other adhesive tapes tested, it did not interfere with the electrochemical high performance liquid chromotography (HPLC) detection of vitamin E. The first two tape strippings were discarded, and tape strippings 3–8 were collected and pooled for determination of the vitamin E concentration. Therefore, due to the thin murine SC, only one SC α- and one γ-tocopherol concentration was measured per animal. For estimation of lipid peroxidation, MDA was analyzed in SC adherent to Scotch Superstrength Mailing Tape (3M; 2.5 cm × 5 cm) as described previously (
). Briefly, after weighing, tape strips were transferred to a 50 ml centrifuge tube containing 2 ml phosphate buffered saline with 1 mM ethylenediamine tetraacetic acid, 50 μl butylated hydroxy toluene (1 mg per ml), 1 ml 0.1 M sodium dodecyl sulfate, and 4 ml ethanol, mixed vigorously, and extracted with 4 ml hexane. Ethanol (2 ml) was added to the hexane to precipitate the glue (from the tape adhesive), which was then removed and discarded. The hexane was taken to dryness under nitrogen; the residue was resuspended in 500 μl of ethanol:methanol (1:1). The sample was then injected into the HPLC system (Shimadzu, Kyoto, Japan), consisting of a SCL-10 A system controller, a LC-10AD pump, and a SIL-10 A autoinjector with sample cooler, an Ultrasphere ODS C-18, 5 μm particle size column (Beckman, Fullerton, CA) with an a LC-4B amperometric electrochemical detector with a glassy carbon electrode (Bioanalytical Systems, West Lafayette, IN). The mobile phase was methanol:ethanol 1:9 (vol/vol) with 20 mM lithium perchlorate, at a flow rate of 1.2 ml per min. For measurement of ubiquinol-10, α- and γ-tocopherols, the electrochemical detector was operated with a 0.5 V potential and the full recorder scale at 50 nA.
MDA was detected in SC by reacting chloroform extracts from tape strippings with thiobarbituric acid (TBA), separating the TBA reactive substances by HPLC, and quantitating the MDA-TBA adduct by fluorimetric detection as described previously (
Statistical analyses were carried out using SuperAnova for the Macintosh (Berkeley, CA). Logarithmic transformation of the mouse data was carried out to normalize variances between treatment groups. The following designs were used: (i) 1-factor ANOVA for comparisons between UV irradiated and control mice, (ii) 1-factor ANOVA with repeated measures designed to analyze the comparisons between successive layers of human SC, (iii) 2-factor ANOVA with repeated measures designed to compare the antioxidant contents in the various layers of human SC, exposed or not exposed to UV irradiation. Least square means comparisons were used to evaluate differences between treatments; p < 0.05 was considered statistically significant. All data shown represent means ± SD.
UV dose-dependent depletion of tocopherols and increase in MDA formation in murine SC
Baseline concentrations of murine SC α-tocopherol were 9.3 ± 1.9 pmol per mg wet weight and of γ-tocopherol were 3.6 ± 1.4 pmol per mg, which are in accordance with our earlier report (
). Exposure to a single dose of solar simulated ultraviolet light dose dependently depleted the SC concentrations of α-tocopherol, and, less markedly, of γ-tocopherol. A suberythemogenic dose of 0.75 MED significantly depleted α-tocopherol from 9.3 ± 1.9 to 1.4 ± 0.3 (p < 0.0001; Figure 2a), whereas 9 MED were required to significantly deplete γ-tocopherol from 3.6 ± 1.4 to 1.2 ± 0.2 pmol per mg (p < 0.001; Figure 2b).
MDA levels in sham irradiated control mice were 4.8 ± 0.5 pmol per mg of SC tissue. The MDA levels after SSUV exposures to 0.75 MED and 1.5 MED were 5.0 ± 0.7 pmol per mg and 5.8 ± 0.8 pmol per mg, respectively. Only after exposure to 9 MED was a significant increase of MDA observed (7.1 ± 0.2; p < 0.01; Figure 2c).
Both UVA and UVB deplete α-tocopherol in murine SC
To evaluate whether the destruction of tocopherols in the SC could be attributed to UVA or UVB irradiation, mice were exposed to none, both, or either UVA or UVB light. UVA and UVB irradiation (2 MED) depleted both α-tocopherol from 7.7 ± 0.9 to 0.5 ± 0.4 pmol per mg (p < 0.0001) and γ-tocopherol from 2.4 ± 0.6 to 1.4 ± 0.7 (p < 0.01); however, the effects of either UVA or UVB on the two tocopherols were different. Using filters such that only UVA light irradiated the mice, α-tocopherol was depleted to 2.1 ± 0.6 pmol per mg (p < 0.001), whereas UVB irradiation depleted α-tocopherol to 1.5 ± 1.0 pmol per mg (p < 0.001). Neither UVA nor UVB alone had an effect on murine SC γ-tocopherol Figure 3.
Vitamin E concentrations increase with depth of human SC
α- and γ-tocopherol concentrations were determined in 10 sequentially tape stripped layers (two tape strippings combined for each successive pair of 20 strippings) obtained from each of 10 human subjects. In all subjects, the outermost SC layers contained the lowest vitamin E concentrations; the highest concentrations were found in the deepest layer Figure 4. Compared with the outermost SC (6.5 ± 1.4 pmol per mg, tape strippings 1 and 2), the α-tocopherol contents were significantly higher in tape strippings 7 and 8 (17 ± 10 pmol per mg, p < 0.005) and in all subsequent layers (p < 0.0001) up to and including the deepest layer (76 ± 12 pmol per mg tape, strippings 19 and 20, p < 0.0001). Thus, α-tocopherol concentrations in the deepest layers of SC were 11.7-fold higher than those in the uppermost SC layer.
γ-Tocopherol levels were also lowest in the outermost SC (2.2 ± 1.3 pmol per mg, tape strippings 1 and 2), and were significantly higher in tape strippings 7 and 8 (3.7 ± 1.3 pmol per mg, p < 0.03) and in subsequent layers (9 and 10, p < 0.006; 11 and 12, p < 0.01; 13 and 14, p < 0.0002; all others, p < 0.0001) up to and including the deepest layer (7.9 ± 3.7 pmol per mg tape, strippings 19 and 20, p < 0.0001); however, the gradient of γ-tocopherol was not as steep as that observed for α-tocopherol. γ-Tocopherol concentrations were only 3.6-fold lower in the outermost as compared with the deepest SC layers Figure 4.
Suberythemogenic UV irradiation depletes human SC tocopherols
In six of the 10 human subjects described above, the α- and γ-tocopherol concentrations were measured in tape strippings of both nonexposed (right upper arm, control site) and UV exposed (left upper arm) skin. The nonexposed control SC Figure 5 exhibited similar results as described for α- and γ-tocopherol in Figure 4. The α-tocopherol concentrations ranged from 6.7 ± 1.1 pmol per mg in tape strippings 1 and 2 to 77 ± 14 pmol per mg in tape strippings 19 and 20, which reflects a 11.5-fold increase in α-tocopherol in the deepest layer. γ-Tocopherol levels in tape strippings 1 and 2 were 2.8 ± 1.0 pmol per mg and increased 3.4-fold to 8.2 ± 3.0 pmol per mg in tape strippings 19 and 20.
Following exposure of the contralateral site to 0.75 MED solar simulated UV radiation, a significant depletion of SC tocopherols was observed. Although it appears that both the α- and the γ-tocopherol concentrations were depleted in the UV exposed site compared with the corresponding layer of the control site Figure 5, these differences in α-tocopherol concentrations only reached statistical significance in tape strippings 9 and 10 (p < 0.0007) and in all subjacent layers (each p < 0.0001); and for γ-tocopherol concentrations in tape strippings 7 and 8 (p < 0.04), 9 and 10 (p < 0.008), 11 and 12 (p < 0.03), 13 and 14 (p < 0.01), 15 and 16 (p < 0.02), 17 and 18 (p < 0.002), and 19 and 20 (p < 0.0002). The largest decrease in α-tocopherol in response to UV irradiation was observed in the deepest SC layers (tape strippings 19 and 20), where the UV treated site contained 35 ± 8 pmol per mg α-tocopherol, a 55% decrease as compared with the control site (77 ± 15 pmol per mg). Similarly, the γ-tocopherol concentrations in tape strippings 19 and 20 after UV irradiation were 5.5 ± 2.1 pmol per mg, a 33% decrease as compared with the control site (8.2 ± 3.0 pmol per mg).
The overall human SC concentration of vitamin E in the tape strips was 33 ± 4 pmol per mg for α-tocopherol and 4.8 ± 0.8 pmol per mg for γ-tocopherol. UV light exposure (0.75 MED) depleted SC α-tocopherol by 45% (to 18 ± 6 pmol per mg; p < 0.003) and γ-tocopherol by 35% (to 3.1 ± 1.0; p < 0.001).
This study demonstrates that inherent vitamin E in human SC is relatively depleted on the surface and increases with SC depth. Remarkably, a single suberythemogenic dose of solar simulated UV light (SSUV; 0.75 MED) depleted human SC α-tocopherol by almost 50% Figure 5, and murine SC α-tocopherol by 85% Figure 2. Previously, SSUV doses equivalent to 3 MED or more were necessary to detect a significant depletion of α-tocopherol in whole epidermis and dermis (
). Similarly, no detectable amounts of ubiquinol-10, the human equivalent of this antioxidant, were found in human SC (detection limit: 0.1 pmol), whereas its concentration in full thickness human epidermis is similar to α-tocopherol (
), in murine and human SC the levels of ascorbate were very low as compared with epidermal and dermal tissue (Thiele JJ, Traber MG, Packer L, unpublished observations), which is not surprising because the SC is a very hydrophobic tissue. Thus, SC appears to lack any antioxidants that are known to recycle vitamin E, potentially explaining the susceptibility of SC vitamin E to oxidative stress.
Whereas α-tocopherol in murine SC was significantly depleted after suberythemogenic UVR, the lipid peroxidation parameter MDA was increased only after an unphysiologically high dose Figure 2c. Similarly, in previous studies on the cutaneous effects of ozone we found that the concentration necessary to detect increased MDA formation in the SC was five times higher than the one needed to deplete vitamin E (
). It is not clear, however, if SC lipid peroxidation only occurs after exposure to high UV doses, or if the techniques currently available are not sensitive enough to detect more subtle changes in vivo. In vitro studies examining low density lipoproteins oxidation show that lipid peroxidation occurs when vitamin E is almost completely depleted, and that linoleic acid is depleted during this process (
Essential function of linoleic acid esterified in acylglucosylceramide and acylceramide in maintaining the epidermal water permeability barrier. Evidence from feeding studies with oleate, linoleate, arachidonate, columbinate and alpha-linolenate.
). Thus, minor amounts of lipid peroxidation of polyunsaturated fatty acids may not be detectable by current methodologies, yet may have severe pathophysiologic consequences.
Remarkably, a vitamin E gradient was found in untreated human SC with lowest tocopherol concentrations at the surface and highest tocopherol concentrations in the deepest SC layers Figure 4. In human epidermis, the ratio of α- to γ-tocopherol is about 10 to one (
). This is in accordance with the α- and γ-tocopherol ratio we have found in the deepest SC layer, with 10-fold higher α-tocopherol concentration compared with the γ-tocopherol concentration Figure 4. The tocopherol gradient in human SC appears to reflect a depletion of tocopherols in surface layers. This seems likely for the following reasons: (i) the irradiance of UVR at the absorption maximum of α-tocopherol (around 295 nm) is highest at the outer surface and decreases within the SC (
), and thus is inversely correlated with the α-tocopherol concentrations within the SC. Because the presence of oxygen is needed for the induction of UVR induced generation of reactive oxygen intermediates, this might contribute to the increasing α-tocopherol depletion towards outer SC layers. (iii) The percutaneous penetration of most molecules, among them noxious, oxidizing xenobiotics, follows a gradient within the SC with highest concentrations in the outer layers (
). Oxidation of α-tocopherol by such exogenous oxidants would therefore also be higher in the outer SC. Besides its protection against lipid peroxidation, vitamin E is suggested to stabilize lipid bilayer structures (
) demonstrated that murine SC α-tocopherol is very susceptible to exposure of ozone, a strong oxidant and a major air pollutant. Interestingly, in murine SC, α-tocopherol, as compared with γ-tocopherol, was more readily depleted by ozone, which is in accordance with the effects of SSUV on murine SC tocopherols demonstrated in this study. Furthermore, the in vivo antioxidant activity of α-tocopherol is thought to be higher than that of γ-tocopherol (
). In theory, the observed SC tocopherol depletion could be caused either directly, by absorption of short wave UVB, and/or indirectly, by excited state singlet oxygen or reactive oxygen intermediates that are generated by photosensitizers upon UV absorption also in the UVA range. Our results indicate that both mechanisms may be relevant, because either UVB or UVA deplete murine SC α-tocopherol Figure 3. The absorption maxima of both tocopherols lie close together, which suggests that mechanisms other than direct photodestruction by UV, e.g., free radical scavenging properties, may account for this difference in their susceptibilities to UV mediated depletion.
Pathophysiologically, the findings of this study may have relevance with respect to (i) the optical barrier function of the SC, (ii) the physical barrier function of the SC, and (iii) signaling cascades that are initiated in the SC and extend into adjacent epidermal and possibly dermal layers. Recent evidence suggests that α-tocopherol protection against UVB induced lipid peroxidation is a consequence of both its ability to scavenge peroxyl radicals and its UVB absorbance resulting in a sunscreen effect (
). Both mechanisms would provide photoprotection in adjacent epidermal layers. Hence, the presence of α-tocopherol in the human SC and its distribution gradient appears to be of major relevance, especially because its absorption maximum is very similar to the action spectrum for UVR induced erythema and carcinogenesis of hairless mice with a flat peak at 293 nm (
). In addition to its antioxidant properties, α-tocopherol was reported to act as a human skin penetration enhancer, resulting from its interactions with the gel phase interstitial lipid region of the SC (
). Consequently, depletion of inherent SC α-tocopherol could affect the barrier function of human SC.
This study demonstrates the specific distribution of inherent vitamin E in the human SC, and its remarkably high susceptibility to suberythemogenic UVR, leading to its depletion prior to visible skin reactions occurring. Thus, vitamin E depletion in the stratum corneum is an early and sensitive in vivo marker of SSUV induced photo-oxidation. Its pathophysiologic relevance, however, needs to be further investigated.
Helpful discussions with Dr. F. Dreher (UCSF) are gratefully acknowledged. N. Espuno and J. H. Choi provided excellent technical assistance. J.J. Thiele (Th 620/1–1) was supported by a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft. Research was in part supported by a grant from Unilever Research U.S.A.
Tissue optics and photoimmunology.
in: Parrish J.A. Photoimmunology. Plenum Medical,
New York1983: 73
Essential function of linoleic acid esterified in acylglucosylceramide and acylceramide in maintaining the epidermal water permeability barrier. Evidence from feeding studies with oleate, linoleate, arachidonate, columbinate and alpha-linolenate.