Proc. Nati. Acad. Sci. USA Vol. 89, pp. 8497-8501, September 1992 Immunology UV exposure reduces immunization rates and promotes tolerance to epicutaneous antigens in humans: Relationship to dose, CDla-DR+ epidermal macrophage induction, and Langerhans cell depletion (photoimmunology/ozone depletion/contact sensitivity/T lymphocytes/antigen presenting cells) K. D. COOPER*t*, L. OBERHELMAN*, T. A. HAMILTON*, 0. BAADSGAARD*§, M. TERHUNE*, G. LEVEE*, T. ANDERSON*t, AND H. KOREN¶ *Immunodermatology Unit, Department of Dermatology, University of Michigan, Ann Arbor, MI 48109; tVeterans Administration Hospital, Ann Arbor, MI 48105; and 'Health Effects Research Labs, Environmental Protection Agency, Chapel Hill, NC 27711 Communicated by Thomas M. Donahue, May 1, 1992 ABSTRACT Increasing UVB radiation at the earth's sur- tibility of UV-exposed mice and the unresponsiveness of face might have adverse effects on in vivo immunologic responses UV-exposed mice to contact allergens were found to be due in humans. We prospectively randomized subjects to test to antigen-specific suppressor T lymphocytes (12, 13). whether epicutaneous immunization is altered by prior admin- UV regulation of murine contact sensitivity has held up istration of biologically equalized doses of UV radiation. Mul- well as a model of immunologic events occurring in photo- tiple doses of antigens on upper inner arm skin (UV protected) carcinogenesis. Epidermal Langerhans cells, an antigen pre- were used to elicit contact sensitivity responses, which were senting population of dendritic cell lineage present in the quantitated by measuring increases in skin thickness. If a dose epidermis (14), have a potent capacity to initiate contact of UVB sufficient to induce redness (erythemagenic) was ad- sensitivity reactions (15, 16), as well as tumor rejection (17). ministered to the immunization site prior to sensitization with However, purified Langerhans cells exposed to UV are no dinitrochlorobenzene (DNCB), we noted a marked reduction in longer able to induce T-lymphocyte proliferation (18), pos- the degree of sensitization (P < 0.0006) that was highly dose sibly through alterations of adhesion molecule expression responsive (r = 0.98). Even suberythemagenic UV (less than a (19). The net result of UV exposure is that immunization (to visible sunburn) resulted in a decreased frequency of strongly contact, tumor, or microbial antigens) through the skin positive responses (32%) as compared to controls (73%) (P = results in persistent antigen-specific unresponsiveness to the 0.019). The rate of immunologic tolerance to DNCB (active antigens (20-23). Depression in the ability of T cells to react suppression ofa subsequent repeat immunization) in the groups to new peptide sequences generated as a result of a UV- that were initially sensitized on skin receiving erythemagenic induced genetic mutation could then result in tolerance rather doses of UV was 31% (P = 0.0003). In addition, a localized than rejection of UV carcinoma cells bearing such abnormal moderate sunburn appeared to modulate immunization with gene products (24). diphenylcyclopropenone through a distant, unirradiated site Previous studies on whether UV modulates the contact (41% weak responses) as compared to the control group (9%) (P sensitization potential of human skin in vivo have suggested = 0.05). Monitoring antigen presenting cell content in the slightly reduced (25-31) or unaltered responses (32, 33). epidermis revealed that erythemagenic regimens induced However, none of these studies is conclusive due to the use CDla-DR+ macrophages and depleted Lrhans cells. In of subjective assessments, to testing ofthe response by using conclusion, relevant and even subcinical levels of UV exposure skin of patients with skin disease or skin that had also have significant down modulatory effects on the ability of recently received UV exposure, to insufficient n for statis- humans to generate a T-celi-mediated response to antigens tical analysis, or to lack of appropriate control groups. introduced through irradiated skin. We modified a highly quantitative and sensitive method for assessing dinitrochlorobenzene (DNCB) contact sensitivity With UVB comprising an increasing proportion of the sun- in human skin (34). Prospectively randomized groups were light reaching the earth's surface, the impact ofsuch a change sensitized through normal skin or skin irradiated with various on human health becomes increasingly important to under- doses ofindividually biologically equalized doses of UV. We stand. In addition to causing photosensitivity diseases (i.e., found that UV exposure in humans resulted in highly signif- lupus, porphyrias, medication reactions) and carcinogenic icant, dose-responsive decreases in immunologic responsive- genetic mutations, mammalian UV exposure alters immuno- ness. Ofadditional concern are our findings that levels ofUV logic responses that normally handle microbial pathogens and exposure below clinical detectability can impair immune UV-induced cancers (1, 2). For instance, although murine responsiveness and that a localized sunburn can alter T-cell UV-induced cancers can be highly antigenic and are rejected responses at distant, unirradiated sites. upon transplantation into normal mice (3), mice exposed to subcarcinogenic doses of UV allow progressive tumor MATERIALS AND METHODS growth (4). Although most other immune functions remained Demographics and UV Exposure of Study Populations. The intact, UV-exposed animals could no longer become immu- Institutional Review Board approved the protocol, adver- nized to normally potent contact allergens (5, 6). UV expo- tisements, and consent document. Upon recruitment, each sure resulted not only in a simple failure ofimmunization, but subject was randomly assigned to a UV administration sched- also in long-term, active suppression of subsequent immuni- zations to the contact allergen through normal skin (toler- Abbreviations: DNCB, dinitrochlorobenzene; MED, minimal ance) (7-11). Both the increased UV-induced tumor suscep- erythemal dosage; DPCP, diphenylcyclopropenone. *To whom reprint requests should be addressed at: Immunoderma- tology Unit, University of Michigan, R5548 Kresge I/Box 0530, The publication costs of this article were defrayed in part by page charge Ann Arbor, MI 48109-0530. payment. This article must therefore be hereby marked "advertisement" §Present address: Department of Dermatology, Gentofte Hospital, in accordance with 18 U.S.C. §1734 solely to indicate this fact. University of Copenhagen, Copenhagen, Denmark. 8497 Downloaded by guest on October 1, 2021 8498 Immunology: Cooper et A Proc. Natl. Acad. Sci. USA 89 (1992) ule. Individuals were of skin type I, II, or III, without history cludes edematous vesicular changes), (v) extreme or spread- of chronic disease and not currently on medication. ing reaction (includes bullous or ulcerative reaction). Individuals received either no UV radiation or one ofthree In addition to the subjective assessment, we used an localized UV exposure regimens. A portable UVB photo- objective assessment of skin edema. The skin fold thickness therapy device (Dermacontrol, Frankfort, IL) containing six was determined by using a micrometer with spring-loaded FS40 bulbs emitted 0.3 mJ/cm2 at a 10-inch (1 inch = 2.54 cm) calipers (Mitutoyo Manufacturing, Tokyo), recording the source-to-skin distance. An LMHO6C meter (National Bio- skin thickness in millimeters at each site before and 48 hr after logical, Cleveland) equipped with an IL SEE1240 detector the patch was applied. The increase in skin thickness over the fitted with a W wide-angle quartz diffuser and a SCS280 filter 48-hr period was calculated for each site by subtracting the (International Light, Newburyport, MA) was used. The prechallenge thickness from the 48-hr postchallenge thick- minimal erythemal dosage (MED) was determined for each ness. The "mm increase sum" is derived from the addition of subject, such that biologically equivalent amounts of UVB the increases in skin thickness of each of the four concen- were administered. Except for the 3 x 5-inch exposure area trations for each allergen and allows the entire dose-response on the left buttock, all areas of the body were draped. Two curve (see Fig. 1) to be approximated as a single value (see groups were exposed 4 days in a row, with the first group Fig. 2). Photographs were also obtained. receiving 0.75 of the MED (0.75MED X 4) and the second Langerhans Cell Quantitation in Sheets. Punch biopsies (4 group receiving 2 times the MED each day (2MED x 4) mm) were taken on the day of sensitization-one from the (Table 1). These subjects were sensitized to contact allergens exposed buttock and one from the protected buttock. A 1 M immediately following the last UV exposure (day 4). A third NaCl-split sheet of epidermis was fixed in acetone and group received 4 MED on the 1st day only (4MED x 1). stained with a combination of fluoresceinated OKT6 (anti These subjects were sensitized on the 3rd day after UV CD1a; Orthomune, Raritan, NJ) and phycoerythrin-conju- exposure (day 3). To control for diminished levels of contact gated anti-HLA-DR (Becton Dickinson). Blinded quantita- sensitization that occurs in menstruating women except tions were performed with the aid of an ocular grid. during midcycle (35), all female subjects were sensitized 14 Statistical Analysis. Differences in challenge response, as days after the onset of menses. measured by increase (mm) in skin thickness and visual Sensitization.