H2O2 Is an Important Mediator of UVB-Induced EGF-ReceptorPhosphorylation in Cultured Keratinocytes

Dominik Peus, Remus A. Vasa, Alexander Meves, Markus Pott, Astrid Beyerle, Karen Squillace, andMark R. PittelkowDepartments of Dermatology, and Biochemistry and Molecular Biology, Mayo Clinic/Foundation, Rochester, Minnesota, U.S.A.

Exposure of human keratinocytes to physiologic doses ofultraviolet B (UVB) radiation induces phosphorylationof the epidermal growth factor receptor (EGFR). Wedemonstrate that H2O2 generated by UVB mediatesEGFR phosphorylation. Using dihydrorhodamine 123 asa specific fluorescent dye probe, we show that UVBirradiation (50–800 J per m2) of keratinocytes leads withinminutes to concentration-dependent intracellular pro-duction of H2O2. A corresponding concentration-dependent increase in the release of extracellular H2O2was measured by using Amplex, a derivative of dihydro-phenoxazine. The levels of intracellular H2O2 that areinduced by UVB irradiation and that stimulate EGFRphosphorylation correlate strongly with the responseinduced by exogenously added H2O2. UVB or H2O2demonstrated concentration- and time-dependent stimu-lation of EGFR phosphorylation that was initiallyobserved within 1–5 min and exhibited a proportionate

The human population is intermittently but chronicallyexposed to ultraviolet radiation (UVR) from the sun.UVR is primarily responsible for more than 1,000,000cutaneous malignancies each year in the U.S.A. alone,making it the most efficient environmental carcinogen

known (Miller and Weinstock, 1994). Acute and chronic effects ofUVR on skin include inflammation, hyperpigmentation, hyperplasia,and skin cancer (Norris et al, 1993; Young, 1993; Gilchrest et al, 1996).UVR has been identified as a complete carcinogen in the skin and, inconjunction with appropriate chemical agents, has been demonstratedto act as both a tumor initiator and a tumor promoter (Epstein, 1978;Matsui and DeLeo, 1991). In addition, skin exposure to ultraviolet B(UVB) wavelength radiation, a minor but active constituent of sunlight,induces direct and indirect biologic effects, including DNA damageand mutations (Sachsenmaier et al, 1994b), systemic immunosuppression(Kripke, 1994), activation of dormant viruses (Herrlich et al, 1992),induction of early and late gene responses (Fornace et al, 1992;Sachsenmaier et al, 1994b; Huang and Adamson, 1995), and cell cyclegrowth arrest (Medrano et al, 1995).

Manuscript received June 9, 1997; revised February 3, 1998; accepted forpublication February 10, 1998.

Reprint requests to: Dr. Mark R. Pittelkow, Mayo Clinic, DermatologyResearch, Gugg. 442D, 200 First Street SW, Rochester, MN 55905.

Abbreviations: BHA, butylated hydroxyanisole; DHR, dihydrorhodamine123; EGFR, epidermal growth factor receptor; NAC, N-acetyl-L-cysteine;PDTC, pyrrolidine dithiocarbamate.

0022-202X/98/$10.50 · Copyright © 1998 by The Society for Investigative Dermatology, Inc.

966

delay for UVB-induced production of H2O2. EGFRphosphorylation induced by UVB or H2O2 declined sig-nificantly toward baseline levels by 4 h and could berestimulated after H2O2 but not after UVB exposure.Phosphorylation of EGFR was inhibited by the structur-ally unrelated antioxidants butylated hydroxyanisole,N-acetyl-L-cysteine, and pyrrolidine dithiocarbamate, orby the H2O2-degrading enzyme catalase. These dataindicate that generation of H2O2 by UVB radiation ofhuman keratinocytes participates in the rapid, ligand-independent phosphorylation of EGFR and implicateH2O2 as a biologic mediator in EGFR activation andregulation of the downstream signaling cascade. UVB-induced H2O2 has the potential to initiate or modulateearly EGFR-mediated signaling events that could play animportant role in the cellular response to oxidative stress.Key word: tyrosine kinase phosphorylation. J Invest Dermatol110:966–971, 1998

The complexity and interactions of cellular responses to UVB andUVR, however, have been only partially delineated. The early signalingevents regulated by growth factor tyrosine kinase receptors appear tobe triggered by highly sensitive intracellular mechanisms within minutesof UVR exposure (Rosette and Karin, 1996). The mediators linkingphoton absorption to activation of signal transduction pathways inUVR-exposed cells are particularly important but remain ill-defined.

The epidermal growth factor receptor (EGFR) is a well characterizedtransmembrane glycoprotein of Mr 180,000 (Carpenter and Cohen,1990; Carpenter, 1993). Upon binding EGF or related ligands, theEGFR cytoplasmic domain is autophosphorylated on tyrosine residuesand initiates multiple signaling responses associated with mitogenesisand cell regulation (Yarden and Ullrich, 1988; Ullrich and Schlessinger,1990; Hunter, 1995). Irradiation of mesenchymal cells with ultravioletC (UVC) wavelength radiation also activates signal transduction path-ways that extend from the cell membrane to the nucleus (Sachsenmaieret al, 1994a; for reviews, see Herrlich et al, 1994; Angel, 1995; Benderet al, 1997).

The current investigations stem from the observations ofSachsenmaier et al (1994a), which demonstrate phosphorylation ofEGFR by UVC and inhibition by pretreatment with EGF, suramin,or expression of a dominant-negative EGFR mutation. EGFRphosphorylation has been shown to be induced by UVA, UVB, orUVC in various immortalized cell lines (Zheng et al, 1993; Miller et al,1994; Warmuth et al, 1994; Sachsenmaier et al, 1994a, b; Coffer et al,1995; Knebel et al, 1996; Huang et al, 1996; Rosette and Karin, 1996),but the biochemical mechanisms mediating this response have notbeen elucidated.

VOL. 110, NO. 6 JUNE 1998 UVB INDUCES EGF-RECEPTOR PHOSPHORYLATION VIA H2O2 967

Reactive oxygen species (ROS) are constitutively produced in mostcell types by specific metabolic processes such as enzymatic oxidationsand aerobic respiration (Cerutti, 1985; Darr and Fridovich, 1994).When cellular levels of ROS exceed the counter-regulatory antioxidantsystem, the redox balance within cells is perturbed, and oxidative stressresults. UVR is a potent inducer of ROS, including superoxide radical(•O2

–), hydrogen peroxide (H2O2), and hydroxyl radical (•OH), whichhave been shown to exert potent activities in stimulating cellulardamage. ROS have been implicated in cutaneous aging as well as thepathogenesis of diseases such as cancer and inflammatory disorders(Fridovich, 1978; Black, 1987; Halliwell, 1989; Floyd, 1990; Janssenet al, 1993; Tyrrell, 1994).

ROS have recently been suggested to mediate induction of growthfactor tyrosine kinase phosphorylation following UVC exposure (Huanget al, 1996). Interestingly, H2O2 has been found to enhance EGFRtyrosine phosphorylation in human squamous carcinoma and other celllines, including NA, A431, HeLa, and A549 (Gamou and Shimizuet al, 1995; Huang et al, 1996; Knebel et al, 1996; Rao, 1996). Theseobservations suggest that oxidative stress may be involved in EGFRphosphorylation. This study focuses on generation of H2O2 in normalhuman keratinocytes by physiologic doses of UVB radiation. We alsoexamine the induction of EGFR phosphorylation by UVB and H2O2.

MATERIALS AND METHODS

Materials and chemicals The UVB source was a FS20 lamp (Westinghouse,Pittsburgh, PA) that emits an energy spectrum with high fluence in the UVBregion and a peak at 313 nm (Toews et al, 1980). The emitted dose wasregularly quantitated by a UVB radiometer and photodetector (IL 443 and SEE240, International Light, Newburyport, MA). N-acetyl-L-cysteine (NAC),butylated hydroxyanisole (BHA), pyrrolidine dithiocarbamate (PDTC), amino-triazole, and bovine catalase were purchased from Sigma (St. Louis, MO).Amplex Red reagent and dihydrorhodamine 123 (DHR) were obtained fromMolecular Probes (Eugene, OR). DHR detects mainly H2O2 productionfollowing conversion to the oxidized product, rhodamine 123, and very littleoxidized compound leaks from cells compared with other probes such asdichlorofluorescein (Heck et al, 1992; Oyama et al, 1994; Los et al, 1995;Sundaresan et al, 1995).

Culture and treatment of normal human keratinocytes Human ker-atinocytes were isolated from neonatal foreskin specimens, and primary cultureswere initiated and maintained in a replicative state with complete, serum freeMCDB 153 medium. Subconfluent keratinocytes from primary cultures wereplated into secondary culture at 1–10 3 103 cells per cm2 and used atsubconfluence or confluence as determined by phase-microscopic observation.Complete medium was supplemented with 0.1 mM calcium, 0.2% (vol/vol)bovine pituitary extract, EGF (10 ng per ml), insulin (5 µg per ml), hydro-cortisone (5 3 10–7 M), ethanolamine (1 3 10–4 M), phosphoethanolamine(1 3 10–4 M), and supplemented amino acids (Wille et al, 1984; Pittelkow andScott, 1986). Cells were grown to confluence, washed, and incubated instandard medium (without growth factors) for 48 h to measure EGFRphosphorylation following UVB irradiation or H2O2 treatment. No differenceswere observed whether cells were irradiated in phosphate buffered saline or instandard culture medium.

Detection of intracellular H2O2 by flow cytometric analysis Intracellularlevels of H2O2 were analyzed by flow cytometry using DHR as a specificfluorescent dye probe (Royal and Ischiropoulos, 1993; Chen et al, 1995).Keratinocytes were loaded with DHR (1 µM) for 45 min before treatmentwith UVB. During the intracellular release of H2O2, reduced DHR is irreversiblyoxidized and converted to the red fluorescent compound rhodamine 123(Rothe et al, 1991). Keratinocytes were detached from the culture plate bytrypsin, and enzyme activity was quenched with 2% fetal calf serum in phosphatebuffered saline. Keratinocytes were fixed in 1% paraformaldehyde (Sigma),placed on ice, and the intracellular rhodamine 123 fluorescence intensity of10,000 cells was measured for each sample using a Becton-Dickenson FACSVantage flow cytometer with the excitation source at 488 nm and emissiondetected with a 580 nm long pass filter. Histograms were analyzed with thesoftware program PC-Lysis (Becton Dickenson).

Extracellular hydrogen peroxide assay Measurement of extracellular H2O2following UVB exposure was performed using Amplex, a derivative ofdehydrophenoxazine, which becomes highly fluorescent upon oxidation by

H2O2 in the presence of horseradish peroxidase. Fluorescence was detectedusing a Cytofluor II microplate fluorometer (Perspectives Biosystems,Framingham, MA). The excitation/emission wavelength filter was set at 590/615. Fluorescence intensity of Amplex was measured within the linear regionof the H2O2 concentration–response curve. Wells without cells containing thesame substrates and exposed to the same treatment represented controls. Thecontrol fluorescent values and background fluorescence of untreated cells weremonitored over time and subtracted from the relative fluorescence intensity ofexperimental samples. These values were converted into absolute pmoles usingan H2O2 standard curve. All H2O2 measurements were performed in 24 wellplates containing confluent keratinocytes, 100 µM Amplex, and 1 U horseradishperoxidase per ml (Mohanty et al, 1997).

Immunoprecipitation, immunoblot, and tyrosine phosphorylationassays Cell lysates were obtained by scraping cultures into 1 ml of Frackeltonbuffer, centrifugation for 5 min, and adding to the supernatant monoclonalantibody LA1 to EGFR (Upstate Biotechnology, Lake Placid, NY) coupled toProtein G-Plus Agarose beads (Oncogene, Cambridge, MA). Beads were rockedfor several hours, repeatedly washed, released after adding 20 µl of samplebuffer by incubation at 100°C for 5 min, and separated by centrifugation. Thesamples were then separated electrophoretically by 10% sodium dodecylsulfate-polyacrylamide gel electrophoresis. Proteins were electro-transferred toImmobilon-P (Millipore, Bedford, MA), and the membrane was blocked with2% bovine serum albumin before addition of antiphosphotyrosine antibody(PY20) coupled to peroxidase (Transduction Laboratories, Lexington, KY).Total EGFR was quantitated by immunoblotting each immunoprecipitationmembrane with EGFR monoclonal antibody 15E11 specific for peptide epitopesof EGFR extracellular domain (Baron et al, 1997). The bands were visualizedusing the ECL detection system (Amersham, Arlington Heights, IL).

RESULTS

H2O2 is dose-dependently induced by UVB H2O2 productionin keratinocytes was analyzed by DHR loading followed by irradiationwith UVB over a range of doses and incubation for 20 min. Comparedwith unirradiated control cells, 100–800 J UVB per m2 induced adose-dependent increase in relative fluorescent intensity (Fig 1A) thatdemonstrated intracellular H2O2 generation. These results correlatedclosely with extracellular release of H2O2 30 min after UVB exposureas measured by the in situ assay using Amplex (Fig 1B).

Correlation of DHR fluorescence after UVB or H2O2 treatmentand inhibition of UVB-induced H2O2 by antioxidants PDTCor BHA Using DHR as an indicator of intracellular H2O2 levels,we found that treatment of keratinocytes with 60–80 µM H2O2resulted in DHR fluorescence similar to 200 J UVB per m2, andexposure to 100 µM H2O2 induced fluorescence equivalent in intensityto 400 J per m2 (Fig 2A). DHR fluorescent intensity of cells exposedto 200 J UVB per m2 was specifically inhibited by 50 µM PDTC(Fig 2B) or 50 µM BHA (Fig 2C). Baseline levels of fluorescence inuntreated cells were also decreased by these antioxidants. These findingsprovide evidence that H2O2- and UVB-induced DHR fluorescenceare strongly correlated, and structurally different antioxidants modulateH2O2 levels in UVB-stimulated cells.

H2O2 and UVB phosphorylate EGFR in a time- and dose-dependent fashion The biologic activity of UVB or H2O2 wasmeasured by phosphorylation of EGFR in keratinocytes. Cells wereexposed to 200 µM H2O2 for 1–20 min or irradiated with 400 J UVBper m2 and incubated for 5–20 min (Fig 3). Within 1 min of exposure,the level of EGFR phosphorylation was markedly enhanced by H2O2,and maximum EGFR phosphorylation was observed by 5 min. UVBirradiation induced a modest increase in EGFR phosphorylation at5 min, and the level of EGFR phosphorylation was further enhancedby 10–20 min. Immunoblotting with EGFR monoclonal antibody15E11 revealed equal loading of protein (Fig 3). A dose-dependentincrease in EGFR phosphorylation was observed over the range of 3–300 µm H2O2 (Fig 4A). Increasing doses of UVB (50–800 J per m2)proportionately enhanced levels of phosphorylated EGFR (Fig 4B).The rapid onset of UVB- or H2O2-induced EGFR phosphorylationimplicates a ligand-independent mechanism of activity.

968 PEUS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 1. Dose-dependent intracellular H2O2 production andextracellular release of H2O2 after UVB exposure. DHR-loaded,subconfluent keratinocytes were irradiated with UVB, released by trypsinization,fixed in 0.5% formaldehyde, and analyzed by flow cytometry as described inthe Materials and Methods. (a) Dose-dependent increase in relative fluorescencefollowing increasing UVB doses and incubation for 30 min. (b) Amplex(100 µM), a dihydrophenoxazine derivative, and horseradish peroxidase (1 Uper ml) were added to confluent keratinocytes in 24 well plates. Thirty minutesafter exposure to different doses of UVB, fluorescence intensity was measuredby Cytofluor II and converted into absolute picomoles of H2O2 using an H2O2standard curve.

Time-dependent decrease in H2O2- and UVB-inducedEGFR phosphorylation and differences in restimulationresponses EGFR phosphorylation decreased markedly by 4 h aftertreatment with H2O2 or UVB (Fig 5A, B). 200 µM H2O2 inducedthe maximum increase in EGFR phosphorylation by 10 min, and asteady decline to baseline levels was observed over 4 h (Fig 5A).Following 400 J UVB per m2 exposure, the maximum level of EGFRphosphorylation was reached after µ1 h and declined progressivelythereafter (Fig 5B). Baseline levels of phosphorylated EGFR variedamong experiments due to cell density and autocrine activity (Peuset al, 1997). H2O2 restimulation strongly enhanced EGFR phosphoryl-ation in keratinocytes (Fig 5A); however, UVB-exposed keratinocytesexhibited a higher baseline level of EGFR phosphorylation thatpersisted at 4 h and showed no significant enhancement in responseto UVB restimulation (Fig 5C).

Catalase decreases and aminotriazole increases EGFR phos-phorylation after UVB Exposure of keratinocytes to 400 J UVBper m2 following 1 h pretreatment with 300 U catalase, a H2O2

Figure 2. Correlation between H2O2- and UVB-induced DHRfluorescence intensity and inhibition by antioxidants PDTC and BHA.DHR-loaded confluent keratinocytes were treated with different doses of H2O2or UVB (a). Pretreatment of cells with (b) PDTC (50 µM) or (c) BHA (50 µM)for 30 min and exposure to UVB (200 J per m2) resulted in decreased relativefluorescence intensity 30 min after UVB irradiation. Error bars, 6 SD.

VOL. 110, NO. 6 JUNE 1998 UVB INDUCES EGF-RECEPTOR PHOSPHORYLATION VIA H2O2 969

Figure 3. H2O2 and UVB phosphorylate EGFR in a time-dependentmanner. Confluent, resting human keratinocytes were exposed to H2O2(200 µM) or UVB (400 J per m2) and incubated for designated times.Total cell protein was extracted, immunoprecipitated with monoclonal EGFRantibody, separated by sodium dodecyl sulfate-polyacrylamide gelelectrophoresis, transferred to nitrocellulose membranes, and immunoblottedwith horseradish peroxidase coupled anti-phosphotyrosine antibody, PY20.Autophosphorylation of EGFR was stimulated by EGF (10 ng per ml) for20 min. Equal loading was demonstrated by immunoblotting with EGFRmonoclonal antibody 15E11 after stripping of PY20 from membrane.

Figure 4. Dose-dependent EGFR phosphorylation by H2O2 and UVB.Confluent, resting keratinocyte cultures were exposed to H2O2 or UVB, andEGFR phosphorylation was quantitated as described in Fig 3. (a) H2O2 doses,3–300 µM for 10 min treatment; (b) UVB radiation, 50–800 J per m2 and20 min incubation or EGF (10 ng per ml) for 20 min.

metabolizing enzyme, per ml, showed a decrease in the level ofphosphorylated EGFR (Fig 6A). By contrast, inhibition of endogenouscatalase by 2 or 20 mM aminotriazole increased EGFR phosphorylationafter UVB exposure (Fig 6B). Treatment of cells with catalase alonehad no effect and 20 mM aminotriazole alone slightly increased thebaseline level of phosphorylated EGFR (Fig 6B). These findingsprovide further evidence that modulating the levels of H2O2 producedin keratinocytes by UVB exposure exerts an effect on EGFR phos-phorylation.

Inhibition of UVB-induced EGFR phosphorylation by anti-oxidants Keratinocyte cultures were pretreated with the structurallyunrelated antioxidants NAC (30 mM or 3 mM) or PDTC (10 µM)prior to 200 J UVB per m2 exposure. These agents partially inhibitedUVB-induced EGFR phosphorylation (Fig 7A). Another antioxidant,BHA (50 µM), also inhibited UVB-induced EGFR phosphorylation(Fig 7B). Addition of both NAC and PDTC or BHA resulted inadditive inhibition (Fig 7C). Non-irradiated cells treated with NAC,PDTC, or BHA in combination or alone, showed no significantdifference in the levels of phosphorylated EGF compared with control.The inhibition of EGFR phosphorylation by antioxidants indicatesthat this process is dependent on ROS generation and the cellularredox state.

Figure 5. Time-dependent decrease and restimulation of H2O2- andUVB-induced EGFR phosphorylation. Resting, confluent keratinocyteswere exposed to: (a) H2O2 (200 µM) for designated times or EGF (10 ng perml) for 20 min, 4 h later H2O2-stimulated cells were restimulated for 10 minwith the same dose of H2O2; or (b) UVB (400 J per m2) exposure andincubation for designated times or EGF (10 ng per ml). (c) Four hours afterUVB (400 J per m2) exposure, cells were restimulated for 20 min with thesame dose of UVB. EGFR phosphorylation analysis was performed as describedin Fig 3.

DISCUSSION

The chemical mediators of various cellular responses to UVR are onlybeginning to be identified and characterized. Our findings provideseveral lines of evidence to indicate that H2O2 functions as a prominentmediator of UVB-induced EGFR phosphorylation. Firstly, EGFRphosphorylation is inhibited by structurally unrelated antioxidantreagents, including PDTC, BHA, and NAC. Secondly, keratinocytestreated with catalase show a substantial reduction of EGFR phosphoryl-ation, whereas aminotriazole, a catalase inhibitor, enhances phosphoryl-ated levels of EGFR. Thirdly, concentrations of exogenous H2O2correlate closely with the intensity of EGFR phosphorylation andconcentrations of intracellular H2O2 that are generated. Fourthly, thetime-course of UVB-induced generation of H2O2 as measured byspecific fluorescence of oxidized DHR correlates closely with thekinetics of increase in EGFR phosphorylation; and lastly, H2O2 directlyand rapidly induces phosphorylation of EGFR, mimicking selectedeffects of UVB.

UVB doses of 50–800 J per m2 stimulate the keratinocyte tophosphorylate EGFR, and these levels of irradiation represent exposuresthat typically induce the physiologic response of solar erythema in skintypes I–III (Norris et al, 1993). This range of UVB irradiation alsoroutinely produces a minimal erythema dose (MED) under standardskin testing conditions. Therefore, the dose range of UVB that inducesEGFR phosphorylation in cultured keratinocytes has physiologicrelevance for epidermis and skin. Based on the low doses of UVB(50–100 J per m2) required to stimulate ligand-independent EGFRphosphorylation, we hypothesize that this keratinocyte response is acellular defence mechanism rather than a toxic insult and serves aprotective role in epidermis.

Generation of ROS has been more extensively studied after UVAradiation than after UVB irradiation. In fibroblasts, production of ROS

970 PEUS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 6. Effects of catalase and aminotriazole on EGFRphosphorylation after UVB. Confluent, resting keratinocytes were pretreatedwith: (a) catalase (cat, 300 U per ml) for 1 h (lanes 3, 4) before UVB (400 Jper m2) exposure for 20 min (lanes 2, 3); or (b) the catalase inhibitor,aminotriazole (20 or 2 mM), for 20 min before UVB exposure (200 J per m2)for 20 min. EGFR phosphorylation analysis was performed as described in Fig 3.

such as singlet oxygen and H2O2 has been reported (Tyrrell andPidoux, 1989; Vile and Tyrrell, 1995); however, little is known aboutthe generation of ROS in keratinocytes. These studies begin to addressthis gap by demonstrating H2O2 generation after UVB irradiation inkeratinocytes. Intracellular production of H2O2, as a cell-generatedmediator of EGFR phosphorylation in response to UVB irradiation,was identified using flow cytometry and cell loading of DHR, a highlysensitive fluorescent dye probe. Within 20 min of UVB exposure, asignificant increase in H2O2 generation was measured in DHR-loadedkeratinocytes that was inhibited by PDTC or BHA. These antioxidantsare also able to inhibit phosphorylation of EGFR that demonstratesthat this process is dependent on ROS and indicates that H2O2 maybe involved as a mediator. In addition, for the induction of DHRfluorescence, similar doses of UVB and H2O2 are needed that arenecessary to induce EGFR phosphorylation. In parallel with thegeneration of intracellular H2O2, we also found a concentration-dependent increase of the release of H2O2 as measured by Amplex.Even though H2O2 freely crosses cell membranes, only a fraction ofintracellularly generated H2O2 is likely to be released. Therefore,intracellular H2O2 may be significantly higher and/or more effectivebecause the site of generation may be in close proximity to the siteof action.

Several recent investigations have examined UVR-induced receptorphosphorylation events and have speculated on the cellular mechanismsthat mediate these responses. Rosette and Karin (1996) showed thatUVB exposure induced receptor clustering followed by internalization.The response was assumed to result from physical perturbation of theplasma membrane or conformational changes in membrane proteinscaused by UVR energy absorption, but a potential role for ROSgenerated by UVB-treated cells was not addressed. In this regard, theearly responses of ligand-induced receptor clustering and aggregationthat is believed to be crucial for transducing a signal (Metzger, 1992),have not been previously elucidated as UVR-mediated mechanisms(Anderson, 1993).

UVR-induced EGFR phosphorylation has been shown to bepredominantly a cell membrane-associated event, rather than arisingfrom the nucleus as had previously been suggested (Sachsenmaier et al,1994b; Herrlich et al, 1994; Huang et al, 1996). UVC as well as UVBstimulated EGFR phosphorylation of isolated cell membranes (Knebel

Figure 7. Antioxidants NAC, PDTC, and BHA inhibit UVB-inducedEGFR phosphorylation. Keratinocytes were irradiated with UVB (200 J perm2) after pretreatment with (a) NAC (30 mM, lane 3, or 3 mM, lane 4) orPDTC (10 µM) for 30 min or (b) BHA (50 µM) for 30 min before UVBexposure and incubation for 20 min. (c) Pretreatment with a combination ofNAC (10 mM) and PDTC (10 µM) or BHA (50 µM) for 30 min before UVB(200 J per m2) exposure and incubation for 30 min exerted additive inhibitionof UVB-induced EGFR phosphorylation. NAC was added 8 h before UVBexposure. EGFR phosphorylation analysis was performed as described in Fig 3.

et al, 1996). The response is independent of ligand binding asshown by EGFR containing a truncated extracellular domain that stillundergoes phosphorylation following UVR (Knebel et al, 1996).Knebel et al also demonstrated that the inhibition of membrane-boundprotein tyrosine phosphatases by UVR and H2O2 may also be a targetto modulate EGFR phosphorylation.

UVB-induced phosphorylation of EGFR is similar to ligand-inducedautophosphorylation of EGFR (Peus et al, 1997); however, a recentreport demonstrates distinct differences in specific sites of phosphoryl-ation of EGFR induced by H2O2 versus the peptide ligand, EGF(Gamou and Shimizu et al, 1995). The relative kinase activity andbiologic potency of phosphorylated EGFR induced by EGF versusH2O2 remain to be determined. Furthermore, it is not yet knownwhether specific EGFR phosphorylation sites induced by UVB areidentical or significantly overlap with those sites phosphorylated byH2O2. In this regard, the role of other ROS, such as •OH and •O2

–

in phosphorylating EGFR and other receptor tyrosine kinases, alsorequire further investigation. These findings imply that UVR- andH2O2-mediated phosphorylation of EGFR are, in large part, rapid,membrane-associated responses induced by oxidative stress.

Additional transmembrane receptor targets such as FGFR andinterleukin-1 and tumor necrosis factor-α receptors of HeLa cells, havealso been shown to be rapidly phosphorylated by UVB, UVC, andosmotic stress (Sachsenmaier et al, 1994a; Rosette and Karin, 1996).Growth factor receptors other than EGFR, such as platelet-derivedgrowth factor receptor, may also activate signalling responses via H2O2or other ROS (Sundaresan et al, 1995). The activation or modulationof multiple receptor families, including T cell receptor-coupled tyrosinekinases (Schieven et al, 1994) in the absence of ligand binding, havethe potential to be mediated by oxidative molecules and likelycontribute to the initiation of specific signaling cascades and stress-

VOL. 110, NO. 6 JUNE 1998 UVB INDUCES EGF-RECEPTOR PHOSPHORYLATION VIA H2O2 971

related responses within cells (Bender et al, 1997). Osmotic stress aswell as UVR are known inducers of ROS. We suggest that ligand-independent H2O2-induced receptor phosphorylation and ‘‘activation’’or modulation of receptor activity may involve multiple cell surfacereceptors in keratinocytes and play roles in coordinating crucialphysiologic responses, such as triggering cellular defence mechanismsor mediating injurious or toxic events.

Based on these findings, we conjecture that the balance betweenpro- and anti-oxidative mechanisms of the cell will determine theactivities of downstream signaling cascades initiated by UVB stimulationof EGFR and other receptor phosphorylation; however, the specificdownstream elements that are activated by UVB- and H2O2-inducedEGFR phosphorylation remain to be identified. Collectively, theseresults provide clearer evidence that H2O2 generated by keratinocytesacts as a proximate second messenger to upregulate the phosphorylationof EGFR in response to UVB. These observations are instrumental infocusing future investigations on specific UVR-related intracellularsignalling pathways of keratinocytes and epidermis. Cellular andmolecular mechanisms that underlie acute and chronic skin responsesto UVR, including solar erythema, skin aging, and carcinogenesis,may be mediated, at least in part, by H2O2 and other ROS producedby epidermal keratinocytes.

The excellent technical assistance provided by N.N. Swanson is gratefully acknowledged.We thank Dr. N. Maihle for the gift of EGFR monoclonal antibody 15E11. R. Vasawas funded by a grant from the Heinrich-Hertz Stiftung, Duesseldorf, Germany. D.Peus is supported by a grant (PE 635/1–2) of the Deutsche Forschungsgemeinschaft(DFG). Additional support was provided by the Mayo Foundation.

REFERENCES

Anderson RG: Caveolae: where incoming and outgoing messengers meet. Proc Natl Acad SciUSA 90:10909–10913, 1993

Angel P: The role and regulation of the jun proteins in response to phorbol ester and UVlight. In: Baeuerle PA (ed.). Inducible Gene Expression, Vol. 1. Birkhauser, Boston, 1995,pp. 62–92

Baron AT, Huntley BK, Lafky JM, et al: Monoclonal antibodies specific for peptide epitopesof the epidermal growth factor receptor’s extracellular domain. Hybridoma 16:259–271, 1997

Bender K, Blattner C, Knebel A, Iordanov M, Herrlich P, Rahmsdorf HJ: UV-induced signaltransduction. J Photochem Photobiol 37:1–17, 1997

Black HS: Potential involvement of free radical reactions in ultraviolet light-mediatedcutaneous damage. Photochem Photobiol 46:213–221, 1987

Carpenter G: EGF: new tricks for an old growth factor. Curr Opin Cell Biol 5:261–264, 1993Carpenter G, Cohen S: Epidermal growth factor. J Biol Chem 265:7709–7712, 1990Cerutti PA: Prooxidant states and tumor promotion. Science 227:375–381, 1985Chen Q, Olashaw N, Wu J: Participation of reactive oxygen species in the lysophosphatidic

acid-stimulated mitogen-activated protein kinase kinase activation pathway. J BiolChem 270:28499–28502, 1995

Coffer PJ, Burgering BM, Peppelenbosch MP, Bos JL, Kruijer W: UV activation of receptortyrosine kinase activity. Oncogene 11:561–569, 1995

Darr D, Fridovich I: Free radicals in cutaneous biology. J Invest Dermatol 102:671–675, 1994Epstein JH: Photocarcinogenesis: a review. Natl Cancer Inst Monographs 50:13–25, 1978Floyd RA: Role of oxygen free radicals in carcinogenesis and brain ischemia. FASEB J

4:2587–2597, 1990Fornace AJ Jr, Jackman J, Hollander MC, Hoffman-Liebermann B, Liebermann DA:

Genotoxic-stress response genes and growth-arrest genes, gadd, MyD, and other genesinduced by treatments eliciting growth arrest. Ann NY Acad Sci 663:139–153, 1992

Fridovich I: The biology of oxygen radicals. Science 201:875–880, 1978Gamou S, Shimizu N: Hydrogen peroxide preferentially enhances the tyrosine

phosphorylation of epidermal growth factor receptor. FEBS Lett 357:161–164, 1995Gilchrest BA, Park HY, Eller MS, Yaar M: Mechanisms of ultraviovlet light-induced

pigmentation. Photochem Photobiol 63:1–10, 1996Halliwell B: Free radicals, reactive oxygen species and human disease: a critical evaluation

with special reference to atherosclerosis. Br J Exp Pathol 70:737–757, 1989Heck DE, Laskin DL, Gardner CR, Laskin JD: Epidermal growth factor suppresses nitric

oxide and hydrogen peroxide production by keratinocytes. Potential role for nitricoxide in the regulation of wound healing. J Biol Chem 267:21277–21280, 1992

Herrlich P, Ponta H, Rahmsdorf HJ: DNA damage-induced gene expression: signaltransduction and relation to growth factor signaling. Rev Physiol Biochem Pharmacol119:187–223, 1992

Herrlich P, Sachsenmaier C, Radler-Pohl A, Gebel S, Blattner C, Rahmsdorf HJ: Themammalian UV response: Mechanism of DNA damage induced gene expression. AdvEnzyme Regul 34:381–395, 1994

Huang R-P, Adamson ED: A biological role for Egr-1 in cell survival following ultra-violetirradiation. Oncogene 10:467–475, 1995

Huang R-P, Wu JX, Fan Y, Adamson ED: UV activates growth factor receptors via reactiveoxygen intermediates. J Cell Biol 133:211–220, 1996

Hunter T: Protein kinases and phosphatases: the yin and yang of protein phosphorylation andsignaling. Cell 80:225–236, 1995

Janssen YM, Van Houten B, Borm PJ, Mossman BT: Cell and tissue responses to oxidativedamage. Lab Invest 69:261–274, 1993

Knebel A, Rahmsdorf HJ, Ullrich A, Herrlich P: Dephosphorylation of receptor tyrosinekinases as target of regulation by radiation, oxidants or alkylating agents. EMBO J15:5314–5325, 1996

Kripke ML: Ultraviolet radiation and immunology: Something new under the sun. CancerRes 54:6102–6105, 1994

Los M, Schenk H, Hexel K, Baeuerle PA, Droge W, Schulze-Osthoff K: IL-2 gene expressionand NF-kB activation through CD28 requires reactive oxygen production by. 5-lipoxygenase. EMBO J 14:3731–3740, 1995

Matsui MS, DeLeo VA: Longwave ultraviolet radiation and promotion of skin cancer. CancerCells 3:8–12, 1991

Medrano EE, Im S, Yang F, Abdel-Malek ZA: Ultraviolet B light induces G1 arrest inhumanmelanocytes byprolonged inhibitionof retinoblastomaproteinphosphorylationassociated with long-term expression of the p21Waf-1/SDI-1/Cip-1 protein. CancerRes 55:4047–4052, 1995

Metzger H: Transmembrane signaling: The joy of aggregation. J Immunol 149:1477, 1992Miller CC, Hale P, Pentland AP: Ultraviolet B injury increases prostaglandin synthesis through

a tyrosine kinase-dependent pathway. Evidence for UVB-induced epidermal growthfactor receptor activation. J Biol Chem 269:3529–3533, 1994

Miller DL, Weinstock MA: Nonmelanoma skin cancer in the United States: incidence. J AmAcad Dermatol 30:774–778, 1994

Mohanty JG, Jaffe JS, Schulman ES, Raible DG: A highly sensitive fluorescent micro-assayof H2O2 release from activated human leukocytes using a dihydroxyphenoxazinederivative. J Immunol Meth 202:133–141, 1997

Norris PG, Gange RW, Hawk JLM: Acute effects of ultraviolet radiation on the skin. In:Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF (eds). Dermatology inGeneral Medicine, 4th edn. McGraw Hill, New York, 1993, pp. 1651–1658

Oyama Y, Hayashi A, Ueha T, Maekawa K: Characterization of 29,79-dichlorofluorescinfluorescence in dissociated mammalian brain neurons: Estimation on intracellularcontent of hydrogen peroxide. Brain Res 635:113–117, 1994

Peus P, Hamacher L, Pittelkow MR: EGF-receptor tyrosine kinase inhibition induceskeratinocyte growth arrest and terminal differentiation. J Invest Dermatol 109:751–756, 1997

Pittelkow MR, Scott RE: New techniques for the in vitro culture of human skin keratinocytesand perspectives on their use for grafting of patients with extensive burns. Mayo ClinProc 61:771–777, 1986

Rao GN: Hydrogen peroxide induces complex formation of SHC-Grb2-SOS with receptortyrosine kinase and activates Ras and extracellular signal-regulated protein kinasesgroup of mitogen-activated protein kinases. Oncogene 13:713–719, 1996

Rosette C, Karin M: Ultraviolet light and osmotic stress: Activation of the JNK cascadethrough multiple growth factor and cytokine receptors. Science 173:1194–1197, 1996

Rothe G, Emmendorffer A, Oser A, Roesler J, Valet G: Flow cytometric measurementof the respiratory burst activity of phagocytes using dihydrorhodamine 123 [letter;comment]. J Immunol Meth 138:133–135, 1991

Royal JA, Ischiropoulos H: Evaluation of 29,79-dichlorofluorescin and dihydrorhodamine123 as fluorescent probes for intracellular H2O2 in cultured endothelial cells. ArchBiochem Biophys 302:348–355, 1993

Sachsenmaier C, Radler-Pohl A, Zinck R, Nordheim A, Herrlich P, Rahmsdorf HJ:Involvement of growth factor receptors in the mammalian UVC response. Cell 78:963–972, 1994a

Sachsenmaier C, Radler-Pohl A, Muller A, Herrlich P, Rahmsdorf HJ: Damage to DNA byUV light and activation of transcription factors. Biochem Pharmacol 47:129–136, 1994b

Schieven GL, Mittler RS, Nadler SG, Kirihara JM, Bolen JB, Kanner SB, Ledtetter JA: ZAP-70 tyrosine kinase, CD45 and T-cell receptor involvement in UV and H2O2 inducedsignal transduction. J Biol Chem 269:20718–20726, 1994

Sundaresan M, Yu ZX, Ferrans VJ, Irani K, Finkel T: Requirement for generation of H2O2for platelet-derived growth factor signal transduction. Science 270:296–299, 1995

Toews GB, Bergstresser PR, Streilein JW: Epidermal Langerhans cell density determineswhether contact hypersensitivity or unresponsiveness follows skin painting withDNFB. J Immunol 134:445–453, 1980

Tyrrell RM: The molecular and cellular pathology of solar ultraviolet radiation. Mol AspectsMed 15:1–77, 1994

Tyrrell RM, Pidoux M: Singlet oxygen involvement in the inactivation of cultured humanfibroblasts by UVA (334 nm, 365 nm) and near-visible (405 nm) radiations. PhotochemPhotobiol 49:407–412, 1989

Ullrich A, Schlessinger J: Signal transduction by receptors with tyrosine kinase activity. Cell61:203–212, 1990

Vile GF, Tyrrell RM: UVA radiation-induced oxidative damage to lipids and proteins in vitroand in human skin fibroblasts is dependent on iron and singlet oxygen. Free Radic BiolMed 18:721–730, 1995

Warmuth I, Harth Y, Matsui MS, Wang N, DeLeo VA: Ultraviolet radiation inducesphosphorylation of the epidermal growth factor receptor. Cancer Res 54:374–376, 1994

Wille JJ Jr, Pittelkow MR, Shipley GR, Scott RE: Integrated control of growth anddifferentiation of normal human prokeratinocytes cultures in serum-free medium:Clonal analyses, growth kinetics and cell cycle studies. J Cell Physiol 121:31–44, 1984

Yarden Y, Ullrich A: Growth factor receptor tyrosine kinases. Ann Rev Biochem 57:443–478, 1988

Young AR: Chronic effects of ultraviolet radiation on the skin: experimental aspects. In:Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF, (eds). Dermatology inGeneral Medicine, 4th edn. 1993, pp. 1658–1660

Zheng ZS, Chen RZ, Prystowsky JH: UVB radiation induces phosphorylation of theepidermal growth factor receptor, decreases EGF binding and blocks EGF inductionof ornithine decarboxylase gene expression in SV40-transformed human keratinocytes.Exp Dermatol 2:257–265, 1993