Printer Friendly

Low doses of repetitive ultraviolet A induce morphologic changes in human skin.

Numerous investigations have documented that exposure of the skin to the ultraviolet B (290-320 nm [UVB]) and UVA (320-400 nm) components of sunlight is the most important cause of cutaneous photodamage, including photocarcinogenesis and photoaging [1-3].

Human skin changes grossly and microscopically in detectable ways over the lifetime of an individual [21. However, exposure to repetitive amounts of solar radiation greatly increases the rate at and the degree to which the skin ages [2-4]. Dermatoheliosis or photoaging is the clinical description of wrinkled. leathery, dyspigmented skin seen in patients chronically exposed to the sun [4]. Photoaging has been induced by chronic exposure to all portions of the solar spectrum including UVA, UVB, and infrared using relatively large doses utilizing animal models [5,6].

Little research has been reported on the effects of controlled repetitive UV radiation and the changes indicative of cutaneous photodamage in human subjects. Additionally, the amount of UVA to which indoor workers are exposed in normal daily activities is considerably less than the levels of irradiation employed in many of the animal studies [5-7]. It is likely a person would be exposed to UVA either filtered through window glass or through many current sunscreens that contain predominantly UVB filters and only partially absorbing UVA sunscreen filters.

One study was designed to deliver amounts of solar-simulating UV and UVA radiation that unprotected skin could be exposed to during relatively brief outdoor sunlight exposure or from UVA filtered through window glass, e.g., non-UV filtering car windows. People using a sunscreen that only partially filters UVA will receive larger amounts of UVA if they spend prolonged periods outdoors. In addition, those individuals receiving tanning salon exposures can receive twice the UVA radiation than the doses utilized in this study (R. Sayre, personal communication).

Materials and methods

Irradiation Equipment

An Oriel 1000-W xenon arc solar simulator was utilized and equipped with a dichroic mirror, a 2-mm Schott WG 345 filter for UVA irradiation, a 1-mm Schott WG 320 filter for irradiation with 290-400 nm wavelengths, and for all irradiations a 1-mm UG 11 was used to attenuate infrared radiation.

The spectral characteristics of the UV apparatus were measured using an Optronics spectral radiometer (R. Sayre, Rapid Precision Testing, Inc.). Using the 1-mm Schott WG 320, there was only [10.sup.9] W/[cm.sup.2]/nm transmission below 290 nm. Using the 2-mm Schott WG-345 filter, there was no measurable transmission below 320 nm.

Output of the xenon arc simulator was measured before each irradiance using an IL 700 (International Light Co.) radiometer peaking at 305 nm for solar-simulating UV. For UVA, a broad absorbing radiometer (Dermalight) with a peak absorption at 360 nm was used. Irradiance was generally 0.05 mW/[cm.sup.2] for solar-simulating UV and 30 mW/[cm.sup.2] for UVA.

Research Volunteers

Normal healthy skin phototype II and III individuals (five male and seven female, mean age 30.9 [+ or -] 3-6) who had no history of previous skin disease, were taking no photosensitizing drugs, and had never exposed the buttock skin to direct sunlight were recruited. These subjects reviewed and signed an informed consent form detailing the investigation prior to enrollment. Buttock skin was examined to confirm the absence of any skin lesions that could interfere with the investigation.

MED Determination

The minimal erythema dosages (MED) were determined on buttock skin for each individual for solar-simulated UV (290-400 nm), as well as UVA (320-400 nm) using the appropriate Schott filters. The MEDs to solar-simulated UV were expressed in mJ/[cm.sup.2] and the MEDs to UVA expressed in J/[cm.sup.2]. They were determined by using a light-proof adhesive-backed foil template with 1-[inch.sup.2] openings that were sequentially uncovered to deliver quantities of UV above and below the expected MED of skin phototype II individuals for solar-simulating UV.

The sites were examined 24 h after irradiation and the MED was determined as the site that showed minimal, uniform perceptible erythema over the entire 1-[inch.sup.2] area. The mean MED for solar-simulated UV was 77 [+ or -] 17 mJ/[cm.sup.2] and 29.2 [+ or -] 1.7 J/[cm.sup.2] for UVA. The suberythemal UVA dose did not produce any visible eiythema. The MEDs were recorded for each individual and a treatment schedule was initiated with twice-weekly repetitive ultraviolet irradiation using the solar simulator and appropriate filters noted above, based on the initial MED for each individual.

Irradiation Protocol

For each volunteer, individualized amounts of solar-simulating UV or UVA, based upon the initial MED determination, were delivered twice weekly for 24 weeks. The irradiations consisted of one MED of solar-simulating UV, range 0.03 to 0.07 J/[cm.sup.2], one MED of UVA, range 25 to 25 J/[cm.sup.2] (high IJVA), or the amount of UVA in one MED of solar-simulating UV, range 3 to 5 J/[cm.sup.2] (low UVA). A non-irradiated control site was also included. To reduce the possibility of site variation, a randomization scheme for the different irradiations was established. Volunteers were irradiated twice weekly with at least 3 d between each irradiation for 24 weeks (i.e., 48 irradiances).

Biopsies and Histologic Stains

Four-millimeter biopsies following 2% plain xylocaine injection were obtained 12, 24, and 36 weeks following the start of irradiation. The 12- and 24-week biopsies were obtained at least 3 d after irradiation. The 36-week biopsies were taken 12 weeks following the final irradiation. The biopsies were placed in 10% buffered formalin, paraffin embedded, and sectioned at 5 microns. They were stained with hematoxylin and eosin for general tissue morphology as well as Gormori's aldehyde fuchsin for elastic tissue and Lillie's ferrous iron for melanin.

Histologic Evaluations

Assessments were made at 10 separate positions for each section, evaluating 10 non-serial sections using 10x and 40x magnification. Histologic measurements were performed on both epidermal and dermal parameters. The number of stratified cell layers was evaluated at 10x by counting the average number of keratinocytes in a vertical direction over 10 fields/section. The stratified cell layer mean for each treatment group, at each time interval, was determined by obtaining the mean for each panelist, then generating the mean for each treatment group.

The number of granular cell layers was determined in the same manner. Stratum corneum thickness and the degree of epidermal dyskeratosis were evaluated by using a four-point scoring system: 0, none; 1, mild; 2, moderate; and 3, severely dyskeratotic or thickened stratum corneum. Epidermal melanization was evaluated on Lillie's ferrous iron-stained sections using a four-point scoring system: 0, none; 1, mild; 2, moderate; and 3. profound melanization. Dermal parameters included evaluation of perivascular inflammatory cell infiltrate and vascular dilation. Dermal parameters were assessed on hematoxylin and eosin-stained sections. Again, scoring was 0, none; 1, mild; 2, moderate; and 3, severe.

Image Analysis Quantification of Elastic Tissue Content

A color image analysis system was employed and consisted of a Videometric 150 Color Image Analysis Software System (Oncor Instrument Systems, San Diego, CA) equipped with a Sony DXC-960MD 3CCD color video camera and a Sony PVM-1344Q Trinitron color video monitor coupled to a Leitz Orthoplan microscope. The system identifies elastic tissue-like staining, based upon its unique color properties, as numerical values. All other numerical values within the field of measurement are ignored.

Elastic tissue-like staining was analyzed at a magnification of 40x. The fields analyzed were inter-follicular regions, manually outlined and defined as the area below the epidermal-dermal function to the mid-reticular dermis, which was the total dermal area visualized at 40x. For each biopsy, 10 non-serial sections were randomly selected and four measurements performed on each section for a total of 40 fields quantified per biopsy. All samples were quantified by the same operator (DPM). Elastic tissue content is expressed as area fraction per cent [+ or -] SD.

Statistical Analysis

The number of stratified cell and granular layers and elastic tissue content were considered to be additive measures and analyzed parametrically. The existence of treatment effect was assessed for each parameter by evaluation by time point using a two-way analysis of variance. All remaining parameters were scored using a visual assessment scale. Because this grading is subjective, these parameters were considered to be ordered categorizations and were analysed using the Friedman's rank sum procedure. All differences reported are at a significance level less than 0.05.



Histologic Assessment of Epidermal Components Shows Altered Morphology

Solar-simulated UV treatment caused an increase in the number of stratified cell and granular cell layers and stratum corneum thickness [Figures 1-3]. This increase was greatest after 24 weeks of irradiation. Twelve weeks after the final irradiation, values were similar to those seen at 12 weeks. For the UVA treatments, the high-UVA group displayed an unexpected response with respect to the number of granular and stratified cell layers. These values were increased at 12 weeks (P<0.05), similar to the control at 24 weeks and showed an increase at 36 weeks. The low UVA elicited the same response for stratified cell layer numbers except that a statistical difference (P<0.05) was seen at 36 weeks. An increase in granular cell layer numbers was seen at all time points but significantly different from the control at 12 weeks only. Both UVA sites showed increases in stratum corneum thickness at all time points; however, the increases in the low-dose UVA were not significant. Solar-simulated UV irradiation resulted in significant (P<0.05) increase in stratum corneum thickness at all time points. All treatments produced increased numbers of dyskeratotic cells although significant (P<0.05) differences were only noted at 36 weeks in the solar-simulated UV and high-UVA groups [Table 1], Increased melanization was seen in all UV-treated groups with the high UVA being different from the control at all time points. The solar-simulated UV produced increases in melanization but no statistical differences were noted [Figure 4]. This appears to be due in part to variability in test subject response to these treatments.

These results suggest that the damage produced by the UVA treatments persists longer than solar-simulate d UV-induced epidermal damage. Most notable were the epidermal changes produced by suberythemal amounts of UVA. These alterations were present after only 12 weeks of twice-weekly irradiation. Although improvement in morphology was seen 12 weeks after the final irradiation, changes were still evident.

Suberythemal UVA Irradiation Produces Dermal Vascular Changes and Inflammation

UVA irradiation resulted in vascular dilation that was significant (P<0.05) at 36 weeks for the low UVA and the high L'VA [Figure 51. All irradiation regimens induced inflammatory cell infiltrate [Figure 6], The high-UVA and solar-simulated UV treatments resulted in marked inflammatory infiltrate at all time points. Interestingly, the suberythemal UVA also caused noticeable vascular dilation and inflammation at all time points. The solar-simulated UV treatment produced significant (P<0.05) vascular dilation at all biopsy intervals, showing a gradual increase in severity with time.




All treatment regimens produced noticeable changes in the dermis that were observed after only 12 weeks of irradiation. These alterations increased in severity with treatment duration and showed only minimal improvement 12 weeks after the final irradiation. Of greatest interest was the ability of suberythemal doses of UVA to produce dermal damage. The dermal changes seen in this group illustrate the potential that even low doses of UVA have in producing dermal photodamage.

UVA Irradiation Resulted in a Decrease in Elastic Tissue Content

Image-analysis quantification, based upon elastic tissue-like staining, showed interesting changes between the irradiated and non-irradiated sites. The solar-simulated UV produced an increase, whereas both the suberythemal and erythemal UVA produced a persistent reduction at 12, 24, and 36 weeks. Suberythemal UVA-irradiated sites showed some recovery in elastic tissue content toward the control at week 36, although the values were still significantly different [Table 2].


The research results described in this manuscript confirm that repetitive irradiation of human skin results in epidermal and dermal photodamage. Previous investigations of repetitive ultraviolet irradiation have been conducted in animal models, in particular the hairless mouse [5-7]. The difference in histologic characteristics between human and hairless mouse skin, especially the much thinner murine epidermis and relative lack of an epidermal pigmentation response to ultraviolet radiation, may make the murine model more sensitive to UV effects than human skin.

We felt it important to perform this investigation in human skin that had not been previously exposed to sunlight. Acute exposure to erythemal amounts of UVA has been shown to be associated with a variety of histologic changes [8], but there has been little research reported documenting the effects of repetitive irradiation with small amounts of solar-simulated UV or UVA on human skin.

We carefully selected the amounts of UV delivered to be compatible with the exposure expected for an individual with an indoor job with brief periods of outdoor activitiy, i.e., driving in a car or taking short walks. Individuals that are outdoors and protected with an efficient UVB sunscreen (sun protection factor 15) would also be exposed to similar amounts of UVA as our suberythemal UVA dose during a 3- to 4-h exposure to Southern California summer sun.

Treatment with solar-simulated UV and UVA resulted in changes in epidermal morphology after only 12 weeks of irradiation and was increased in magnitude after 24 weeks. Following 12 weeks with no treatment, epidermal morphology showed noticeable improvement. Although the dermis responded in a similar manner during the 24-week irradiation period, 12 weeks after irradiation, observable changes were still evident. This suggests that the dermal components are damaged by low doses of both solar-simulated UV and UVA and it recovers at a much slower rate than does the epidermis.

Unexpected were the changes in the skin irradiated with suberythemal doses of UVA. Alterations were noted in both the epidermis and dermis after only 12 weeks of treatment (see Results). The most interesting finding in this and all UV treatments was the effect on elastic tissue. Previous animal UV irradiations have shown changes in elastic fiber structure in animal skin [7]. In our studies, the effects on elastic tissue were determined by computerized image analysis of elastic tissue content. Changes in elastic tissue content were observed in all UV-treated groups, including low UVA, after only 12 weeks of irradiation. Epidermal melanization appears to have played a minor photoprotective role with regards to elastic tissue damage. Recent observations confirmed the damage caused by suberythemogenic UVA to elastic tissue measured by image-analysis studies and by measuring elastin lysosomal deposition [9]. Recent immunohistochemical findings have confirmed our findings that suberythemogenic UVA is detrimental to elastic tissue. Lavker et al. have shown significant lysosomal deposition on elastic tissue of skin irradiated with suberythemogenic doses of UVA [9]. The elastic tissue content as measured by computerized image analysis of elastic tissue-like staining shows a quantitative difference in effect between solar-simulated UV and UVA radiation. The methods utilized to measure dermal elastic tissue content have been reported previously [10,11] and proved reliable in our studies. There was a slight increase in the dermal elastic tissue content with solar-simulated UV and a decrease in both the low and high UVA. We observed no quantitative differences in the inflammatory responses elicited but there may be qualitative differences that potentiate differing effects on elastic tissue; however, we did not investigate this possibility.



Another potential reason for the difference is that UVA may have a more direct effect on elastic tissue leading to a reduction in content. The differing effects on elastic tissue from light from different portions of the ultraviolet spectrum is not known at this time and requires further investigation. Although abnormal elastotic material has been investigated in previous studies of sun-damaged skin [12], the cause of this abnormal tissue is not clear.

Previous studies of repetitive irradiation in animals and humans have failed to use a true solar-simulating apparatus. One shortcoming has been the FS sunlamp light sources utilized in previous human and some animal studies [13-15]. These contain a significant amount of UVC in addition to UVB and UVA, Therefore the relevance of results obtained from studies utilizing these bulbs is unclear for natural solar exposure. In our studies, care was taken to select and measure the spectral output of the solar simulator utilized. Full spectral characteristics were determined for the different filter arrangements to make certain that no UVC was being delivered and sites receiving UVA received only UVA without any relevant contaminant UVB, In addition, previous studies performed on animals [5,7,12-14] may have exaggerated the effects of UV on human skin because mouse skin has a much thinner epidermis and stratum corneum with less pigmenting ability compared to human skin.

The individuals in this study were irradiated twice weekly for 24 weeks. The results reported here suggest that the relatively low levels of UV delivered would be expected to lead to even greater photodamage and photoaging if continued for longer periods of time. Our results confirm the damaging effects of small amounts of solar-simulate d UV and UVA.

The results are particularly compelling with respect to suberythemal UVA, which produced observable photodamage. The public is easily exposed to comparable amounts of UVA even when using many current sunscreens that provide predominantly UVB protection. Our findings clearly demonstrate there is a need for routine daily photoprotection with efficient UVB and UVA absorbing sunscreens and other methods of protection such as photoprotective clothing.

Reprinted with kind permission of the publishers from Journal of Investigative Dermatology, 1995, 105, 739-743.

This work was supported by a research grant from Unilever Research, Edgewater, New Jersey.

We would like to acknowledge Dr Mark Presser for his excellent assistance in conducting the statistical analysis.


[1.] Kligman AM. Early destructive effects of sunlight on human skin. JAMA, 1969, 210, 2377-2380.

[2.] Lavker RM. Structural alterations in exposed and unexposed aged skin. J Invest Dermatol, 1979. 73. 59-66.

[3.] Know JM Cockerell EG, Freeman RG. Etiological factors and premature aging. JAMA, 1962, 179, 136-142.

[4.] Lavker RM, Kligman AM. Chronic heliodermatitis, A morphologic evaluation of chronic actinic damage with emphasis on the role of mast cells. J Invest Dermatol, 1988. 90. 325-330.

[5.] Kligman LH, Akin FJ, Kligman AM. The contributions of UVA and UVB to connective tissue damage in hairless mice. J Invest Dermatol, 1985, 94. 272-276.

[6.] Kligman LH. Intensification of ultraviolet-induced dermal damage by infrared radiation. Arch Dermatol Res, 1982. 272. 229-238.

[7.] Hilrose R, Kligman LH. An ultrastructural study of ultraviolet-induced elastic fiber damage in hairless mouse skin. J Invest Dermatol. 1988, 90, 697-702.

[8.] Gilchrest BA, Soter NA, Hawk JLM et at. Histologic changes associated with ultraviolet-A induced erythema in normal human skin. J Am Acad Dermatol, 1983, 9. 213-219.

[9.] Lavker RM, Gerberick GF, Veres D et al. Cumulative effects from repeated exposures to suberythemal doses of UVB and UVA in human skin. J AM Acad Dermatol, 1995, 32, 53-62.

[10.] Uiito J, Pau JL, Brockley K et al. Elastic fibers in human skin: quantification of elastic fibers by computerized digital image analyses and determination of elastin by radioimmunoassay of demosine. Lab Invest, 1983, 49, 499-505.

[11.] Godeau G, Gonnord G, Jolivet O et al. A selective histochemical method for the quantitative estimation of elastic fibers by computerized morphometric analysis. International Academy of Cytology, Anal Quant Cytol Histol. 1986. 8, 321-325.

[12.] Chen VL, Fleischmajer R, Schwartz E et al. Immunochemistry of elastotic material in .sun damaged skin. J Invest Dermatol, 1986, 87, 334-337.

[13.] Plastow, SR, Lovell CR, Young AR. UVB-induced collagen changes in the skin of the hairless albino mouse. J Invest Dermatol, 1987, 88, 145-148.

[14.] Schwartz E, Cruickshank FA, Perlish JS, Fleischmajer R. Alterations in dermal collagen in ultraviolet irradiated hairless mice. J Invest Dermatol, 1989, 93. 142-146.

[15.] Kaidbey KH, Kligman AM. The acute effects of longwave ultraviolet radiation on human skin. J Invest Dermatol, 1979, 72. 253-256.

Nicholas J Lowe (1), David P Meyers (2), Joshua M Wieder (1), Debra Luftman (1), Teresa Borget (1), Marjorie D Lehman (2), Anthony W Johnson (2) and Ian R Scott (2)

(1) Skin Research Foundation of California, Santa Monica, California, and

(2) Unilever Research US, Edgewater, New Jersey, USA
Table 1: Effect of repetitive irradiation on
epidermal dyskeratosis *

 12-week 24-week
Treatment value value *
group (K = 12) (n = 12)

1 MED solar simulated 0.38 [+ or -] 0.06 0.31 [+ or -] 0.61
Untreated control 0.00 [+ or -] 0.00 0.15 [+ or -] 0.36

1 MED UVA 0.56 [+ or -] 0.79 0.46 [+ or -] 0.75
UVA in solar simulation 0.44 [+ or -] 0.50 0.15 [+ or -] 0.36

Treatment ([dagger])
group (K = 12)

1 MED solar simulated 0.36 [+ or -] 0.48
Untreated control 0.09 [+ or -] 0.29

1 MED UVA 0.55 [+ or -] 0.50
UVA in solar simulation 0.36 [+ or -] 0.48

* Human buttock skin irradiated twice weekly for 24 weeks.
Biopsies obtained 12, 24 and 36 weeks after the initial
irradiation The degree of dyskeratosis visually assessed on
hematoxylin and eosin-stained sections using a grading scale
of 0, none; 1, mild; 2, moderate; and 3, severe.

([dagger]) Data are mean [+ or -] SD

Table 2: Effect of repetitive irradiation on area
fraction percent of elastic fibers *

 12-week value 24-week value
 ([dagger]) ([dagger])
Treatment group (n=12) (n=12)

1 MED solar simulated 4.66 [+ or -] 0.29 4.65 [+ or -] 0.67
Untreated control 4.53 [+ or -] 0.24 4.37 [+ or -] 0.72
1 MED UVA 3.77 [+ or -] 0.29 2.87 [+ or -] 0.32
UVA in solar simulation 4.10 [+ or -] 0.23 3.14 [+ or -] 0.40

 36-week value
Treatment group (n=12)

1 MED solar simulated 5.20 [+ or -] 0.49
Untreated control 4.49 [+ or -] 0.45
1 MED UVA 2.94 [+ or -] 0.44
UVA in solar simulation 3.78 [+ or -] 0.68

* Human buttock skin Irradiated twice weekly for 24 weeks.
Biopsies obtained 12. 24 and if: weeks alter the initial
irradiation and elastic tissue-like staining quantified via
image analysis

([dagger]) Data are mean [+ or -] SD.
COPYRIGHT 2009 Mediscript Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Leading Article
Author:Lowe, Nicholas J.; Meyers, David P.; Wieder, Joshua M.; Luftman, Debra; Borget, Teresa; Lehman, Marj
Publication:Clinical Dermatology
Article Type:Reprint
Date:Jun 1, 2009
Previous Article:Clinical dermatology: retinoids & other treatments--25 years.
Next Article:Botulinum A exotoxin for glabellar folds: a double-blind, placebo-controlled study with an electromyographic injection technique.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |