In Vivo UVA-1 and UVB Irradiation Differentially Perturbs theAntigen-Presenting Function of Human EpidermalLangerhans Cells

Henning C. Dittmar, Johannes M. Weiss, Christian C. Termeer, Ralf W. Denfeld, Marcus B. Wanner,*Lone Skov,† Jonathan NWN Barker,‡ Erwin Schopf, Ole Baadsgaard,† and Jan C. SimonDepartment of Dermatology, University of Freiburg, Freiburg, Germany; *Department of Surgery, University of Basel, Basel, Switzerland; †Departmentof Dermatology, University of Copenhagen, Gentofte Hospital, Hellerup, Denmark; ‡St. Johns Institute of Dermatology, UMDS, St. Thomas’Hospital, London, U.K.

Ultraviolet B (UVB, 290–320 nm) radiation is knownto suppress the immune function of epidermalLangerhans cells. We have recently described thatin vitro UVB irradiation perturbs the antigen-presentingcell function of Langerhans cells by inhibiting theirexpression of functional B7 costimulatory molecules(B7–1, B7–2). The aim of this study was to determinewavelength-specific UV effects on Langerhans cellsfunction in vivo, specifically UVB and UVA-1. Toaddress this issue, volunteers were irradiated on thesunprotected volar aspects of their forearms with 3 Hminimal erythema dose of UVB (Philips TL-12) andUVA-1 (UVASUN 5000 Mutzhaas). Langerhans cellswere isolated from suction blister roofs immediatelyfollowing irradiation. Langerhans cells isolated fromUVB- but not from UVA-1-irradiated skin failed toactivate naıve resting allogeneic T cells (mixed epi-dermal cell leukocyte reaction) or primed tetanustoxoid reactive autologous T cells. Langerhans cells

Non-ionizing UV radiation modulates cell-mediatedimmune responses in humans and animal models(Ullrich, 1995a; Cooper, 1996; Elmets andAnderson, 1996; Krutmann et al, 1996). UVradiation reaching the skin consists of different

wavelengths, its main constituents being UVB (290–320 nm) andUVA (UVA-1, 340–400 nm, and UVA-2, 320–340). In contrast toUVA radiation, UVB light is almost completely absorbed withinthe epidermis. Epidermal cells are therefore considered to be theprincipal target for the immunosuppressive effects of both UVBand UVA radiation. Studies concerning the effects of UVB radiationon epidermal cells have shown that keratinocytes, macrophages,and Langerhans cells are influenced by UVB light (Schwarzand Luger, 1989; Krutmann and Trefzer, 1992; Beissert andGranstein, 1996).

Manuscript received April 9, 1998; revised June 29, 1998; accepted forpublication November 13, 1998.

Reprint requests to: Dr. Jan C. Simon, Department of Dermatology,University of Freiburg, Hauptstrasse 7, D-79104 Freiburg, Germany.

Abbreviations: APC, antigen-presenting cell; MED, minimal erythemadose; MFI, mean fluorescence intensity.

0022-202X/99/$10.50 · Copyright © 1999 by The Society for Investigative Dermatology, Inc.

322

isolated from sham-irradiated or UVA-1-irradiatedskin strongly upregulated B7–2 molecules during short-term tissue culture. By contrast, Langerhans cellsfrom UVB-irradiated skin did not upregulate B7–2molecules. Furthermore, exogenous stimulation of theB7 pathway by anti-CD28 stimulatory monoclonalantibodies restored the capacity of UVB-irradiatedLangerhans cells to activate both alloreactive andtetanus toxoid-reactive T cells, implying suppressedantigen-presenting cell activities and perturbed B7expression of Langerhans cells isolated from UVB-irradiated skin are related. Those studies demonstratethat in vivo UVB, but not UVA-1, interferes with theactivation-dependent upregulation of B7 moleculeson Langerhans cells, which in turn is of functionalrelevance for the perturbed antigen-presenting cellfunction of Langerhans cells within UVB- but notUVA-1-irradiated skin. Key words: CD28/CD80/CD86/tetanus toxoid. J Invest Dermatol 112:322–325, 1999

Epidermal Langerhans cells are potent antigen-presenting cells(APC) and due to their localization in the suprabasal layers of theepidermis they are susceptible to the effects of UVB radiation. Itis well established that in vitro UVB irradiation affects the antigen-presenting function of Langerhans cells (Simon et al, 1990, 1991a,b, 1992; Ullrich, 1995a, b). On the other hand, in vivo UVAirradiation had no effect on CD1a and ATPase expression byhuman Langerhans cells and did not suppress contact hypersensitivityresponses in humans (Koulu et al, 1985; Skov et al, 1997); however,in vitro UVA radiation, like UVB radiation, induces the release ofimmunomodulatory cytokines by keratinocytes (Krutmann andGrewe, 1995). In this study we therefore wished to compare theeffects of in vivo UVB and UVA-1 irradiation on the APC functionof human Langerhans cells.

MATERIALS AND METHODS

Media and chemicals Complete-RPMI 1640 (Gibco, Eggenstein,Germany) was supplemented with 10% heat-inactivated fetal calf serum(Gibco), 25 mM N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid(HEPES) (Sigma, Munchen, Germany), 50 µg penicillin-streptomycin(Gibco) per ml. Trypsin (Gibco) was used in a concentration of 0.25%in phosphate-buffered saline supplemented with DNAse I (Boehringer,Mannheim, Germany, 80 U per ml) to dissociate the epidermis ofsuction blisters.

VOL. 112, NO. 3 MARCH 1999 IN VIVO UVA-1 AND UVB DIFFERENTIALLY PERTURBS B7 MOLECULES ON HUMAN LANGERHANS CELLS 323

Monoclonal antibodies (MoAb) PE-conjugated anti-HLA-DR (L243,mouse IgG2a) and PE-conjugated IgG2a control MoAb were obtained fromBecton Dickinson (Heidelberg, Germany), and IgG1 control antibodieswere from Dianova (Hamburg, Germany). MoAb against B7–1 (cloneL307.4, mouse IgG1) was procured from Becton Dickinson, and anti-B7–2 (clone IT2.2, mouse IgG2b) from PharMingen (San Diego, CA). MoAb9.3 (mouse IgG2a) was a kind gift of Dr. Linsley (Bristol Myers SquibbPharmaceutical Research Institute, Seattle, WA). Anti-CD14 (mouseIgG2b), CD19 (mouse, IgG1, both Becton Dickinson), and HLA-DR andCD8 (ATCC HB145, mouse IgG1 and OKT8, mouse IgG2) were usedfor ‘‘negative selection’’ enrichment of T cells (Skov and Baadsgaard,1996a). Cross-linking sheep anti-mouse polyclonal antibodies (IgG, code-number 515–005–062) was purchased from Dianova. Sheep anti-mouseIgG MoAb coupled with Dynal beads were kindly provided by Dynal(Hamburg, Germany).

UV irradiation After informed consent, healthy volunteers (skin typesII and III) with no history of skin cancer or photosensitivity were testedduring the winter months for minimal erythema dose (MED) of UVA-1and UVB irradiation. Subsequently the volar aspects of the forearm of eachindividual were irradiated with 3 3 MED of UVA-1 (UVASUN 5000Mutzhaas, Munchen, Germany, 340–400 nm) and UVB with four unfilteredTL-12 fluorescent tubes (Philips, Hamburg, Germany, 250–400 nm witha peak at 313 nm) placed 46 cm above the skin (Simon et al, 1991a). TheUVA-1 dosimetry was performed with a Waldmann radiometer (Waldmann,Schwennigen, Germany); the UVB dose was measured with a UV researchradiometer equipped with a SCS 280 photodetector (International Light,Newburyport, MA). The study protocol was approved by the local ethicscommittee.

Epidermal cell suspensions from suction blisters Suction blisterswere raised on normal or UVA-1 and UVB irradiated (3 3 MED) skin,as described (Kiistala and Mustakallio, 1967; Skov and Baadsgaard, 1996b).Epidermal cell suspensions were generated by limited trypsinization (Weisset al, 1995). Epidermal cells were either analyzed directly by fluorescence-activated cell sorter (FACS) analysis or cultured in supplemented RPMI1640 (37°C, 5% CO2) for 48 h.

Immunostaining and flow cytometry For three color FACS analysiscells were stained in phosphate-buffered saline (4°C) with one of thefollowing unlabeled primary MoAb: anti-B7–1, B7–2, or appropriateisotype control MoAb, followed by DTAF-labeled goat anti-mouse (Fab9)2(H 1 L) MoAb (Dianova), followed by a blocking step with mouse serum(Dianova) and finally with HLA-DR PE labeled MoAb or nonreactive PElabeled IgG2a control antibodies (Weiss et al, 1995). 7-AminoactinomycinD (7-AAD) (2.5 mg per ml, Sigma) was added to exclude nonviable cellsby life gating. Epidermal cell suspensions were analyzed by FACScan,applying the CellQuest research software (both Becton Dickinson)collecting 5.000 HLA-DR1 viable cells. Mean fluorescence intensity (MFI)was determined with the FACScan CellQuest Software.

APC assays APC assays were performed according to published pro-cedures (Weiss et al, 1995). In brief, autologous or allogeneic T cells wereenriched by plastic adherence and immunomagnetic depletion with HLA-DR, CD14, CD19, and CD8 MoAb primary antibodies and second stepsheep anti-mouse IgG MoAb coupled to Dynal beads. The resulting Tcells were .95% CD31 as determined by FACS analysis. T cells (1 3 105)were cocultured (mixed epidermal cell leukocyte reaction, supplementedRPMI in 5% CO2, 37°C, 168 h) with in vivo UVA-1, UVB irradiated, orsham-irradiated autologous (tetanus toxoid) or allogeneic epidermal cells(5 3 104) from suction blisters in 96 well round-bottom microtiterplates (Costar, Cambridge, MA). To assess tetanus toxoid-specific T cellproliferation, tetanus toxoid (Seruminstitut, Copenhagen, Denmark) wasadded to the cocultures at 2 µg per ml. For exogenous stimulation studieswith MoAb, anti-CD28 MoAb 9.3 or the isotype control MoAb wereadded at a concentration of 10 µg per ml at the initiation of the cocultures.In addition, cross-linking sheep anti-mouse polyclonal antibodies wereadded at a concentration of 10 µg per ml. [3H]Thymidine (1 µCi) wasadded to each well for the final 16 h of coculture. Plates were harvestedwith a Canberra Packard Filter Mate (Canberra Packard, Frankfurt,Germany) and incorporation of [3H]thymidine was determined by liquidscintillation spectroscopy using a Top-Count (Canberra Packard).

Statistical analysis The differences in T cell proliferation inducedby irradiated Langerhans cells were compared with those induced byunirradiated Langerhans cells by Student’s t test; p values , 0.05 wereconsidered significant.

Figure 1. In vivo UVA-1 and UVB radiation differentially affects theAPC function of human Langerhans cells. After MED testing eachvolunteer was in vivo irradiated with 33MED UVA-1, UVB, and shamtreated. Suction blisters were raised from which epidermal cell suspensionswere generated immediately following irradiation. Epidermal cells (5 3 104)were cocultured with allogeneic (a) or autologous (b) T cells (1 3 105).tetanus toxoid (2 µg per ml) was added to the autologous (b) cultures. Tcell proliferation was determined after a 7 d culture by [3H]TdR uptake.*p , 0.0001, relative to sham-irradiated epidermal cells.

RESULTS

In vivo UVB but not UVA-1 radiation inhibits the APCfunction of Langerhans cells To investigate the effects of in vivoUVA-1 and UVB radiation on the APC function of Langerhanscells, we applied 3 3 MED of either UV wavelength to the sun-protected forearm of volunteers. Immediately following irradiation,epidermal cell suspensions containing 1000 Langerhans cells perwell were investigated for their capacity to stimulate the proliferationof alloreactive naıve T cells and of tetanus toxoid reactive primedT cells. Epidermal cell suspensions from sham- or UVA-1-irradiatedskin were fully capable of stimulating the proliferation of bothalloreactive and tetanus toxoid reactive T cells (Fig 1a, b). Theseresults demonstrate that the APC function of human Langerhanscells is not affected by in vivo UVA-1 irradiation. By contrast,epidermal cells isolated from UVB-irradiated skin failed to stimulateboth types of T cell responses (Fig 1a, b).

In vivo UVB but not UVA-1 irradiation perturbs the func-tional expression of B7 costimulatory molecules on Langer-hans cells We were also interested in determining whether thecostimulatory molecule B7 and the suppressed APC function ofin vivo UVB-irradiated Langerhans cells are related. Epidermal cellsfrom suction blisters suspensions were analyzed fresh or after 48 hof tissue culture for their B7–1 and B7–2 expression pattern. Onlyviable HLA-DR1 7-AAD– cells were analyzed by FACS. In vivosham-irradiated HLA-DR1 epidermal cells strongly upregulatedB7–2 (MFI 119) during culture, whereas B7–1 (MFI 18) expressionwas low. In vivo UVB irradiation inhibited the culture-inducedupregulation of B7–2 (MFI 25). In contrast, in vivo UVA-1irradiation had no effect (MFI 118) (Fig 2).

After this finding we questioned whether the reduced APCactivity of in vivo UVB-irradiated Langerhans cells was related tothe perturbed expression of functional B7 molecules. Provision ofexogenous B-7 costimulation by the addition of the stimulatinganti-CD28 MoAb 9.3 to epidermal cells isolated from UVB-irradiated skin, partially restored the proliferation of alloreactiveand tetanus toxoid reactive T cells (Fig 3). T cell proliferation,however, could not be reconstituted completely to the levelsinduced by MoAb 9.3 added to sham-irradiated Langerhans cells,suggesting that, albeit the perturbed APC function of UVB-irradiated Langerhans cells may be related (or appears to be related

324 DITTMAR ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 2. In vivo UVB but not UVA-1 irradiation disturbs culture-induced upregulation of B7–2 on Langerhans cells. Skin on the volarforearm was sham treated (a–d) or irradiated with 33MED UVA-1 (e, f)and UVB (g, h). Epidermal cells were generated from suction blisters.Epidermal cells were labeled with MoAb against HLA-DR (PE) (b–h)and IgG (a, b), B7–1 (DTAF) (c, e, g), or B7–2 (DTAF) (d, f, h). Viable(7-AAD–), HLA-DR1 cells (5000 cells per sample) were analyzed byFACScan. The mean fluorescence intensities of B7–1 and B7–2 expressionare displayed in the upper right corner (n 5 3).

in part) to their failure to upregulate B7–2 molecules, othersuppressor mechanisms also contribute to this effect of UVBradiation.

DISCUSSION

This paper documents the differential effect of in vivo UVB andUVA-1 irradiation on the APC function of human Langerhans cells.Langerhans cells obtained from UVB-irradiated (UVB-irradiatedLangerhans cells) but not UVA-1-irradiated skin (UVA-1-irradiatedLangerhans cells) failed to stimulate alloreactive and tetanus toxoidspecific T cells. This perturbed APC function of UVB-irradiatedLangerhans cells appears to be related to their failure to upregulateB7–2 costimulatory molecules, because this defect could be partiallyrestored by exogenous B7 stimulation with anti-CD28 MoAb.These findings extend previous observations by us and others(Weiss et al, 1995; Rattis et al, 1996) that in vitro UVB irradiationsuppresses the APC function of Langerhans cells by perturbingtheir expression of B7 costimulatory molecules. On the other hand,a recent study by Laihia and Jansen (1997) showed that in vivoirradiation of human skin with a solar simulator leads to anupregulated B7–2 expression on Langerhans cells 24 h after radia-tion, returning to baseline levels after 48 h. The authors speculatethat this may be due to an influx of dermal Langerhans cellprecursors into the epidermal compartment, and a subsequentshedding of the costimulatory molecules (Laihia and Jansen, 1997).By contrast, in this study we isolated the cells directly afterirradiation to prevent migration of dermal cells. Furthermore thecontrasting effects of UVB on B7 expression in the three studies

Figure 3. CD28 ligation overcomes the suppression of T cellproliferation induced by Langerhans cells from UVB-irradiatedskin. Epidermal cells (5 3 104) from UVB-irradiated (33MED) orunirradiated skin were cocultured with allogeneic or autologous T cells(1 3 105) as detailed in Fig 1. In addition to cross-linking sheep anti-mouse polyclonal antibodies (10 µg per ml), anti-CD28 MoAb 9.3 or IgGcontrol were applied at concentrations of 10 µg per ml. T cell proliferationwas determined after a 7 d culture by [3H]TdR uptake. *p , 0.004,relative to sham-irradiated epidermal cells (n 5 3).

could be due to differences in the light sources, solar-simulatorcompared with UVB irradiation, or to differences in doses ofirradiation used, 1 3 MED in the study by Laihia and Jansen(1997) compared with 3 3 MED in our study.

Our finding that Langerhans cells isolated from in vivo UVB-irradiated skin showed suppressed B7–2 upregulation upon in vitroculture, whereas B7–1 expression remained undetactable during48 h culture, suggests that a perturbed B7–2 expression onLangerhans cells is more important than B7–1 for the UVB-inducedimmunosuppression. This notion is supported by previous reportsindicating that although both B7–1 (CD80) and B7–2 (CD86)bind to CD28 and CTLA-4, they differ in their expression patternsand signaling properties (June et al, 1994). For example, B7–1 incontrast to B7–2 is expressed at low/undetectable levels on restingLangerhans cells (Simon et al, 1994; Yokozeki et al, 1996) and,unlike B7–1, B7–2 is rapidly upregulated upon Langerhans cellsstimulation (Inaba et al, 1994). There is evidence from previousstudies by Peguet-Navarro et al (1995) that B7–2 is more importantthan B7–1 for T cell activation by human Langerhans cells.Furthermore, findings by Rattis et al (1996) demonstrated thatMoAb against B7–2, but not B7–1, suppressed the allostimulatorycapacity of Langerhans cells. Based on these observations it seemslikely that a perturbed expression of B7–2 molecules is morerelevant than B7–1 for the UVB-induced suppression of Langerhanscells APC function.

We note that in allogeneic and tetanus toxoid specific T cellsystems, exogenous B7 costimulation could only partially overcomethe APC defect of UVB-irradiated Langerhans cells. This supportsthe notion that in addition to the defective B7 signaling otherfactors contribute to the failure of UVB-irradiated Langerhans cellsto activate T cells. Previous data by Tang and Udey (1992)demonstrated that ICAM-1 is also suppressed by UVB on murineepidermal Langerhans cells. Furthermore, IL-10 released fromirradiated keratinocytes (Rivas and Ullrich, 1992) was shown toinfluence the APC function of Langerhans cells by suppressingthe alloreactive T cell stimulatory function (Enk et al, 1993;Peguet-Navarro et al, 1994). For the human system, Caux et al(1994) demonstrated that IL-10 inhibits the cytokine productionof dendritic cells generated from CD341 hematopoietic progenitor

VOL. 112, NO. 3 MARCH 1999 IN VIVO UVA-1 AND UVB DIFFERENTIALLY PERTURBS B7 MOLECULES ON HUMAN LANGERHANS CELLS 325

cells. Hence IL-10 plays an important role in the downregulationof cell-mediated immune response.

Furthermore, this study demonstrates an UV-wavelengthdependency for the immunosuppressive effects on Langerhanscells. Specifically, long-wave UVA-1 irradiation (340–400 nm), incontrast to broad-band UVB (290–320 nm), was without effect onLangerhans cells B7 expression and did not perturb their T cellstimulatory capacity. These findings extend previous findings ofKoulu et al (1985) demonstrating that broad-band UVA did notinfluence the surface expression of ATPase and CD1a on Langerhanscells, and of Skov et al (1997) that UVA-1 in contrast to UVBfailed to suppress contact hypersensitivity in humans. Previousstudies by Baadsgaard et al (1989) have demonstrated that UVA, incontrast to UVB, allowed a rapid recovery of human Langerhanscells alloreactivity, without induction of autoreactivity. Thistemporary effect of UVA in the latter study as compared with ourscould be due to a higher irradiation protocoll (43MED), punchbiopsies, and trypsination overnight employed by Baadsgaard et al(1989). A recent paper by LeVee et al (1997) showed that short-wave UVA-2 (320–340 nm) like UVB suppresses the immunefunction of epidermal APC. In aggregate, these findings supportthe notion that epidermal Langerhans cells are particularly sensitiveto the immediate effects of in vivo UVB and short-wave UVAradiation (290–340 nm), while being relatively resistant to long-wave UVA-1 radiation (340–400 nm).

In conclusion, this paper documents the differential effect ofin vivo UVB and UVA-1 radiation on the APC function of humanLangerhans cells. Langerhans cells residing within skin irradiatedwith 33MED of UVB, but not with 33MED of UVA-1, displaydefective B7–2 costimulation, which in turn impairs their capacityto stimulate T cell responses.

This research was supported by the European Union (EVV-CT94–0563). Wethank Dr. Linsley, Bristol-Myers Squibb, Seattle, WA for the gift of MoAb andDr. Lappin for helpful discussions and critical reading of the manuscript.

REFERENCESBaadsgaard O, Lisby S, Wantzin GL, Wulf HC, Cooper KD: Rapid recovery of

Langerhans cell alloreactivity, without induction of autoreactivity, after in vivoultraviolet A, but not ultraviolet B exposure of human skin. J Immunol142:4213–8, 1989

Beissert S, Granstein RD: UV-induced cutaneous photobiology. Critical Reviews inBiochemistry and Molecular Biology 31:381–404, 1996

Caux C, Massacrier C, Vanbervliet B, Barthelemy C, Liu YJ, Banchereau J:Interleukin 10 inhibits T cell alloreaction induced by human dendritic cells.International Immunology 6:1177–85, 1994

Cooper KD: Cell-mediated immunosuppressive mechanisms induced by UVradiation. Photochem Photobiol 63:400–6, 1996

Denfeld RW, Tesmann JP, Dittmar HC, Weiss JM, Schopf E, Weltzien HU, SimonJC: Further characterization of UVB radiation effects on Langerhans cells:altered expression of the costimulatory molecules B7-1 and B7-2. PhotochemPhotobiol 67:554–560, 1998

Elmets CA, Anderson CY: Sunscreens and photocarcinogenesis: an objectiveassessment. Photochem Photobiol 63:435–40, 1996

Enk AH, Angeloni VL, Udey MC, Katz SI: Inhibition of Langerhans cell antigen-presenting function by IL-10. A role for IL-10 in induction of tolerance. JImmunol 151:2390–8, 1993

Inaba K, Witmer-Pack M, Inaba M, et al: The tissue distribution of the B7–2costimulator in mice: abundant expression on dendritic cells in situ and duringmaturation in vitro. J Exp Med 180:1849–60, 1994

June CH, Bluestone JA, Nadler LM, Thompson CB: The B7 and CD28 receptorfamilies. Immunol Today 15:321–31, 1994

Kiistala U, Mustakallio KK: Dermo-epidermal separation with suction. Electronmicroscopic and histochemical study of initial events of blistering on humanskin. J Invest Dermatol 48:466–77, 1967

Koulu L, Jansen CT, Viander M: Effect of UVA and UVB irradiation on humanepidermal Langerhans cell membrane markers defined by ATPase activity andmonoclonal antibodies (OKT 6 and anti-Ia). Photo-Dermatology 2:339–46, 1985

Krutmann J, Grewe M: Involvement of cytokines, DNA damage, and reactiveoxygen intermediates in ultraviolet radiation-induced modulation ofintercellular adhesion molecule-1 expression. J Invest Dermatol 105:67S–70S,1995

Krutmann J, Trefzer U: Modulation of the expression of intercellular adhesionmolecule-1 (ICAM-1) in human keratinocytes by ultraviolet (UV) radiation.Spring Sem Immunopath 13:333–44, 1992

Krutmann J, Ahrens C, Roza L, Arlett CF: The role of DNA damage and repair inultraviolet B radiation-induced immunomodulation: relevance for humanphotocarcinogenesis. Photochem Photobiol 63:394–6, 1996

Laihia JK, Jansen CT: Up-regulation of human epidermal Langerhans’ cell B7–1 andB7–2 co-stimulatory molecules in vivo by solar-simulating irradiation. Eur JImmunol 27:984–9, 1997

LeVee GJ, Oberhelman L, Anderson T, Koren H, Cooper KD: UVA2 exposure ofhuman skin results in decreased immunization capacity, increased induction oftolerance and a unique pattern of epidermal antigen-presenting cell alteration.Photochem Photobiol 65:622–9, 1997

Peguet-Navarro J, Moulon C, Caux C, Dalbiez-Gauthier C, Banchereau J, SchmittD: Interleukin-10 inhibits the primary allogeneic T cell response to humanepidermal Langerhans cells. Eur J Immunol 24:884–91, 1994

Peguet-Navarro J, Dalbiez-Gauthier C, Rattis FM, van Kooten C, Banchereau J,Schmitt D: Functional expression of CD40 antigen on human epidermalLangerhans cells. J Immunol 155:4241–7, 1995

Rattis FM, Peguet-Navarro J, Staquet MJ, Dezutter-Dambuyant C, CourtellemontP, Redziniak G, Schmitt D: Expression and function of B7–1 (CD80) andB7–2 (CD86) on human epidermal Langerhans cells. Eur J Immunol 26:449–53, 1996

Rivas JM, Ullrich SE: Systemic suppression of delayed-type hypersensitivity bysupernatants from UV-irradiated keratinocytes. An essential role forkeratinocyte-derived IL-10. J Immunol 149:3865–71, 1992

Schwarz T, Luger TA: Effect of UV irradiation on epidermal cell cytokine production.Journal of Photochemistry and Photobiology B – Biology 4:1–13, 1989

Simon JC, Cruz PD Jr, Bergstresser PR, Tigelaar RE: Low dose ultraviolet B-irradiated Langerhans cells preferentially activate CD41 cells of the T helper2 subset. J Immunol 145:2087–91, 1990

Simon JC, Cruz PD Jr, Tigelaar RE, Sontheimer RD, Bergstresser PR: Adhesionmolecules CD11a, CD18, and ICAM-1 on human epidermal Langerhans cellsserve a functional role in the activation of alloreactive T cells. J Invest Dermatol96:148–51, 1991a

Simon JC, Tigelaar RE, Bergstresser PR, Edelbaum D, Cruz PD Jr: Ultraviolet Bradiation converts Langerhans cells from immunogenic to tolerogenic antigen-presenting cells. Induction of specific clonal anergy in CD41 T helper 1 cells.J Immunol 146:485–91, 1991b

Simon JC, Krutmann J, Elmets CA, Bergstresser PR, Cruz PD Jr: Ultraviolet B-irradiated antigen-presenting cells display altered accessory signalling for T-cellactivation: relevance to immune responses initiated in skin. J Invest Dermatol98:66S–69S, 1992

Simon JC, Dietrich A, Mielke V, et al: Expression of the B7/BB1 activation antigenand its ligand CD28 in T-cell-mediated skin diseases. J Invest Dermatol 103:539–43, 1994

Skov L, Baadsgaard O: MHC class II1 keratinocytes from IFN gamma-treatedhuman skin activate T cells in the presence of staphylococcal superantigendespite UVB irradiation. Archives of Dermatological Research 288:255–7, 1996a

Skov L, Baadsgaard O: Ultraviolet B-exposed major histocompatibility complex classii positive keratinocytes and antigen-presenting cells demonstrate a differentialcapacity to activate T cells in the presence of staphylococcal superantigens. BrJ Dermatol 134:824–30, 1996b

Skov L, Hansen H, Barker JN, Simon JC, Baadsgaard O: Contrasting effects ofultraviolet-A and ultraviolet-B exposure on induction of contact sensitivity inhuman skin. Clin Exp Immunol 107:585–8, 1997

Tang A, Udey MC: Effects of ultraviolet radiation on murine epidermal Langerhanscells: doses of ultraviolet radiation that modulate ICAM-1 (CD54) expressionand inhibit Langerhans cell function cause delayed cytotoxicity in vitro. J InvestDermatol 99:83–9, 1992

Ullrich SE: Cutaneous biologic responses to ultraviolet radiation. Curr OpinionDermatol 95:225–30, 1995a

Ullrich SE: Modulation of immunity by ultraviolet radiation: key effects on antigenpresentation. J Invest Dermatol 105:30S–36S, 1995b

Weiss JM, Renkl AC, Denfeld RW, de Roche R, Spitzlei M, Schopf E, Simon JC:Low-dose UVB radiation perturbs the functional expression of B7.1 and B7.2co-stimulatory molecules on human Langerhans cells. Eur J Immunol 25:2858–62, 1995

Yokozeki H, Katayama I, Ohki O, et al: Functional CD86 (B7–2/B70) on culturedhuman Langerhans cells. J Invest Dermatol 106:147–53, 1996