Ultraviolet Radiation-induced Apoptosis Is Mediated by Activation of CD-95 (Fas/APO-1)*

Exposure to ultraviolet light (UV) can induce apoptosis in mammalian cells. The mechanism by which UV radiation engages the suicide apparatus is unclear. Here we demonstrate that UV radiation can activate the Fas pathway via receptor aggregation and subsequent recruitment of the death adaptor molecule FADD/MORT1. UV radiation-induced apoptosis was inhibited by both a dominant negative version of FADD (FADD-DN) and the caspase inhibitor CrmA. Thus, activation of the Fas pathway represents a physiologic mechanism by which UV-damaged cells are eliminated.

Depletion of the ozone layer by chloro/fluorohydrocarbon pollutants and subsequent increase in exposure to ultraviolet (UV) radiation threatens to significantly increase the incidence of skin cancer. In addition, exposure to UV radiation can lead to photokeratitis, photoconjunctivitis, and permanent retinal blindness. UV irradiation of cells elicits a complex cellular response termed the UV response, which includes the posttranslational activation of pre-existing transcription factors including NF-B and AP-1 (1,2). Activation of these transcription factors results in the subsequent induction of proinflammatory gene products IL-1 1 and TNF-␣ (3,4). Prolonged UV exposure results in inhibition of DNA synthesis and subsequent resumption of the cell cycle following repair or apoptosis, a mechanism to rid the organism of irreversibly damaged and potentially cancerous cells. Little is known about the mechanism by which UV radiation triggers apoptosis.
It is now evident that the effector arm of the death pathway is composed of caspases, cysteine proteases that cleave death substrates including PARP and lamin B following aspartate residues (reviewed in Ref. 5). To date ten caspases have been identified. They are normally present as single polypeptide inactive zymogens and require cleavage at internal aspartic acid residues to generate the 2-subunit catalytically competent protease. The best understood pathway leading to caspase activation is engagement of the death receptor, CD-95 (Fas/APO-1). Upon activation, an adapter molecule termed FADD is re-cruited to the cytoplasmic segment of the Fas receptor through a homophilic interaction between a stretch of 60 -80 amino acids, dubbed the death domain that is present in both molecules. The N-terminal segment of FADD encodes a distinct binding module termed the death effector domain (6). Remarkably, an equivalent domain, is present within the prodomain of FLICE (caspase-8). Interaction between the death effector domains allows for the recruitment of the death protease FLICE to the receptor signaling complex. Following conversion to the active dimeric species, FLICE is free to proteolytically activate other downstream zymogen caspases (7), leading to cleavage of death substrates and subsequent apoptotic demise.
The ability of enucleated cells to mount a UV response has led to the suggestion that membrane or cytosolic events likely mediate the response (8). UV exposure for example induces rapid tyrosine phosphorylation of the EGF receptor, suggesting a prominent role for membrane initiated events in the UV response (9). Additionally, it has recently been demonstrated that UV-irradiation results in clustering and subsequent activation of the EGF, IL-1, and TNF receptors, and this contributes to the activation of the JNK cascade observed within minutes of exposure to UV radiation (10). Given this, we hypothesized that UV-irradiation may induce apoptosis by activation of cell surface death receptors, the prototypic example being CD-95.

MATERIALS AND METHODS
Cell Lines and Culture Conditions-The MCF7 and BJAB stably transfected cell lines used in this study have been described previously (11). The cells were maintained in RPMI 1640 containing 10% heatinactivated fetal calf serum, 1% nonessential amino acids, 1% L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin at 37°C in an atmosphere of 5% CO 2 . The transfected MCF7 cell lines were grown in the presence of 0.5 mg/ml G418 sulfate (Life Technologies, Inc.), and the transfected BJAB cell lines were grown in 3 mg/ml G418 sulfate.
Irradiations-Freshly seeded cells were irradiated in at room temperature without growth media with a germicidal UV light (254 nm). The fluence of the UV light source was measured prior to each experiment with a UVX radiometer (UVP, Inc.). Following irradiation, the cells were supplied with media and incubated for the indicated time.
Morphological Analysis-Apoptotic morphology was assessed by using DNA-staining dyes. For propidium iodide (Sigma) staining, MCF7 cells were plated at a density of 5 ϫ 10 4 cells onto chamber slides (Nunc, Inc.). Following irradiation, both treated and control slides were incubated for 48 h prior to staining. Slides were rinsed 2 times with PBS, fixed in 4% paraformaldehyde at room temperature for 30 min, rinsed 3 times with PBS, and stained at room temperature for 10 min in a 100 g/ml solution of propidium iodide in PBS. After staining, the slides were rinsed 2 times with PBS, blotted dry, and mounted using Vectashield mounting medium (Vector Laboratories). BJAB cells (1-2 ϫ 10 6 ) were stained using Hoechst 33342 (Sigma). After a 48-h incubation, both control and irradiated cells were rinsed 2 times with PBS, stained at room temperature for 30 min in a 0.8 mg/ml solution of Hoechst 33342 in PBS, rinsed 2 times with PBS, and wet mounted using 20 l of the cell suspension. Slides were examined immediately after staining using a Leitz Laborlux S microscope.
Quantitative Analysis-Apoptotic cells were quantitated based on * This work was partly supported by a Michigan Memorial-Phoenix Project award (to A. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ Fellow of the Medical Scientist Training Program supported by the Experimental Immunopathology Training Grant NIGMS T32.
nuclear morphology using fluorescence microscopy, and the percentage of apoptotic cells was calculated. A minimum of 100 cells were counted for each sample, and each experiment was done in triplicate.
Pulse Field Gel Electrophoresis (PFGE)-Logarithmically growing cells were irradiated with 10 joules of UV radiation and incubated at 37°C. Cells were washed once in PBS and resuspended at 2 ϫ 10 7 cells/ml. They were then mixed with an equal volume of a 1.4% suspension of low melting point agarose (Bio-Rad). The mixture was poured in a mold and allowed to solidify. The agar blocks were treated for 24 h at 56°C with lysis solution (100 mM EDTA, 10 mM Tris, 20 mM NaCl) containing proteinase K (1 mg/ml) and Sarkosyl (1%), after which lysis solution was replaced with storage buffer (50 mM EDTA, 10 mM Tris).
PFGE was performed using a CHEF Mapper (Bio-Rad) at 10°C in the 2-state mode with linear ramping. State 1 was 30 h with a 120°included angle, 1.9 V/cm, and a switching interval of 90 s. State 2 was 51 h with an included angle of 120°, 1.9 V/cm, and a switching interval of 40 min. The gel was then stained with ethidium bromide (5 g/ml) and photographed.
CD-95 Cross-linking and Immunoprecipitation-15 million BJAB cells were mock irradiated or treated with 15 or 30 joules of UV irradiation and incubated for 20 min. Cells were pelleted and washed twice in PBS and resuspended in 500 l of PBS and treated with 2 mM cleavable cross-linker 3,3Ј-dithiobis[sulfosuccinimidyl proprionate] (Pierce) for 15 min on ice, and the reaction was quenched with 10 mM The basic principle is that, when antibody is limiting, more molecules of the Fas-receptor will be immunoprecipitated if aggregation occurs after UV-treatment. In contrast, when antibody is not limiting, equal amounts of Fas-receptor should be immunoprecipitated. B, BJAB cells that were unirradiated controls or irradiated with 15 and 30 joules were treated with the cross-linking reagent 3,3Ј-dithiobis[sulfosuccinimidyl proprionate] and lysed for immunoprecipitation using a Fas-specific antibody (ApoI) as described above under antibody (Ab) limiting conditions (top panel) or antibody excess conditions (bottom panel). Western blot analysis of the immunoprecipitate was done using a Fas-specific rabbit polyclonal antibody (Santa Cruz). C, control and UV-irradiated BJAB cells (30 joules) were fixed with 4% paraformaldehyde and stained for indirect immunofluorescense using a Fas-specific antibody as described above. Cells were viewed and photographed using a confocal microscope with a ϫ 60 objective. ammonium acetate for 10 min. The cells were pelleted and washed twice in PBS and lysed using 500 l of lysis buffer (20 mM Tris (pH 7.4), 140 mM NaCl, 10% glycerol, 1% Triton X-100, and 2 mM EDTA) containing a protease inhibitor mixture (Boehringer Mannheim) on ice for 30 min with agitation. The lysates were then used for immunoprecipitations in the presence of an anti-Fas antibody (Apo-1, Kamiya Labs, CA) at 0.5 g/ml (antibody limiting) or 10 g/ml (antibody excess). Immune complexes were precipitated using protein A-Sepharose (Pharmacia Biotech Inc.) and washed three times in lysis buffer. The precipitate was resuspended in Laemmli buffer, boiled for 5 min, and resolved on SDS-polyacrylamide gel electrophoresis. The resolved samples were transferred onto a polyvinylidene difluoride (Millipore Corp.) membrane for Western blot analysis. The presence of CD-95 or associated molecules was then detected using the appropriate primary and secondary antibodies followed by detection using chemiluminescence (Pierce).
Analysis of CD-95 Using Immunofluorescence-BJAB cells were UV irradiated with 30 joules or were mock-irradiated. After 20 min, they were fixed with 4% paraformaldehyde and stained for immunofluorescence using an anti-CD-95 antibody (Transduction Laboratories) followed by an fluorescein isothiocyanate-conjugated secondary antibody. Fas-receptor specific immunostaining was visualized and photographed using a Bio-Rad MRC-1000 confocal microscope with a ϫ 60 objective.

RESULTS AND DISCUSSION
Ligand dependent activation of CD-95 requires multimerization though ligand independent activation can occur upon oligomerization following overexpression (12). To test the hypothesis that UV-induced apoptosis used components of the CD-95 death pathway, we analyzed MCF-7 breast carcinoma cells and its transfected derivative line MCF-7-Fas, which constitutively expresses approximately 5-fold more CD-95 receptor (12). MCF-7-Fas cells were significantly more sensitive to UV-induced apoptosis in a time-and dose-dependent manner (Fig. 1,  A and B). To examine if UV irradiation induced CD-95 multimerization, we utilized an immunoprecipitation protocol (Fig.  2, A and B) employing limiting antibody conditions that would preferentially immunoprecipitate oligomerized receptor. At limiting antibody concentrations, there appeared to be a linear relationship between the amount of CD-95 immunoprecipitated and the UV dose. As expected, this relationship was not observed under conditions of antibody excess (Fig. 2B). UVinduced CD-95 multimerization was not a cell line-specific phenomenon as it was also observed in both BJAB and Jurkat cells (data not shown). Indirect immunofluorescence experiments using CD-95-specific antibodies were also performed to visualize CD-95 oligomerization upon UV irradiation (Fig. 2C). In the absence of treatment, CD-95 was visualized heterogenously on the cell surface. Upon treatment with UV, CD-95 immunoreactivity was detected as very large, brightly staining aggregates.
To investigate if UV irradiation-mediated multimerization of CD-95 resulted in recruitment of the adapter molecule FADD, co-immunoprecipitation experiments were performed (Fig. 3). Immunoprecipitation of CD-95 followed by immunoblotting using a FADD-specific antiserum, revealed a UV dose-dependent increase in the association of FADD with CD-95. This phenomenon was observed in Jurkat (data not shown) and BJAB cells.
FADD serves as a conduit for death signals from other death receptors including TNFR1 and DR3 (13). A dominant negative version of FADD (FADD-DN) that lacks the death effector domain and is therefore unable to recruit FLICE inhibits cell death induced by all three receptors (CD95, TNFR1, and DR3). If UV irradiation was inducing apoptosis by activating death receptors, then the observed cell death should be inhibited by FADD-DN. Indeed, cells expressing FADD-DN (MCF7-FADD-DN) were significantly more resistant to UV-induced apoptosis than their vector-transfected counterparts (Fig. 4, A-C). Fortyeight h following UV irradiation, vector-transfected MCF-7 cells had morphological features typical of apoptosis (Fig. 4A,  top panel), including cytoplasmic shrinkage and nuclear condensation. The ability of FADD-DN to protect from UV-induced apoptosis was also observed with BJAB cells (Fig. 4D). In contrast to differences seen with UV irradiation, the sensitivity to apoptosis of CD-95-overexpressing cell lines, as well as FADD-DN-expressing cells, did not differ significantly from control cell lines upon ␥ irradiation (Fig. 4E). These results indicate that CD-95 and FADD are not involved in ␥ radiationinduced apoptosis.
Consistent with the involvement of the FADD-FLICE axis in UV irradiation-induced cell death was finding the CrmA, a poxvirus-encoded serpin that preferentially inhibits FLICEblocked UV-induced cell death. Expression of CrmA, but not an active site mutant, CrmA-mut (11), resulted in attenuation of UV-induced apoptosis as determined by inhibition of DNA fragmentation (Fig. 5).
The results presented here for the first time delineate the pathway involved in UV-induced apoptosis. UV-mediated olimerization of CD-95 and likely other death receptors is the initiating event that triggers the downstream FADD-FLICE death effector pathway. The mechanism by which UV irradiation induces receptor oligomerization remains unclear. We speculate, however, that energy transfer to cell surface receptors may induce conformational changes that allow for oligomerization and engagement of the downstream signaling machinery.