j. Soc. Cosmet. Chem., 47, 129-155 (May/June 1996) Broad-spectrum sunscreens with UVAI and UVA II absorbers provide increased protection against solar-simulating radiation-induced dermal damage in hairless mice LORRAINE H. KLIGMAN, Department of Dermatology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104-6142, PATRICIA P. AGIN, Research and Development, Schering-Plough Health Care Products, 3030Jackson Avenue, Memphis, TN 38151, and ROBERT M. SAYRE, Rapid Precision Testing Laboratory, Cordova, TN 38018-1342. Accepted jSr publication June 19, 1996. Synopsis Previous experiments designed to examine sunscreen protection against chronic UV radiation-induced skin damage in hairless mice have used radiation sources emitting mainly UVB or UVA radiation. Because humans are exposed to full-spectrum solar radiation, we were interested in examining the efficacy of three sunscreens, with increasing spectral absorption into the UVA range, against chronic solar-simulating radiation (SSR). Three groups of hairless mice received a cumulative SSR dose of 10 and 16 times a previously determined minimal photoaging dose (MPD) over periods of 18 and 30 weeks. Each twice- weekly exposure was designed to equal the SPF value of the first sunscreen, an SPF-7 sunscreen containing the UVB absorber octyl methoxycinnamate. The second sunscreen, in addition to the UVB absorber, contained a UVA II absorber (oxybenzone) and had an SPF of 16. The third, with an SPF of 18, contained the UVB and UVA II absorbers plus a UVA I absorber (avobenzone). These conditions allowed assessment of the effects of UVB and UVA radiation that are normally transmitted through all sunscreens. Although none of the sunscreen-treated mice developed erythema, considerable dermal matrix damage occurred in the SPF-7 group, with greater damage at 16 MPD than at 10 MPD. The SPF-16 sunscreen allowed less but clearly recognizable damage at both dose points. The SPF-18 sunscreen with the broadest spectral absorp- tion provided the greatest protection. These results support the need for high-SPF broad-spectrum sun- screen protection that includes the entire UVA spectrum to reduce photodamage that results from chronic exposure to sunlight. INTRODUCTION The hairless mouse has proved a useful model for demonstrating the protective effect of sunscreens against the diverse consequences of chronic ultraviolet radiation. For exam- ple, sunscreens with high sun protection factors (SPF) have been shown to prevent UVB-induced photocarcinogenesis (1,2) and the dermal connective tissue damage char- 129

130 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS acteristic of photoaging (3-5). Sunscreens applied to mice after the induction of dermal damage allowed matrix repair despite continued UVB exposure (6). UVA-induced damage has also been effectively prevented by sunscreens containing UVA-absorbing molecules (7-9). Except for one study (10) that used solar-simulating radiation (SSR), all other studies used radiation sources that emitted predominantly UVB or UVA and that were not representative of the complete solar ultraviolet spectrum. In summary, laboratory studies with hairless mice have shown sunscreens to provide nearly complete protection against photodamage. Despite appropriate application and use of sunscreen products, small amounts of UVB and UVA penetrate to the viable layers of the skin. An SPF-15 sunscreen will block about 93% of incident radiation. Sunburn may be prevented, but the small amount of transmitted radiation may have serious consequences with regard to chronic photodam- age. We were, therefore, interested in examining the effect of high doses of SSR on connective tissue damage in sunscreen-protected hairless mice. On the basis of previous work to establish an action spectrum for elastosis (! 1), it was determined, using a solar simulator, that a cumulative dose of --! J/cm 2 of the UVB portion of the emission allows a cumulative UVA dose of -- 170 J/cm 2. This dose, achieved over a period of nine weeks, produced a 50% increase in elastic fibers, as quantified by image analysis. This was seen, histologically, as mild elastic fiber hyperplasia. We thought it appropriate to define this dose as a minimal photoaging dose (MPD). In the current study we irradiated hairless mice until 10 and 16 MPDs were accumu- lated, based on UVB calculations (1 J/cm 2 = ! MPD). At the rate of two weekly exposures of seven minimal erythema doses (MEDs) each, these two dose points required 18 and 30 weeks, respectively. The 7-MED dose was chosen to equal the SPF value of the basic sunscreen, which contained only a UVB absorber (octyl methoxycinnamate). Two other sunscreens contained UVA absorbers in addition to the UVB absorber. One had an SPF of 16, with oxybenzone as the UVA absorber, and the other, an SPF of 18, with oxybenzone plus avobenzone (Parsol 1789) as UVA absorbers. The experimental conditions enabled us to examine the long-term consequences of UVA on UVB sun- screen-protected skin, the effects of the small amounts of radiation that are not blocked by the sunscreen, and the increase in protection provided by the addition of UVA absorbers to a UVB-absorbing sunscreen. MATERIALS AND METHODS ANIMALS Female albino hairless mice (Skh-hairless-1), ages 10-12 weeks, were obtained from the Temple University Health Sciences Center, Philadelphia, PA. Three groups of 18 mice each were individually housed but irradiated nine at a time in a cage designed to prevent mice from shielding each other. Room lighting (12-h on/off cycle) was with General Electric F-40 "GO" gold fluorescent tubes, which emit no measurable UV radiation. SOURCE AND SCHEDULE OF ULTRAVIOLET RADIATION A forced-air cooled 4200 W compact xenon arc solar simulator (Spectral Energy Corp.,